QUICK FACTS
Created Jan 0001
Status Verified Sarcastic
Type Existential Dread
Keywords
definite integrals, calculus, indefinite integral, antiderivatives, integers, integral (disambiguation), definite integral, function, signed area

Integral

“This article delves into the concept of definite integrals within the realm of calculus. For a discussion of the indefinite integral, which is fundamentally...”

Contents
  • 1. Overview
  • 2. Etymology
  • 3. Cultural Impact

Operation in mathematical calculus

This article delves into the concept of definite integrals within the realm of calculus . For a discussion of the indefinite integral , which is fundamentally tied to the inverse operation, please refer to the dedicated article on antiderivatives . If you’re looking for the specific set of numbers, you might want to consult the entry on integers . For other uses and broader interpretations of the term, the Integral (disambiguation) page awaits your perusal. Furthermore, while the phrase “Area under the curve” often evokes the geometric intuition of integration, for other uses or more specialized contexts, see Area under the curve (disambiguation) .

A definite integral of a function isn’t just a number; it’s a sophisticated representation of the signed area of the region meticulously bounded by the function’s graph and the horizontal axis. As illustrated in the accompanying graph, consider the integral of any given function, say, $f(x)$. The total value is derived by subtracting the yellow (negative) area from the blue (positive) area. It’s a precise accounting, a balance sheet of geometric contributions.

Part of a series of articles about Calculus

The essence of calculus can often be distilled to a single, elegant truth: the Fundamental theorem of calculus . This theorem, a cornerstone of mathematical analysis, states that the process of integration and differentiation are, in fact, inverse operations. Specifically, for a function $f’(t)$ representing the instantaneous rate of change, its definite integral from $a$ to $b$ yields the net change in its antiderivative, $f(t)$, over that interval:

$$\int _{a}^{b}f’(t),dt=f(b)-f(a)$$

This profound connection underpins much of the practical application of integrals.

Fundamental Concepts of Calculus

Differential Calculus

Definitions

Concepts

Rules and identities

Integral Calculus

Definitions

Integration by (Techniques)

Series

Convergence tests

Vector Calculus

Theorems

Multivariable Calculus

Formalisms

Definitions

Advanced Topics

Specialized Integrals

Miscellanea

In the grand scheme of mathematics , an integral stands as the continuous counterpart to a sum . It’s the mechanism by which we elegantly calculate areas , volumes , and their more abstract, higher-dimensional generalizations. Integration, the precise art of computing an integral, is one of the two foundational operations of calculus – a fact so fundamental it almost doesn’t bear repeating, yet here we are. The other, of course, being differentiation . Initially, integration was a specialized tool, deployed to tackle pressing problems in mathematics and physics , such as precisely determining the elusive area beneath a curve, or discerning an object’s total displacement from its velocity profile. Its utility, however, proved far too profound to be confined, and its application rapidly broadened across an expansive array of scientific and engineering disciplines.

A definite integral provides a quantifiable measure: the signed area of a specific region within the Cartesian plane. This region is meticulously delineated by the graph of a given function and bounded between two designated points on the real line . By convention, areas that reside above the horizontal axis of the plane are accorded a positive value, while those that dip below are considered negative. This “signed” aspect is crucial for representing net changes. Beyond this geometric interpretation, integrals also refer to the concept of an antiderivative , which is essentially a function whose derivative is the original given function; in this context, they are often referred to as indefinite integrals. The profound fundamental theorem of calculus serves as the conceptual bridge, elegantly relating definite integration to differentiation. More importantly, it furnishes a remarkably efficient method for computing the definite integral of a function, provided its antiderivative is known, solidifying the understanding that differentiation and integration are intrinsically inverse operations.

While the sophisticated methods for calculating areas and volumes might seem like a modern invention, their conceptual roots stretch back to ancient Greek mathematics . However, the truly systematic and robust principles of integration, as we understand them today, were independently formulated by the intellectual titans Isaac Newton and Gottfried Wilhelm Leibniz in the latter half of the 17th century. Their revolutionary insight involved conceiving the area under a curve as an infinite summation of infinitesimally thin rectangles, a concept that, while intuitive, lacked the rigorous formalization it would later receive. That rigor arrived with Bernhard Riemann , who, centuries later, provided a precise and widely adopted definition of integrals. His approach was predicated on a limiting procedure, meticulously approximating the area of a curvilinear region by dissecting it into an infinite number of vanishingly narrow vertical slabs. Yet, even Riemann’s elegant framework had its limitations. In the nascent years of the 20th century, Henri Lebesgue further generalized Riemann’s formulation, introducing what is now universally known as the Lebesgue integral . This monumental achievement proved more encompassing than Riemann’s, capable of integrating a significantly broader class of functions, thereby expanding the very landscape of what could be considered “integrable.”

The concept of integrals is also remarkably flexible, capable of being generalized based on both the intrinsic nature of the function being analyzed and the specific domain over which the integration is performed. For instance, a line integral extends the idea to functions of two or more variables, where the traditional interval of integration is replaced by a continuous curve connecting two points in space. Similarly, in a surface integral , the curve is further elevated to a piece of a surface embedded within three-dimensional space , allowing for calculations over complex geometries.

History

See also: History of calculus

To truly appreciate the concept of integration, one must grudgingly acknowledge its protracted historical development. It wasn’t simply conjured into existence.

Pre-calculus integration

The earliest documented systematic methodology capable of approximating and, in some cases, precisely determining integrals can be traced back to the method of exhaustion . This ingenious technique was pioneered by the ancient Greek astronomer Eudoxus and later refined by the philosopher Democritus around 370 BC. Their goal was to ascertain the areas and volumes of complex shapes by conceptually dissecting them into an infinite multitude of smaller divisions, each of whose area or volume was already a known quantity. This method, a precursor to modern integration, was significantly developed and masterfully employed by Archimedes in the 3rd century BC. His groundbreaking work utilized exhaustion to calculate the exact area of a circle , the surface area and volume of a sphere , the area of an ellipse , the area lying beneath a parabola , the volume of a segment of a paraboloid of revolution, the volume of a segment of a hyperboloid of revolution, and even the area of a spiral . It was an impressive feat of geometric ingenuity, considering the mathematical tools available at the time.

A remarkably similar methodological approach emerged independently in China around the 3rd century AD, attributed to the mathematician Liu Hui , who successfully applied it to determine the area of the circle. This method found further application in the 5th century, when the esteemed Chinese father-and-son mathematicians Zu Chongzhi and Zu Geng leveraged it to calculate the volume of a sphere, demonstrating parallel intellectual developments across distinct cultures.

Moving to the Middle East, the polymath Hasan Ibn al-Haytham, known in Latin as Alhazen (c. 965 – c. 1040 AD), made significant contributions. He derived a sophisticated formula for the sum of fourth powers , a non-trivial achievement. Alhazen’s work extended to determining the equations necessary to calculate the area bounded by the curve represented by $y = x^k$ (which, in contemporary notation, translates to the integral $\int x^k,dx$), for any specified non-negative integer value of $k$. He then astutely applied these results to perform what would now be recognized as an integration of this function. Crucially, his formulae for the sums of integral squares and fourth powers allowed him to precisely calculate the volume of a paraboloid .

Despite these early sparks of genius, the next truly significant advancements in integral calculus lay dormant until the 17th century. It was during this period that the work of Cavalieri , with his groundbreaking method of indivisibles , alongside the independent contributions of Fermat , began to meticulously construct the intellectual scaffolding for modern calculus. Cavalieri, notably, computed the integrals of $x^n$ for powers up to $n = 9$, a considerable achievement encapsulated in Cavalieri’s quadrature formula . The peculiar case of $n = -1$, however, necessitated the invention of an entirely new function – the hyperbolic logarithm – which was ultimately achieved through the quadrature of the hyperbola in 1647.

