{
  "id": "acceleration",
  "title": "Acceleration",
  "lang": "en",
  "source_lang": "en",
  "summary": "Acceleration is the rate of change of velocity with time, describing changes in an object’s speed, direction of motion, or both.",
  "category": "science",
  "subcategory": "physics",
  "type": "concept",
  "aliases": [],
  "generated_at": "2026-11-03",
  "model": "pro/gpt-6.1-sol",
  "url": "https://aiwiki.page/en/acceleration/",
  "translations": {
    "zh-Hans": {
      "url": "https://aiwiki.page/zh/acceleration/",
      "json": "https://aiwiki.page/zh/acceleration.json"
    }
  },
  "sections": [
    {
      "heading": "",
      "text": "Acceleration is the rate at which an object’s velocity changes with time. Because velocity specifies both speed and direction, an object accelerates whenever it speeds up, slows down, or changes direction. Acceleration is therefore a vector quantity: a car braking along a straight road and a satellite moving at constant speed around a circular orbit are both accelerating, although in different ways. Its unit in the International System of Units is the metre per second squared, written \\(\\mathrm{m\\,s^{-2}}\\).\n\nAcceleration connects the description of motion to its physical causes. In classical mechanics, Newton’s laws of motion relate an object’s acceleration to the forces acting on it and its mass. In relativity, a distinction between acceleration measured in a coordinate system and acceleration registered by an instrument carried with the object becomes essential. The concept also underlies the analysis of rotating machinery, vehicle motion, earthquakes, fluid flow, and particle beams."
    },
    {
      "heading": "Mathematical definition and physical meaning",
      "text": "For an object whose position relative to a chosen reference frame is \\(\\mathbf r(t)\\), velocity and instantaneous acceleration are successive time derivatives:\n\n\\[\\mathbf v(t)=\\frac{d\\mathbf r}{dt}, \\qquad \\mathbf a(t)=\\frac{d\\mathbf v}{dt} =\\frac{d^2\\mathbf r}{dt^2}.\\]\n\nHere \\(t\\) denotes time, and the derivatives describe the limiting rates of change at an instant. Average acceleration over a finite interval is instead\n\n\\[\\overline{\\mathbf a} =\\frac{\\mathbf v(t_2)-\\mathbf v(t_1)}{t_2-t_1}.\\]\n\nAverage acceleration need not equal the acceleration at any particular instant as a vector. For example, an object completing one full circular revolution returns to its initial velocity, giving zero average acceleration over the revolution even though its instantaneous acceleration is never zero.\n\nIn fixed Cartesian coordinates, the components are \\(a_x=d^2x/dt^2\\), \\(a_y=d^2y/dt^2\\), and \\(a_z=d^2z/dt^2\\). The magnitude is \\(a=\\sqrt{a_x^2+a_y^2+a_z^2}\\). Acceleration has dimensions of length divided by time squared, \\(LT^{-2}\\); an acceleration of \\(1\\ \\mathrm{m\\,s^{-2}}\\), maintained along a straight line in the direction of motion, increases speed by \\(1\\ \\mathrm{m\\,s^{-1}}\\) each second. The unit does not mean a distance travelled “per square second”; it expresses how quickly velocity changes.\n\nThe sign of an acceleration component depends on the chosen coordinate direction, not on whether the object is speeding up. A vehicle moving in the negative \\(x\\)-direction and accelerating in that same direction has negative velocity and negative acceleration while increasing its speed. More generally,\n\n\\[\\frac{d}{dt}\\left(\\frac{v^2}{2}\\right) =\\mathbf v\\cdot\\mathbf a.\\]\n\nA positive dot product indicates increasing speed, a negative one decreasing speed, and a zero value no instantaneous change in speed. At zero speed, this expression alone does not determine how speed changes immediately afterward. The term deceleration usually means acceleration that reduces speed, rather than acceleration with an intrinsically negative sign.\n\nThe rate of change of acceleration is called jerk, \\(\\mathbf j=d\\mathbf a/dt\\), with units \\(\\mathrm{m\\,s^{-3}}\\). Although many elementary problems assume constant acceleration, actual machines and vehicles often vary acceleration smoothly. Jerk then becomes important because two motions with the same maximum acceleration can impose quite different transient loads and sensations."
    },
    {
      "heading": "Historical development",
      "text": "The modern concept of acceleration emerged from efforts to replace qualitative classifications of motion with quantitative relationships among distance, time, and speed. In Aristotle’s natural philosophy, terrestrial and celestial motions belonged to different explanatory schemes, and motion through a medium was discussed through relationships among motive action, resistance, and the nature of bodies. This was not a formulation of instantaneous acceleration in the modern mathematical sense.\n\nDuring the fourteenth century, scholars associated with Merton College, Oxford, investigated uniformly changing motion. Their mean-speed theorem states, in modern terms, that an object undergoing uniformly accelerated motion travels the same distance in a given interval as an object moving throughout that interval at the average of its initial and final speeds. Nicole Oresme developed geometrical representations of changing qualities in which the area of a diagram could represent an accumulated quantity. Such constructions anticipated graphical methods for relating speed and distance, although they did not constitute a complete experimental dynamics.\n\nGalileo Galilei made uniformly accelerated motion central to the study of falling bodies. In Discourses and Mathematical Demonstrations Relating to Two New Sciences (1638), he presented the relationship between elapsed time and distance for motion from rest under uniform acceleration. In modern notation it is \\(s=\\tfrac12 at^2\\). Consequently, distances covered during successive equal time intervals are proportional to the odd numbers \\(1,3,5,\\ldots\\). Inclined planes slowed the motion sufficiently to make timing and comparison more practicable than direct observation of rapid vertical falls.\n\nGalileo’s work also separated the mathematical description of ideal motion from complications such as air resistance. Projectile motion could be analyzed by combining uniform horizontal motion with uniformly accelerated vertical motion, producing a parabola when resistance and changes in gravitational acceleration are neglected. This decomposition helped establish that different components of motion can be treated independently under specified conditions.\n\nIsaac Newton connected changes of motion to forces in Philosophiæ Naturalis Principia Mathematica (1687). The development of calculus by Newton and Gottfried Wilhelm Leibniz supplied systematic methods for treating instantaneous rates of change. Newton’s framework unified falling bodies and orbital motion within one theory, making acceleration a central quantity in mathematical physics rather than merely a feature of terrestrial falling.\n\nIn the twentieth century, Albert Einstein altered the interpretation of accelerated motion through special and general relativity. Acceleration remained measurable, but its relation to coordinates, time, and gravitation required a broader framework. In particular, free fall came to be understood as motion without proper acceleration, even when a chosen coordinate system assigns the falling object a changing velocity."
    },
    {
      "heading": "Motion under constant and variable acceleration",
      "text": "Kinematics describes motion without specifying the forces that produce it. For constant acceleration in a fixed Cartesian frame, integration gives\n\n\\[\\mathbf v(t)=\\mathbf v_0+\\mathbf a t,\\]\n\n\\[\\mathbf r(t)=\\mathbf r_0+\\mathbf v_0t+\\frac12\\mathbf a t^2,\\]\n\nwhere the initial time has been set to zero. In one-dimensional motion, eliminating time yields\n\n\\[v^2=v_0^2+2a(x-x_0).\\]\n\nThese equations apply only when acceleration is constant over the interval considered. A constant magnitude is not sufficient if the acceleration’s direction changes, as it does in uniform circular motion.\n\nFor a numerical example, a vehicle starting at \\(10\\ \\mathrm{m\\,s^{-1}}\\) and accelerating uniformly at \\(2\\ \\mathrm{m\\,s^{-2}}\\) for five seconds reaches \\(20\\ \\mathrm{m\\,s^{-1}}\\) and travels \\(75\\ \\mathrm m\\). Its average speed is \\(15\\ \\mathrm{m\\,s^{-1}}\\). If it instead brakes from \\(20\\ \\mathrm{m\\,s^{-1}}\\) at a constant acceleration of \\(-5\\ \\mathrm{m\\,s^{-2}}\\), with the positive axis chosen along its initial motion, it stops in four seconds after travelling \\(40\\ \\mathrm m\\). This idealized braking distance excludes reaction time and changes in tyre–road conditions.