Further critical steps were taken in the early 17th century by Barrow and Torricelli . Their investigations provided the very first tantalizing hints of a profound connection between integration and differentiation . Barrow, in particular, is credited with providing one of the earliest proofs of the fundamental theorem of calculus , laying crucial groundwork. Wallis then generalized Cavalieri’s method, extending the computation of integrals of $x$ to a general power, encompassing both negative and fractional exponents, thereby expanding the scope of integrable functions considerably.

Leibniz and Newton

The monumental leap forward in integration, the one that truly reshaped mathematics, occurred in the 17th century with the independent, and famously contentious, discovery of the fundamental theorem of calculus by Leibniz and Newton . This theorem, a testament to simultaneous intellectual breakthroughs, definitively established the intrinsic connection between integration and differentiation. This revelation, coupled with the comparatively simpler mechanics of differentiation, could now be exploited to calculate integrals with unprecedented efficiency. Crucially, the fundamental theorem of calculus empowered mathematicians to solve a vastly broader and more complex class of problems than ever before. Of equally profound importance was the comprehensive mathematical framework that both Leibniz and Newton meticulously developed. This framework, christened infinitesimal calculus, permitted a precise and rigorous analysis of functions defined over continuous domains. It is this very framework that ultimately evolved into modern calculus , and it is from Leibniz’s original work that the now ubiquitous notation for integrals is directly derived.

Formalization

While Newton and Leibniz undeniably provided a systematic and powerful approach to integration, their initial work, like many pioneering efforts, suffered from a certain lack of rigour . This deficiency did not escape critical notice. Bishop Berkeley , with characteristic philosophical acuity, famously assailed the “vanishing increments” employed by Newton, derisively labeling them as “ghosts of departed quantities.” The intellectual discomfort with such intuitive, yet ill-defined, concepts persisted until calculus eventually acquired a much firmer and more unshakeable footing with the systematic development of limits . Integration itself was first rigorously formalized, using the precise language of limits, by Riemann . His definition, while a monumental step, still had its boundaries. Although all bounded piecewise continuous functions are indeed Riemann-integrable on a bounded interval, mathematicians soon encountered more general and often “pathological” functions – particularly within the demanding context of Fourier analysis – to which Riemann’s definition simply did not apply. It was in response to these limitations that Lebesgue formulated his distinct definition of the integral, one firmly rooted in measure theory , a sophisticated subfield of real analysis . This new definition proved significantly more expansive, capable of integrating a wider class of functions. Since then, other definitions of integrals have been proposed, each extending Riemann’s and Lebesgue’s approaches to address specific mathematical challenges. While these approaches, grounded in the real number system, remain the most prevalent today, alternative methodologies exist, such as the definition of an integral as the standard part of an infinite Riemann sum, which draws upon the somewhat esoteric hyperreal number system.

Historical notation

The notation that has become synonymous with the indefinite integral, the elongated ‘S’ symbol (∫), was first introduced by Gottfried Wilhelm Leibniz in the year 1675. He ingeniously adapted the integral symbol , ∫, from the letter Ćż (long s ), which was itself an abbreviation for summa (often written as Ćżumma in Latin), meaning “sum” or “total.” This choice perfectly encapsulated the idea of integration as an infinite summation. The modern notation for the definite integral, featuring the limits of integration positioned conspicuously above and below the integral sign, saw its initial use by Joseph Fourier in the MĂ©moires of the French Academy, appearing sometime between 1819 and 1820, and subsequently reprinted in his seminal work of 1822.

Interestingly, Isaac Newton , in his own development of calculus, employed a different system. He indicated integration by placing a small vertical bar above a variable, or, perhaps even more unconventionally, by enclosing the variable within a box. However, these notations proved problematic. The vertical bar could be easily confused with symbols like $\dot{x}$ or $x’$, which were already established for indicating differentiation. The box notation, while visually distinct, posed considerable difficulties for printers of the era to reproduce accurately. Consequently, Newton’s preferred notations for integration did not gain widespread adoption, and Leibniz’s more elegant and practical symbol ultimately prevailed.

First use of the term

The actual term “integral” itself first appeared in print, in Latin, by Jacob Bernoulli in 1690, in the phrase: “Ergo et horum Integralia aequantur” (“Therefore their Integrals are equal”). It’s almost as if he knew it would stick.

Terminology and notation

In general, the integral of a real-valued function $f(x)$ with respect to a real variable $x$ over a specified interval $[a, b]$ is formally written as:

$$\int _{a}^{b}f(x),dx.$$

Let’s dissect this notation, since people invariably get it wrong. The prominent integral sign, ∫, is the universal emblem representing the operation of integration itself. The symbol $dx$, often referred to as the differential of the variable $x$, serves a crucial purpose: it unequivocally indicates that $x$ is the variable with respect to which the integration is being performed. Without it, you’re just staring at a squiggle. The function $f(x)$ is known as the integrand – it’s the function whose properties (like area or volume) we are interested in accumulating. The values $a$ and $b$ are designated as the limits (or bounds) of integration, defining the specific segment of the real line over which the integration is to occur. Consequently, the integral is said to be “over the interval $[a, b]$,” which is, rather uncreatively, called the interval of integration. A function earns the descriptor “integrable” if its integral over its domain yields a finite value. When these precise limits ($a$ and $b$) are explicitly stated, the integral is categorized as a definite integral .

Conversely, when these limits are conspicuously absent, as in the expression:

$$\int f(x),dx,$$

the integral assumes the designation of an indefinite integral. This form does not represent a single numerical value, but rather an entire class of functions – specifically, the antiderivatives – each of which possesses the integrand $f(x)$ as its derivative . The profound utility of the fundamental theorem of calculus truly shines here, as it provides the indispensable link between the evaluation of definite integrals and the determination of these indefinite integrals. It is worth noting that the notation for integrals has been further extended and adapted to encompass integration over unbounded domains and across multiple dimensions, topics which will, unfortunately for me, be explored in later sections of this article.

In more advanced mathematical contexts, particularly when the simple Riemann integral is the implied default or the precise type of integral is not germane to the discussion, it is not uncommon to omit the $dx$ term. For example, one might encounter an expression like:

$$\int {a}^{b}(c{1}f+c_{2}g)=c_{1}\int {a}^{b}f+c{2}\int _{a}^{b}g$$

This concise notation is often employed to articulate the fundamental property of linearity, a characteristic shared by the Riemann integral and all its subsequent generalizations. It’s a shorthand, a nod to efficiency, but one earned through understanding.

Interpretations

Integrals, for all their intimidating notation, manifest in a surprising number of practical scenarios. Consider, if you must, the mundane task of determining the volume of water a swimming pool can hold. For a simple rectangular pool with a flat bottom, basic geometry (length, width, depth) suffices. But introduce an oval shape with a gracefully rounded bottom, and suddenly, integrals become not just useful, but absolutely essential for deriving precise and rigorously accurate values for its volume, surface area, and even the length of its curved edge. In such complex cases, the underlying principle remains consistent: one conceptually dissects the desired quantity into an infinite series of infinitesimal pieces, and then, by meticulously summing these vanishingly small contributions, one can achieve an arbitrarily accurate approximation of the total. It’s the ultimate “divide and conquer” strategy.