\n\nNear Earth’s surface, unobstructed free fall is often approximated by a constant downward acceleration. Taking upward as positive gives \\(a_y=-g\\). A projectile then obeys\n\n\\[x=x_0+v_{0x}t, \\qquad y=y_0+v_{0y}t-\\frac12gt^2.\\]\n\nThe horizontal and vertical equations share the same time variable. The approximation neglects aerodynamic forces, Earth’s rotation, and the variation of gravity with position. It is useful over modest distances but becomes inadequate for high-altitude trajectories or objects strongly affected by air resistance.\n\nWhen acceleration varies with time, velocity and position follow from integration:\n\n\\[\\mathbf v(t)=\\mathbf v_0+\\int_0^t\\mathbf a(\\tau)\\,d\\tau,\\]\n\n\\[\\mathbf r(t)=\\mathbf r_0+\\int_0^t\\mathbf v(\\tau)\\,d\\tau.\\]\n\nFor one-dimensional motion in which acceleration is expressed as a function of position, the chain rule gives \\(a=v\\,dv/dx\\). This is often useful when a force depends on displacement rather than explicitly on time.\n\nGraphs provide complementary interpretations. The slope of a velocity–time graph is acceleration, while the signed area under an acceleration–time graph is the change in velocity. Real trajectories frequently require numerical integration because acceleration depends simultaneously on position, velocity, and time. Computer simulations therefore update the state in small time steps; the choice of integration method and step size affects accumulated error and the preservation of physical properties such as energy."
    },
    {
      "heading": "Curved paths, circular motion, and oscillation",
      "text": "Acceleration along a curved path can be separated into changes of speed and changes of direction. Let \\(v\\) be speed, \\(\\mathbf T\\) the unit tangent to the path, \\(\\mathbf N\\) the principal normal directed toward the local center of curvature, and \\(\\rho\\) the radius of curvature. At points where these quantities are defined,\n\n\\[\\mathbf a=\\frac{dv}{dt}\\mathbf T+\\frac{v^2}{\\rho}\\mathbf N.\\]\n\nThe tangential component \\(dv/dt\\) changes speed. The normal component \\(v^2/\\rho\\) changes the direction of velocity. Their perpendicularity allows a moving object to have substantial acceleration without any instantaneous increase or decrease in speed.\n\nIn uniform circular motion, the radius \\(r\\) and speed are constant, and acceleration points toward the circle’s center:\n\n\\[a_{\\mathrm c}=\\frac{v^2}{r}=\\omega^2r,\\]\n\nwhere \\(\\omega\\) is angular speed. This inward acceleration is called centripetal acceleration. It is a description of the motion, not a separate type of interaction. A string’s tension, gravity, friction, or an electromagnetic force may supply the required inward force, depending on the system.\n\nFor example, an object moving at \\(10\\ \\mathrm{m\\,s^{-1}}\\) around a circle of radius \\(20\\ \\mathrm m\\) has a centripetal acceleration of \\(5\\ \\mathrm{m\\,s^{-2}}\\). Doubling its speed at the same radius raises that acceleration to \\(20\\ \\mathrm{m\\,s^{-2}}\\). The square dependence explains why curve negotiation becomes much more demanding as vehicle speed increases. Banking a road or railway curve allows a component of the support force to contribute to the inward acceleration.\n\nFor planar rotation with changing angular speed, a point fixed at radius \\(r\\) has tangential acceleration \\(r\\alpha\\), where \\(\\alpha=d\\omega/dt\\), as well as radial acceleration \\(\\omega^2r\\). In polar coordinates, the complete expressions are\n\n\\[a_r=\\ddot r-r\\dot\\theta^2, \\qquad a_\\theta=r\\ddot\\theta+2\\dot r\\dot\\theta.\\]\n\nThe extra terms arise because the radial and angular unit vectors themselves change direction. Differentiating only the scalar coordinate rates would omit these contributions.\n\nIn simple harmonic motion, acceleration is proportional to displacement from equilibrium and directed oppositely:\n\n\\[a=-\\omega^2x.\\]\n\nA solution is \\(x=A\\cos(\\omega t+\\phi)\\). Acceleration has its greatest magnitude at the turning points, where speed is zero, and vanishes at equilibrium, where speed is greatest. An ideal mass–spring system has \\(\\omega=\\sqrt{k/m}\\), with spring constant \\(k\\) and mass \\(m\\). Real oscillators also experience damping and external driving, which change the acceleration law and can produce resonance when driving frequencies approach a system’s natural frequencies."