Let’s illustrate with another example, since abstract concepts tend to elude some. Imagine the task of finding the precise area of the region bounded by the graph of the function $f(x) = \sqrt{x}$ between $x = 0$ and $x = 1$. A rudimentary approach might involve dividing the interval $[0, 1]$ into a finite number of segments, say, five pieces (specifically, from 0 to 1/5, 1/5 to 2/5, and so on, up to 1). You could then construct rectangles on each of these subintervals, using the height of the function at the right endpoint of each piece (i.e., $\sqrt{1/5}$, $\sqrt{2/5}$, …, $\sqrt{1}$). Summing the areas of these rectangles would yield an approximation:

$$\textstyle {\sqrt {\frac {1}{5}}}\left({\frac {1}{5}}-0\right)+{\sqrt {\frac {2}{5}}}\left({\frac {2}{5}}-{\frac {1}{5}}\right)+\cdots +{\sqrt {\frac {5}{5}}}\left({\frac {5}{5}}-{\frac {4}{5}}\right)\approx 0.7497,$$

which, as you might observe, is slightly larger than the true value. Alternatively, if you were to use the height of the function at the left endpoint of each piece, the approximation would be too low; with twelve such subintervals, the approximated area would only be 0.6203. However, the true power of integration emerges when the number of these “pieces” is allowed to approach infinity. As the width of each segment shrinks to an infinitesimal size, the sum of these infinitely many, vanishingly thin rectangles converges precisely to a definite limit , which represents the exact value of the desired area (in this particular case, 2/3). This process is succinctly captured by the integral notation:

$$\int _{0}^{1}{\sqrt {x}},dx={\frac {2}{3}},$$

This expression fundamentally means that 2/3 is the result of a weighted sum of the function values, $\sqrt{x}$, each multiplied by its corresponding infinitesimal step width, denoted by $dx$, across the interval $[0, 1]$. It’s the ultimate precision, achieved by embracing the infinite.

Darboux sums

Formal definitions

There are, regrettably, numerous ways to formally define an integral, and to complicate matters further, not all of them are perfectly equivalent. These distinctions primarily arise from the need to accommodate various special cases of functions that might not be integrable under more restrictive definitions, though occasionally, pedagogical convenience also plays a role. The definitions most commonly encountered and widely utilized in modern mathematics are the Riemann integral and the Lebesgue integral .

Riemann integral

Main article: Riemann integral

The Riemann integral is formally constructed using Riemann sums of functions, meticulously defined with respect to “tagged partitions” of an interval. A tagged partition of a closed interval $[a, b]$ on the real line is essentially a finite sequence of points that both subdivides the interval and selects a representative point within each subinterval. Specifically, it looks like this:

$$a=x_{0}\leq t_{1}\leq x_{1}\leq t_{2}\leq x_{2}\leq \cdots \leq x_{n-1}\leq t_{n}\leq x_{n}=b.,!$$

This intricate arrangement partitions the overarching interval $[a, b]$ into $n$ smaller, non-overlapping sub-intervals, denoted $[x_{i-1}, x_i]$, each uniquely indexed by $i$. Crucially, each of these sub-intervals is “tagged” with a specific, chosen point $t_i$, where $t_i$ must lie within its respective sub-interval $[x_{i-1}, x_i]$. A Riemann sum for a function $f$ with respect to such a meticulously tagged partition is then defined as:

$$\sum {i=1}^{n}f(t{i}),\Delta _{i};$$

Here, each individual term in the sum represents the area of a rectangle. The height of this rectangle is determined by the function’s value, $f(t_i)$, at the chosen tag point within the sub-interval, and its width is precisely the width of that sub-interval, $\Delta_i = x_i - x_{i-1}$. The ‘mesh’ of such a tagged partition is defined as the width of the largest sub-interval created by the partition, specifically $\max_{i=1…n} \Delta_i$. The Riemann integral of a function $f$ over the interval $[a, b]$ is said to be equal to a value $S$ if, and only if:

For every $\varepsilon > 0$ (no matter how small, representing an acceptable error margin), there exists a $\delta > 0$ (a sufficiently small mesh size) such that, for any tagged partition of $[a,b]$ with a mesh size less than $\delta$, the absolute difference between $S$ and the corresponding Riemann sum is less than $\varepsilon$:

$$\left|S-\sum {i=1}^{n}f(t{i}),\Delta _{i}\right|<\varepsilon .$$

This definition, in essence, captures the idea that as the partitions become infinitely fine (i.e., the mesh approaches zero), the Riemann sum converges to a unique value, $S$. It’s a testament to the power of limits. When the chosen tag points $t_i$ are consistently either the maximum or minimum value of the function within each respective interval, the Riemann sum transforms into an upper or lower Darboux sum , respectively. This highlights the inherent and very close relationship between the Riemann integral and the Darboux integral , which often serve as equivalent formalizations.

Lebesgue integral

Main article: Lebesgue integration

Lebesgue integration

It is frequently of considerable interest, both in the development of mathematical theory and in its practical applications, to be able to interchange the operation of taking a limit with the operation of integration. For example, one might construct a sequence of functions that, in some appropriate sense, approximates the solution to a complex problem. Ideally, the integral of this exact solution function should then be the limit of the integrals of its approximations. However, a significant number of functions that can be generated as limits of sequences are, unfortunately, not Riemann-integrable . Consequently, many crucial limit theorems, which are indispensable for advanced analysis, simply fail to hold when one is restricted to the Riemann integral . This glaring deficiency underscored the pressing need for a more expansive definition of the integral, one capable of integrating a wider and more diverse class of functions.

This more powerful and encompassing integral is the Lebesgue integral . Its ingenuity lies in exploiting a rather counter-intuitive but profound mathematical insight: if the values of a function are merely rearranged over its domain , the fundamental value of the function’s integral should remain unaltered. Building upon this insight, Henri Lebesgue introduced the integral that now bears his name, famously illustrating its core principle in a letter to Paul Montel with a vivid analogy that, even I admit, makes the concept somewhat palatable:

“I have to pay a certain sum, which I have collected in my pocket. I take the bills and coins out of my pocket and give them to the creditor in the order I find them until I have reached the total sum. This is the Riemann integral. But I can proceed differently. After I have taken all the money out of my pocket I order the bills and coins according to identical values and then I pay the several heaps one after the other to the creditor. This is my integral.”

As Folland so succinctly articulates, “To compute the Riemann integral of $f$, one partitions the domain $[a, b]$ into subintervals,” whereas in the Lebesgue integral , “one is in effect partitioning the range of $f$.” This shift in perspective is the critical difference. The formal definition of the Lebesgue integral thus commences with the introduction of a measure , denoted $\mu$. In its most straightforward application, the Lebesgue measure $\mu(A)$ of an interval $A = [a, b]$ is simply its width, $b - a$. In this basic scenario, the Lebesgue integral perfectly aligns with the (proper) Riemann integral whenever both exist. However, in more intricate and mathematically challenging contexts, the sets being measured can be extraordinarily fragmented, exhibiting no discernible continuity and bearing no resemblance whatsoever to simple intervals. It’s where the Riemann definition falters, and Lebesgue shines.

Employing this “partitioning the range of $f$” philosophy, the integral of a non-negative function $f: \mathbb{R} \to \mathbb{R}$ is conceptualized as the sum, over all possible values $t$, of the areas enclosed by infinitesimally thin horizontal strips positioned between $y = t$ and $y = t + dt$. The area of such a strip is precisely $\mu{x : f(x) > t}, dt$. Let’s define an auxiliary function $f^*(t) = \mu{x : f(x) > t}$. With this, the Lebesgue integral of $f$ is then formally defined by:

$$\int f=\int _{0}^{\infty }f^{*}(t),dt$$

Here, the integral on the right-hand side is an ordinary improper Riemann integral . This is permissible because $f^*$ is a strictly decreasing positive function, which inherently possesses a well-defined improper Riemann integral. For a suitable and specific class of functions, known as measurable functions , this formulation precisely defines the Lebesgue integral .

A general measurable function $f$ is considered Lebesgue-integrable if the sum of the absolute values of the areas of the regions between the graph of $f$ and the x-axis is finite. That is, if:

$$\int _{E}|f|,d\mu <+\infty .$$

In such a case, the integral, mirroring the approach of the Riemann integral , is precisely the difference between the total area residing above the x-axis and the total area residing below the x-axis:

$$\int _{E}f,d\mu =\int _{E}f^{+},d\mu -\int _{E}f^{-},d\mu$$

where $f^+(x)$ and $f^-(x)$ represent the positive and negative parts of the function $f(x)$ respectively:

$${\begin{aligned}&f^{+}(x)&&{}={}\max{f(x),0}&&{}={}{\begin{cases}f(x),&{\text{if }}f(x)>0,\0,&{\text{otherwise,}}\end{cases}}\&f^{-}(x)&&{}={}\max{-f(x),0}&&{}={}{\begin{cases}-f(x),&{\text{if }}f(x)<0,\0,&{\text{otherwise.}}\end{cases}}\end{aligned}}$$

This nuanced decomposition allows the Lebesgue integral to handle functions that oscillate wildly or exhibit discontinuities in ways that would render them non-integrable under Riemann’s definition, making it a far more robust and versatile tool for advanced mathematical analysis.

Other integrals

While the Riemann and Lebesgue integrals represent the two most dominant and widely utilized definitions of the integral in contemporary mathematics, they are by no means the only formulations. A multitude of other integral definitions exist, each developed to address specific mathematical challenges, generalize existing concepts, or provide alternative theoretical foundations. These include, but are not limited to:

Each of these integral definitions serves a distinct purpose, pushing the boundaries of what can be meaningfully “summed” in a continuous sense, and demonstrating the rich diversity of mathematical thought in response to complex problems.

Properties

One would hope that something as fundamental as an integral would possess some useful properties. Fortunately, it does.

Linearity

The collection of Riemann-integrable functions defined on a closed interval $[a, b]$ forms a vector space under the standard operations of pointwise addition and multiplication by a scalar. Furthermore, the operation of integration itself, represented as:

$$f\mapsto \int _{a}^{b}f(x);dx$$

is a linear functional on this vector space. What this means, in plain terms, is that the set of integrable functions is inherently closed under the formation of linear combinations . Consequently, the integral of a linear combination of functions is precisely the linear combination of their individual integrals. This elegant property can be formally expressed as:

$$\int _{a}^{b}(\alpha f+\beta g)(x),dx=\alpha \int _{a}^{b}f(x),dx+\beta \int _{a}^{b}g(x),dx.,$$

Here, $\alpha$ and $\beta$ represent arbitrary scalar constants, and $f$ and $g$ are any two Riemann-integrable functions over the interval $[a, b]$. This isn’t merely a convenient shortcut; it’s a fundamental structural characteristic that simplifies many calculations.

Similarly, the set of real-valued Lebesgue-integrable functions on a given measure space $E$ with measure $\mu$ also exhibits this crucial property. It is closed under the formation of linear combinations, and thus also constitutes a vector space. The Lebesgue integral , represented as:

$$f\mapsto \int _{E}f,d\mu$$

is likewise a linear functional on this vector space, leading to the same linearity property:

$$\int _{E}(\alpha f+\beta g),d\mu =\alpha \int _{E}f,d\mu +\beta \int _{E}g,d\mu .$$

More generally, one can consider the vector space encompassing all measurable functions on a measure space $(E, \mu)$, where these functions take values in a locally compact complete topological vector space $V$ over a locally compact topological field $K$. For such functions, $f: E \to V$, an abstract integration map can be defined, assigning to each function $f$ an element of $V$ or the symbol $\infty$:

$$f\mapsto \int _{E}f,d\mu ,,$$

This map is designed to be compatible with linear combinations. In this sophisticated setting, the linearity property holds true specifically for the subspace of functions whose integral results in an element of $V$ (i.e., those whose integral is “finite” in some appropriate sense). The most significant special cases of this general framework arise when $K$ is the set of real numbers ($\mathbb{R}$), complex numbers ($\mathbb{C}$), or a finite extension of the field $\mathbb{Q}_p$ of p-adic numbers , and $V$ is a finite-dimensional vector space over $K$. Another crucial instance is when $K = \mathbb{C}$ and $V$ is a complex Hilbert space .

The property of linearity, when combined with certain natural continuity properties and a normalization condition for a specific class of “simple” functions, can actually be used to construct an alternative, axiomatic definition of the integral itself. This is precisely the approach championed by Daniell for real-valued functions on a set $X$, and subsequently generalized by Nicolas Bourbaki to functions whose values reside in a locally compact topological vector space . For a comprehensive axiomatic characterization of the integral, one might consult Hildebrandt (1953). It’s all very neat, if you’re into that sort of thing.

Inequalities

A selection of rather useful general inequalities applies to Riemann-integrable functions defined on a closed and bounded interval $[a, b]$. These inequalities are not merely curiosities; they provide fundamental bounds and relationships that are indispensable in analysis and can often be generalized to other notions of integral, such as the Lebesgue and Daniell integrals .

  • Upper and lower bounds: An integrable function $f$ on $[a, b]$ is, by its very nature, necessarily bounded on that interval. This means there exist real numbers $m$ and $M$ such that $m \leq f(x) \leq M$ for all $x$ within $[a, b]$. Since the lower and upper sums of $f$ over $[a, b]$ are consequently bounded by $m(b-a)$ and $M(b-a)$ respectively, it logically follows that the integral itself must lie within these bounds:

    $$m(b-a)\leq \int _{a}^{b}f(x),dx\leq M(b-a).$$

    This provides a quick sanity check for any integral calculation.

  • Inequalities between functions: If, for every $x$ in $[a, b]$, the function $f(x)$ is always less than or equal to $g(x)$ (i.e., $f(x) \leq g(x)$), then it stands to reason that the integral of $f$ will be less than or equal to the integral of $g$. Specifically, each of the upper and lower sums of $f$ will be bounded above by the corresponding upper and lower sums of $g$. Thus:

    $$\int _{a}^{b}f(x),dx\leq \int _{a}^{b}g(x),dx.$$

    This property is a direct generalization of the upper and lower bounds mentioned above, as $M(b-a)$ can be seen as the integral of a constant function with value $M$ over $[a, b]$. Moreover, if the inequality between the functions is strict – meaning $f(x) < g(x)$ for every $x$ in $[a, b]$ – then the inequality between their integrals also becomes strictly less than:

    $$\int _{a}^{b}f(x),dx<\int _{a}^{b}g(x),dx.$$

  • Subintervals: If $[c, d]$ is a subinterval contained within $[a, b]$, and the function $f(x)$ is non-negative for all $x$ in its domain, then the integral of $f$ over the smaller subinterval cannot exceed its integral over the larger interval:

    $$\int _{c}^{d}f(x),dx\leq \int _{a}^{b}f(x),dx.$$

    This is intuitively obvious: if you’re accumulating positive quantities, integrating over a smaller region will yield a smaller (or equal) total.

  • Products and absolute values of functions: When dealing with two functions, $f$ and $g$, one can naturally consider their pointwise products , powers, and absolute values :

    $$(fg)(x)=f(x)g(x),;f^{2}(x)=(f(x))^{2},;|f|(x)=|f(x)|.$$

    If $f$ is Riemann-integrable on $[a, b]$, then its absolute value, $|f|$, is also Riemann-integrable on that interval, and a crucial inequality holds: the absolute value of the integral of $f$ is less than or equal to the integral of the absolute value of $f$:

    $$\left|\int _{a}^{b}f(x),dx\right|\leq \int _{a}^{b}|f(x)|,dx.$$

    This is a fundamental result. Furthermore, if both $f$ and $g$ are Riemann-integrable , then their product $fg$ is also Riemann-integrable, and we encounter the powerful Cauchy–Schwarz inequality :

    $$\left(\int _{a}^{b}(fg)(x),dx\right)^{2}\leq \left(\int _{a}^{b}f(x)^{2},dx\right)\left(\int _{a}^{b}g(x)^{2},dx\right).$$

    This inequality holds profound significance, particularly in Hilbert space theory, where the left-hand side is elegantly interpreted as the inner product of two square-integrable functions $f$ and $g$ over the interval $[a, b]$.

  • Hölder’s inequality: This is a generalization of the Cauchy–Schwarz inequality . Suppose $p$ and $q$ are two real numbers such that $1 \leq p, q \leq \infty$ and they satisfy the reciprocal relationship $1/p + 1/q = 1$. If $f$ and $g$ are two Riemann-integrable functions , then the functions $|f|^p$ and $|g|^q$ are also integrable, and the following Hölder’s inequality applies:

    $$\left|\int f(x)g(x),dx\right|\leq \left(\int \left|f(x)\right|^{p},dx\right)^{1/p}\left(\int \left|g(x)\right|^{q},dx\right)^{1/q}.$$

    It’s worth noting that when $p = q = 2$, Hölder’s inequality elegantly reduces to the familiar Cauchy–Schwarz inequality , demonstrating its broader scope.

  • Minkowski inequality: Another crucial inequality, particularly relevant in the study of Lp spaces . Suppose $p \geq 1$ is a real number , and $f$ and $g$ are Riemann-integrable functions . In this scenario, the functions $|f|^p$, $|g|^p$, and $|f+g|^p$ are also Riemann-integrable, and the following Minkowski inequality holds:

    $$\left(\int \left|f(x)+g(x)\right|^{p},dx\right)^{1/p}\leq \left(\int \left|f(x)\right|^{p},dx\right)^{1/p}+\left(\int \left|g(x)\right|^{p},dx\right)^{1/p}.$$

    An analogous version of this inequality for the Lebesgue integral is absolutely fundamental in the construction and understanding of Lp spaces , which are essential function spaces in modern analysis. These inequalities aren’t just abstract theorems; they are powerful tools for bounding, estimating, and understanding the behavior of functions and their integrals.

Conventions

In this section, we’ll consider $f$ to be a real-valued Riemann-integrable function . The integral:

$$\int _{a}^{b}f(x),dx$$

over an interval $[a, b]$ is inherently defined under the explicit condition that $a < b$. This prerequisite implies that the upper and lower sums of the function $f$ are evaluated on a partition $a = x_0 \leq x_1 \leq \dots \leq x_n = b$ where the values $x_i$ are strictly increasing. Geometrically, this convention signifies that integration proceeds “left to right” along the real line , evaluating $f$ within consecutive intervals $[x_i, x_{i+1}]$ such that an interval with a higher index is always positioned to the right of one with a lower index. This directional aspect is critical for consistency. The values $a$ and $b$, which delineate the end-points of the interval , are formally termed the limits of integration of $f$.

However, the definition is extended to cover cases where $a > b$ through a crucial convention:

$$\int _{a}^{b}f(x),dx=-\int _{b}^{a}f(x),dx.$$

This convention ensures that reversing the direction of integration simply negates the value of the integral. It maintains consistency with the idea of signed area; if moving from left to right yields a positive area, moving right to left over the same region must yield a negative one.

Furthermore, with $a = b$, the definition implies a rather straightforward consequence:

$$\int _{a}^{a}f(x),dx=0.$$

This means that an integral taken over a degenerate interval—an interval of zero width, effectively a single point —must evaluate to zero . There’s no “area” to accumulate over a single point.

The first of these conventions (for $a > b$) is essential when one considers integrating over subintervals of a larger interval $[a, b]$. The second (for $a=b$) simply ensures consistency. More importantly, integrals possess a fundamental additive property: if $c$ is any element within the interval $[a, b]$, then the integral from $a$ to $b$ can be decomposed into a sum of integrals over the subintervals $[a, c]$ and $[c, b]$:

$$\int _{a}^{b}f(x),dx=\int _{a}^{c}f(x),dx+\int _{c}^{b}f(x),dx.$$

With the introduction of the first convention (regarding reversed limits), this additive property becomes remarkably robust and universally applicable. The resulting relation:

$${\begin{aligned}\int _{a}^{c}f(x),dx&{}=\int _{a}^{b}f(x),dx-\int _{c}^{b}f(x),dx\&{}=\int _{a}^{b}f(x),dx+\int _{b}^{c}f(x),dx\end{aligned}}$$

is then well-defined for any cyclic permutation of $a$, $b$, and $c$, regardless of their relative magnitudes. This ensures a consistent and coherent framework for manipulating integrals across various segments of the real line , simplifying complex problems considerably. It’s a small detail, but one that prevents utter chaos.

Fundamental theorem of calculus

Main article: Fundamental theorem of calculus

The fundamental theorem of calculus is not merely a theorem; it is the conceptual bedrock upon which much of calculus rests. It articulates the profound truth that differentiation and integration are intrinsically inverse operations : if a continuous function is first subjected to integration and then subsequently to differentiation, the original function is perfectly recovered. This elegant duality is what gives calculus its immense power and coherence. An extraordinarily important consequence of this theorem, often referred to as the second fundamental theorem of calculus, provides an indispensable practical method for computing definite integrals by leveraging an antiderivative of the function in question. It transforms a potentially arduous summation process into a simple evaluation.

First theorem

Let $f$ be a continuous real-valued function defined on a closed interval $[a, b]$. Consider a new function, $F$, meticulously defined for all $x$ in $[a, b]$ as the definite integral of $f$ from $a$ to $x$:

$$F(x)=\int _{a}^{x}f(t),dt.$$

Under these conditions, $F$ possesses several crucial properties: it is continuous across the entire interval $[a, b]$, it is differentiable on the open interval $(a, b)$, and its derivative is precisely the original function $f(x)$:

$$F’(x)=f(x)$$

for all $x$ in $(a, b)$. This part of the theorem essentially states that every continuous function has an antiderivative, and that antiderivative can be constructed via integration.

Second theorem

Now, let $f$ be a real-valued function defined on a closed interval $[a, b]$ that is known to admit an antiderivative $F$ on $[a, b]$. This means that $f$ and $F$ are functions such that for every $x$ in $[a, b]$, the derivative of $F$ is $f$:

$$f(x)=F’(x).$$

If, in addition, $f$ is integrable on $[a, b]$, then the definite integral of $f$ from $a$ to $b$ can be computed with remarkable simplicity by evaluating the antiderivative $F$ at the upper and lower limits of integration and subtracting the results:

$$\int _{a}^{b}f(x),dx=F(b)-F(a).$$

This is the part that makes calculating definite integrals practical. Instead of wrestling with Riemann sums or infinitely thin rectangles, you merely find an antiderivative and plug in the bounds. It’s almost too easy, which is why it’s so celebrated.

Extensions

The concept of integration, despite its inherent sophistication, is remarkably adaptable. Its foundational ideas can be extended and generalized to address scenarios that deviate from the standard assumptions of the basic Riemann integral .

Improper integrals

Main article: Improper integral

The improper integral

$$\int _{0}^{\infty }{\frac {dx}{(x+1){\sqrt {x}}}}=\pi $$

demonstrates a situation where the integration interval is unbounded for both the domain and the range of the function.

A “proper” Riemann integral operates under the strict assumption that the integrand is both defined and finite across a closed and bounded interval specified by its limits of integration. An improper integral arises precisely when one or more of these foundational conditions are violated. This can occur if the interval of integration is unbounded (extending to infinity), or if the integrand itself becomes unbounded (approaches infinity) at one or more points within the interval. In certain cases, such seemingly problematic integrals can still be rigorously defined by carefully considering the limit of a sequence of proper Riemann integrals evaluated over progressively larger or more refined intervals.

If the interval of integration is unbounded, for instance, extending indefinitely at its upper end, the improper integral is then defined as the limit as that endpoint approaches infinity:

$$\int _{a}^{\infty }f(x),dx=\lim _{b\to \infty }\int _{a}^{b}f(x),dx.$$

Similarly, if the integrand is only defined or remains finite on a half-open interval, such as $(a, b]$ (meaning it might be undefined or infinite at point $a$), a limit procedure can again yield a finite result:

$$\int _{a}^{b}f(x),dx=\lim _{\varepsilon \to 0}\int _{a+\varepsilon }^{b}f(x),dx.$$

The essence here is that an improper integral is fundamentally the limit of proper integrals as one or both endpoints of the integration interval approach either a specific real number where the function is problematic, or positive infinity ($\infty$), or negative infinity ($-\infty$). In more complex scenarios, limits might be required at both endpoints simultaneously, or even at internal points within the interval where the integrand exhibits singular behavior. It’s a way of salvaging meaning from integrals that initially seem ill-behaved.

Multiple integration

Main article: Multiple integral

A double integral computes the volume under a surface $z=f(x,y)$.

Just as the definite integral of a positive function of a single variable geometrically represents the area of the region nestled between the function’s graph and the x-axis, the double integral of a positive function of two variables elegantly represents the volume of the region situated between the surface defined by the function ($z = f(x,y)$) and the plane containing its domain . For instance, consider a function in two dimensions, dependent on two real variables , $x$ and $y$. If we wish to integrate this function $f$ over a rectangular region $R$ defined as the Cartesian product of two intervals, $R = [a,b] \times [c,d]$, the integral can be written as:

$$\int _{R}f(x,y),dA$$

where the differential $dA$ explicitly indicates that the integration is being performed with respect to area. This double integral can be rigorously defined using Riemann sums , much like its one-dimensional counterpart, and it precisely quantifies the (signed) volume residing beneath the graph of $z = f(x,y)$ over the specified domain $R$. Under suitable conditions (for example, if $f$ is a continuous function ), Fubini’s theorem provides a powerful simplification: it states that this multiple integral can be equivalently expressed as an iterated integral:

$$\int _{a}^{b}\left[\int _{c}^{d}f(x,y),dy\right],dx.$$

This remarkable result effectively reduces the complex task of computing a double integral to the more familiar process of computing a sequence of one-dimensional integrals. Due to this reduction, another common notation for the integral over region $R$ employs a double integral sign:

$$\iint _{R}f(x,y),dA.$$

The concept extends seamlessly to higher dimensions. Integration over more general, non-rectangular domains is also entirely possible, though it often requires more sophisticated techniques for setting up the limits of integration. The integral of a function $f$, with respect to volume, over an $n$-dimensional region $D$ of $\mathbb{R}^n$ is typically denoted by symbols such as:

$$\int _{D}f(\mathbf {x} )d^{n}\mathbf {x} \ =\int _{D}f,dV.$$

These higher-dimensional integrals are indispensable in physics, engineering, and advanced mathematics for calculating volumes, masses, and other aggregated quantities in multi-dimensional spaces.

Line integrals and surface integrals

Main articles: Line integral and Surface integral

A line integral sums together elements along a curve.

The fundamental concept of an integral can be further generalized to encompass more complex and geometrically intricate domains of integration . This includes integrating along curved lines and over surfaces embedded within higher-dimensional spaces. Such integrals are, rather descriptively, known as line integrals and surface integrals , respectively. These extensions possess profound and vital applications in physics , particularly when one is dealing with vector fields , where quantities vary not just in magnitude but also in direction across space.

A line integral (sometimes, perhaps more evocatively, called a path integral) is a type of integral where the function to be integrated is meticulously evaluated along a specific curve . There are several distinct varieties of line integrals in common use, depending on the nature of the function and the desired outcome. In the special case where the curve forms a closed loop, it is also referred to as a contour integral .

The function designated for integration in a line integral can be either a scalar field (where a single numerical value is assigned to each point in space) or a vector field (where a vector is assigned to each point). The resulting value of the line integral represents the sum of the values of the field at all points along the curve, with each contribution weighted by some scalar function defined on the curve. This weighting is commonly the arc length element ($ds$) or, in the case of a vector field, the scalar product of the vector field with an infinitesimal vector element of the curve ($d\mathbf{s}$). This crucial weighting mechanism is what distinguishes the line integral from simpler integrals defined solely on intervals . Many foundational formulas in physics find their natural continuous analogs in terms of line integrals. For example, the elementary concept that work ($W$) is equal to force ($\mathbf{F}$) multiplied by displacement ($\mathbf{s}$) can be elegantly expressed (using vector quantities) as:

$$W=\mathbf {F} \cdot \mathbf {s} .$$

For an object traversing a path $C$ within a vector field $\mathbf{F}$ (such as an electric field or a gravitational field ), the total work performed by the field on the object is obtained by summing up the differential work done as the object moves from $\mathbf{s}$ to $\mathbf{s} + d\mathbf{s}$. This leads directly to the line integral :

$$W=\int _{C}\mathbf {F} \cdot d\mathbf {s} .$$

This integral precisely captures the cumulative effect of the force along the entire trajectory.

The definition of a surface integral relies on splitting the surface into small surface elements.

A surface integral is a natural generalization of double integrals to the more complex scenario of integration over a surface , which itself may be a curved entity embedded within space . It can be conceptualized as the double integral equivalent of the line integral . As with line integrals, the function to be integrated can be either a scalar field or a vector field . The value of the surface integral represents the cumulative sum of the field’s values over all points on the surface. This is achieved by conceptually subdividing the surface into infinitesimally small surface elements, which then provide the necessary partitioning for forming Riemann sums .

As an illustrative application of surface integrals , consider a vector field $\mathbf{v}$ defined over a surface $S$. For every point $\mathbf{x}$ on $S$, $\mathbf{v}(\mathbf{x})$ represents a vector quantity. Imagine a fluid flowing through the surface $S$, where $\mathbf{v}(\mathbf{x})$ determines the velocity of the fluid at point $\mathbf{x}$. The flux is then defined as the total quantity of fluid flowing through $S$ per unit of time. To calculate this flux, one must take the dot product of the velocity vector $\mathbf{v}$ with the unit surface normal vector (a vector perpendicular to the surface) at each point on $S$. This operation yields a scalar field , which is then integrated over the entire surface:

$$\int _{S}{\mathbf {v} }\cdot ,d{\mathbf {S} }.$$

The fluid flux in this example need not be restricted to physical fluids like water or air; it can also represent abstract quantities such as electrical or magnetic flux. Consequently, surface integrals find extensive and critical applications across various domains of physics , most notably within the classical theory of electromagnetism , where they are indispensable for formulating fundamental laws.

Contour integrals

Main article: Contour integration

In the fascinating realm of complex analysis , the integrand takes on a different nature: it is a complex-valued function of a complex variable $z$, rather than a real function of a real variable $x$. When such a complex function is integrated along a specified curve , denoted $\gamma$, within the complex plane , the integral is symbolized as follows:

$$\int _{\gamma }f(z),dz.$$

This distinct type of integral is universally known as a contour integral . Contour integrals are immensely powerful tools, forming the backbone of much of complex analysis, enabling the evaluation of many real integrals that are intractable by real methods, and playing a crucial role in areas like fluid dynamics and quantum field theory.

Integrals of differential forms

Main article: Differential form

See also: Volume form and Density on a manifold

A differential form is a sophisticated mathematical construct rooted in the fields of multivariable calculus , differential topology , and the theory of tensors . These forms are meticulously organized by their “degree,” which corresponds to the dimension of the geometric object they measure. For instance, a one-form is essentially a weighted sum of the differentials of the coordinates, such as:

$$E(x,y,z),dx+F(x,y,z),dy+G(x,y,z),dz$$

where $E$, $F$, and $G$ are functions defined in three dimensions. A differential one-form can be integrated over an oriented path (a curve with a specified direction), and the resulting integral is simply an alternative, more generalized way of expressing a line integral . Here, the fundamental differentials $dx$, $dy$, and $dz$ serve to measure infinitesimal oriented lengths that are parallel to the three coordinate axes.

A differential two-form, by contrast, is a sum structured in the following manner:

$$G(x,y,z),dx\wedge dy+E(x,y,z),dy\wedge dz+F(x,y,z),dz\wedge dx.$$

In this case, the fundamental two-forms, namely $dx\wedge dy$, $dz\wedge dx$, and $dy\wedge dz$, are used to measure oriented areas that are parallel to the respective coordinate two-planes (e.g., the $xy$-plane, $xz$-plane, and $yz$-plane). The symbol $\wedge$ denotes the wedge product , an operation that shares conceptual similarities with the cross product in vector calculus . Specifically, the wedge product of two forms representing oriented lengths yields an oriented area. A two-form can be integrated over an oriented surface, and this resulting integral is mathematically equivalent to the surface integral that quantifies the flux of the vector field $E\mathbf{i} + F\mathbf{j} + G\mathbf{k}$.

A crucial distinction from the cross product and the traditional three-dimensional vector calculus is that the wedge product and the entire calculus of differential forms are rigorously defined and make perfect sense in arbitrary dimensions and on more generalized manifolds (which are abstract spaces that locally resemble Euclidean space, encompassing curves, surfaces, and their higher-dimensional analogs). Within this framework, the exterior derivative takes on the roles traditionally played by the gradient and curl operators in vector calculus. Furthermore, Stokes’ theorem emerges as a powerful, unifying generalization that simultaneously subsumes and extends the three major integral theorems of vector calculus: the divergence theorem , Green’s theorem , and the Kelvin-Stokes theorem . It’s an elegant, if somewhat abstract, way to unify seemingly disparate concepts.

Summations

Main article: Summation § Approximation by definite integrals

The discrete counterpart to integration is, quite simply, summation . While they appear distinct, these two fundamental mathematical operations can be unified under a common theoretical umbrella through the sophisticated framework of Lebesgue integrals or the increasingly relevant field of time-scale calculus . This inherent connection underscores that integration is, at its heart, a continuous summation process.

Functional integrals

Main article: Functional integration

An integration that is performed not over a conventional variable (like $x$, or, in physics , over a spatial or temporal dimension), but rather over an entire space of functions , is referred to as a functional integral . These integrals are particularly prominent in advanced theoretical physics, such as quantum field theory and statistical mechanics, where one often needs to sum over all possible configurations or “paths” a system might take.

Applications

Integrals, despite their abstract nature, are deployed extensively across an impressive array of fields. Their utility extends far beyond the confines of pure mathematics, proving indispensable in diverse scientific, engineering, and economic disciplines.

For instance, in the realm of probability theory , integrals are the essential tools used to determine the likelihood of some random variable falling within a specific range of values. This is achieved by integrating the probability density function over that particular range. Furthermore, a fundamental property dictates that the integral of an entire probability density function over its entire domain must precisely equal 1. This provides a crucial test for verifying whether a given function that yields no negative values can legitimately serve as a density function.

Beyond probability, integrals are foundational for computing the area of any two-dimensional region that possesses a curved boundary, as well as for computing the volume of three-dimensional objects similarly bounded by curved surfaces. The area of a two-dimensional region, as previously discussed, is directly calculable using the definite integral . The volume of a three-dimensional object, such as a disc or a washer, can be efficiently computed using disc integration . This method leverages the familiar equation for the volume of a cylinder, $\pi r^2 h$, where $r$ is the radius. In the specific case of a simple disc generated by rotating a curve $y = f(x)$ around the x-axis, the radius is given by $f(x)$, and its infinitesimal height is the differential $dx$. By applying an integral with bounds $a$ and $b$, the volume of the resulting solid of revolution is equal to:

$$\pi \int _{a}^{b}f^{2}(x),dx.$$

Integrals are also absolutely central to physics , finding applications in areas such as kinematics , where they are used to determine quantities like displacement , time , and velocity . For example, in rectilinear motion (motion along a straight line), the net displacement of an object over a time interval $[a, b]$ is given by:

$$x(b)-x(a)=\int _{a}^{b}v(t),dt,$$

where $v(t)$ represents the object’s velocity expressed as a function of time. Similarly, the work done by a force $F(x)$ (given as a function of position) as it moves an object from an initial position $A$ to a final position $B$ is calculated as:

$$W_{A\rightarrow B}=\int _{A}^{B}F(x),dx.$$

This is a continuous generalization of the simple “force times distance” formula. Integrals are also deployed in thermodynamics , where the technique of thermodynamic integration is employed to precisely calculate the difference in free energy between two specified thermodynamic states, a critical quantity in chemical and physical processes. In economics , integrals can be used to calculate consumer and producer surplus, or the total cost of production given a marginal cost function. In engineering , they are vital for calculating moments of inertia, centers of mass, fluid flow rates, and stress distributions. The list, frankly, is quite extensive.

Computation

Actually computing these things is often where the real fun begins, or ends, depending on your perspective.

Analytical

The most straightforward and widely taught technique for computing definite integrals of a single real variable is directly derived from the profound fundamental theorem of calculus . The process involves a few key steps: Let $f(x)$ be the function you wish to integrate over a given interval $[a, b]$. The first, and often most challenging, step is to find an antiderivative of $f$; that is, a function $F$ such that its derivative , $F’$, is precisely $f$ across the entire interval. Provided that both the integrand $f(x)$ and its antiderivative $F(x)$ exhibit no singularities along the path of integration (i.e., within the interval $[a, b]$), the fundamental theorem of calculus dictates that the definite integral can be effortlessly evaluated as:

$$\int _{a}^{b}f(x),dx=F(b)-F(a).$$

This elegant result transforms the complex problem of summation into a mere evaluation of a function at two points. However, finding an antiderivative isn’t always trivial. Sometimes, it becomes necessary to employ one of the many specialized techniques that have been painstakingly developed over centuries to evaluate integrals. Most of these techniques operate on the principle of transforming a given, possibly intractable, integral into a different one that is, hopefully, more manageable. These indispensable techniques include integration by substitution (a reverse chain rule), integration by parts (a reverse product rule), integration by trigonometric substitution (useful for expressions involving square roots of quadratic forms), and integration by partial fractions (for rational functions).

For more complex integrals, alternative methods are required. Many nonelementary integrals – those whose antiderivatives cannot be expressed in terms of elementary functions – can often be expanded into a Taylor series and then integrated term by term. Occasionally, the resulting infinite series can be summed analytically, leading to a closed-form solution. The method of convolution utilizing Meijer G-functions can also be a powerful tool, provided the integrand can be expressed as a product of such functions. Furthermore, there are numerous less common, almost esoteric, methods for calculating definite integrals; for instance, Parseval’s identity can be leveraged to transform an integral over a rectangular region into an infinite sum. Occasionally, an integral can be evaluated by a clever “trick,” a moment of mathematical insight; for a prime example of this, one need only look at the famous Gaussian integral .

The computations of volumes of solids of revolution are typically handled with either disk integration or shell integration , each offering a complementary geometric approach to slicing the solid. A vast compendium of specific results, derived using these various techniques, has been collected and formalized in extensive lists of integrals , serving as invaluable reference tools for mathematicians and scientists alike.

Symbolic

Main article: Symbolic integration

In a multitude of problems encountered in mathematics , physics , and engineering , the ultimate goal is not merely a numerical value, but an explicit, analytical formula for the integral itself. To this end, extensive tables of integrals have been meticulously compiled and published over centuries, serving as indispensable reference works. However, with the pervasive spread of computers, many professionals, educators, and students have increasingly turned to sophisticated computer algebra systems (CAS). These systems are specifically engineered to automate and perform difficult or tedious mathematical tasks, and symbolic integration stands as one of their primary capabilities. Indeed, the very ambition of symbolic integration was a principal driving force behind the initial development of some of the earliest and most influential CAS, such as Macsyma and Maple .

A significant mathematical hurdle in the pursuit of symbolic integration lies in the often-unfortunate reality that many relatively simple functions do not possess antiderivatives that can be expressed in closed form using only elementary functions . This includes the familiar rational and exponential functions, the logarithm , trigonometric functions , inverse trigonometric functions , and the basic operations of multiplication and composition. The Risch algorithm represents a monumental achievement in this domain, providing a general criterion to definitively determine whether the antiderivative of a given elementary function is itself elementary, and, if so, to then compute that integral. However, functions with closed-form antiderivatives are, regrettably, the exception rather than the rule. Consequently, even the most advanced computerized algebra systems cannot realistically hope to discover an antiderivative for an arbitrarily constructed elementary function.

On a more optimistic note, if the set of “building blocks” (i.e., the allowed functions and operations) for antiderivatives is fixed in advance, it may still be possible to determine whether the antiderivative of a given function can be expressed using these specified components and operations, and to then find that symbolic answer if it exists. The Risch algorithm , as implemented in powerful systems like Mathematica , Maple , and other computer algebra systems , performs precisely this task for functions and antiderivatives constructed from rational functions, radicals , logarithms, and exponential functions.

More recently, a novel approach has emerged, leveraging the concept of D-finite functions . These are functions that are solutions to linear differential equations with polynomial coefficients. The vast majority of elementary and special functions fall into the category of D-finite functions. Crucially, the integral of a D-finite function is also a D-finite function. This insight provides an algorithmic framework to express the antiderivative of a D-finite function as the solution to a differential equation. This theory further allows for the computation of the definite integral of a D-finite function as the sum of a series derived from its initial coefficients, and provides an algorithm to compute any specific coefficient of that series.

Rule-based integration systems also considerably facilitate the process of integration. Rubi, a prominent computer algebra system that employs rule-based integration, utilizes an extensive system of symbolic integration rules to pattern match and integrate a remarkably wide variety of integrands. This system, for example, employs over 6600 distinct integration rules to compute integrals, a testament to the complexity and breadth of the problem. The method of brackets offers another powerful, generalized approach, extending Ramanujan’s master theorem to a broad spectrum of univariate and multivariate integrals. This method applies a specific set of rules to the coefficients and exponential terms of the integrand’s power series expansion to determine the integral, and it is intimately related to the Mellin transform . It’s a testament to human ingenuity to automate what was once considered the exclusive domain of mathematical virtuosos.

Numerical

Main article: Numerical integration

Numerical quadrature methods: rectangle method , trapezoidal rule , Romberg’s method (at varying piece count), Gaussian quadrature .

When an analytical solution for a definite integral proves elusive or impossible to find, or when the function itself is only known through discrete data points, numerical methods become indispensable. Definite integrals can be effectively approximated using a variety of techniques collectively known as numerical integration or quadrature. The simplest approach is the rectangle method , which, conceptually, relies on dividing the region under the function’s graph into a series of rectangles. The height of each rectangle is determined by the function’s value at a chosen point within each subinterval (e.g., left endpoint, right endpoint, or midpoint), and this height is then multiplied by the fixed step width (the width of the subinterval) to find the area of each rectangle. Summing these areas provides an approximation of the integral.

A more refined approach, the trapezoidal rule , improves upon the rectangle method by replacing the rectangular strips of a Riemann sum with trapezoids. This is achieved by connecting the function values at the endpoints of each subinterval with a straight line, thereby forming a trapezoid. The trapezoidal rule typically weights the function values at the first and last points by one-half, then multiplies by the step width to yield a more accurate approximation. The underlying principle here is that more accurate local approximations of the function will naturally lead to better global approximations of the integral. This idea can be further extended: Simpson’s rule elevates the approximation by using a piecewise quadratic function to model the integrand over each pair of subintervals, often yielding significantly greater accuracy for smooth functions.

The Riemann sums , trapezoidal rule , and Simpson’s rule are all examples belonging to a broader family of quadrature rules known as the Newton–Cotes formulas . A Newton–Cotes quadrature rule of degree $n$ approximates the integrand on each subinterval using a polynomial of degree $n$. This polynomial is specifically chosen to interpolate the values of the function at a set of equally spaced points within the interval. While higher-degree Newton–Cotes approximations can offer increased accuracy, they often demand more function evaluations and can, unfortunately, suffer from numerical instability due to Runge’s phenomenon when using high degrees. One elegant solution to this issue is Clenshaw–Curtis quadrature , where the integrand is approximated by expanding it in terms of Chebyshev polynomials , which have superior convergence properties.

Romberg’s method is an adaptive technique that systematically halves the step widths incrementally, generating a sequence of trapezoidal approximations denoted by $T(h_0)$, $T(h_1)$, and so on, where $h_{k+1}$ is half of $h_k$. For each new, smaller step size, only half of the new function values actually need to be computed, as the others are carried over from previous calculations. The method then employs a process of polynomial interpolation through these approximations and extrapolates to the limit $T(0)$, which represents the theoretical exact integral. Gaussian quadrature is a particularly powerful and efficient numerical integration technique. Instead of using equally spaced points, it strategically evaluates the function at the roots of a set of orthogonal polynomials . An $n$-point Gaussian method is exact for polynomials of degree up to $2n - 1$, offering remarkable precision with a minimal number of function evaluations.

For the computation of higher-dimensional integrals (such as complex volume calculations), methods like Monte Carlo integration become extremely important. These probabilistic methods are especially effective in high dimensions where deterministic quadrature rules become computationally prohibitive. It’s not just about getting an answer, but getting a good answer efficiently.

Mechanical

In a more tangible, less abstract realm, the area of an arbitrary two-dimensional shape can be determined with surprising precision using a specialized measuring instrument known as a planimeter . This device, a marvel of mechanical engineering, traces the boundary of the shape and provides a direct reading of its area. Similarly, the volume of irregular three-dimensional objects, particularly those with complex geometries that defy simple formulas, can be measured with considerable accuracy by observing the fluid displaced when the object is carefully submerged in a liquid. These methods, while perhaps seemingly rudimentary compared to computational algorithms, represent practical applications of integral concepts.

Geometrical

Main article: Quadrature (mathematics)

Area, in some select cases, can be determined through ingenious geometrical compass-and-straightedge constructions that yield an equivalent square . This classical problem of “squaring” a given shape (finding a square with the same area) has a rich history, particularly in ancient Greek mathematics, and represents a foundational link between geometry and the concept of area that foreshadowed modern integration.

Integration by differentiation

Kempf, Jackson, and Morales have unveiled a series of intriguing mathematical relations that permit an integral to be calculated through the seemingly inverse operation of differentiation . Their innovative calculus framework incorporates the ubiquitous Dirac delta function and the partial derivative operator, specifically $\partial_x$. This conceptual inversion of operations is not limited to conventional integrals; it can also be powerfully applied to functional integrals , enabling their computation through the equally specialized technique of functional differentiation . It’s a rather elegant, if somewhat mind-bending, example of how seemingly opposite operations can be intertwined.

Examples

Using the fundamental theorem of calculus

Let’s illustrate the sheer utility of the fundamental theorem of calculus with a classic example, since some concepts require explicit demonstration. It allows for the remarkably straightforward calculation of integrals for even basic trigonometric functions:

$$\int {0}^{\pi }\sin(x),dx=-\cos(x){\Big |}{x=0}^{x=\pi }=-\cos(\pi )-{\Big (}-\cos(0){\Big )}=2.$$

Here, $-\cos(x)$ is the antiderivative of $\sin(x)$. By simply evaluating this antiderivative at the upper limit ($\pi$) and subtracting its value at the lower limit (0), we arrive at the precise result. No tedious Riemann sums , no infinitesimally thin rectangles; just a direct, elegant calculation. It’s almost disappointingly simple, once you know the trick.

See also

Notes

  • ^ Integral calculus is a very well established mathematical discipline for which there are many sources. See Apostol 1967 and Anton, Bivens & Davis 2016, for example.