    },
    {
      "heading": "Forces, mass, and energy transfer",
      "text": "In an inertial reference frame, Newtonian dynamics relates acceleration to the net external force. For a particle of constant mass,\n\n\\[\\sum\\mathbf F=m\\mathbf a.\\]\n\nThe more general momentum statement is \\(\\sum\\mathbf F=d\\mathbf p/dt\\), applied to an appropriately defined system. A force of one newton gives a one-kilogram mass an acceleration of \\(1\\ \\mathrm{m\\,s^{-2}}\\). Forces must be added vectorially: a large individual force need not imply a large acceleration if other forces nearly balance it.\n\nA book resting on a table illustrates this distinction. Gravity acts downward, while the table exerts an upward contact force; their balance gives zero coordinate acceleration in the usual stationary laboratory approximation. When a support is removed, the unbalanced gravitational force accelerates the book downward. Conversely, an object moving with constant velocity needs no net force to maintain that velocity. This differs from everyday circumstances in which continuing propulsion is necessary to counter friction or drag.\n\nThe relation to energy is expressed through power. For a constant-mass particle,\n\n\\[\\frac{dK}{dt} =\\mathbf F_{\\mathrm{net}}\\cdot\\mathbf v =m\\mathbf a\\cdot\\mathbf v,\\]\n\nwhere \\(K=\\tfrac12mv^2\\) is kinetic energy. Only the force component parallel to the instantaneous velocity changes kinetic energy. An ideal centripetal force remains perpendicular to velocity and changes direction without doing work. Acceleration therefore does not by itself imply an increase in kinetic energy.\n\nResistance often makes acceleration depend on speed. With downward taken as positive, a simple model of a falling body under quadratic drag is\n\n\\[m\\frac{dv}{dt}=mg-cv^2.\\]\n\nThe acceleration decreases as speed rises and approaches zero at terminal speed \\(v_{\\mathrm t}=\\sqrt{mg/c}\\), provided the parameters remain approximately constant. The body continues falling at terminal speed because gravity and drag balance, not because gravity has ceased to act. At sufficiently low Reynolds number, a linear dependence on speed is more appropriate than quadratic drag.\n\nFor extended systems, the center of mass obeys \\(M\\mathbf a_{\\mathrm{CM}}=\\mathbf F_{\\mathrm{ext}}\\) when total mass is constant. Internal forces can produce complicated motions without changing the center-of-mass acceleration. Variable-mass problems require additional care: a rocket accelerates by exchanging momentum with expelled propellant, and applying \\(F=ma\\) to the remaining rocket without accounting for momentum flux gives an incomplete equation."
    },
    {
      "heading": "Reference frames and apparent forces",
      "text": "Acceleration is always described relative to a reference frame, but its transformation depends on the relationship between frames. In Newtonian mechanics, two nonrotating frames moving at constant velocity relative to one another are related by a Galilean transformation. Their measured velocities differ by a constant, while their measured accelerations agree. Thus acceleration, unlike velocity, is invariant under a Galilean boost.\n\nIf a nonrotating frame’s origin has acceleration \\(\\mathbf A\\) relative to an inertial frame, an object’s acceleration in that frame is \\(\\mathbf a'=\\mathbf a-\\mathbf A\\). Newton’s second law can retain its familiar form in the accelerating frame if an apparent force \\(-m\\mathbf A\\) is included. Passengers in an accelerating vehicle may describe a backward apparent force, while an observer in an approximately inertial ground frame attributes their forward acceleration to contact forces exerted by seats or other supports.\n\nRotation introduces additional terms. Let \\(\\mathbf r'\\), \\(\\mathbf v'\\), and \\(\\mathbf a'\\) denote position, velocity, and acceleration measured in a frame rotating with angular velocity \\(\\boldsymbol\\Omega\\). Expressing all vectors in a common basis, the inertial acceleration is\n\n\\[\\mathbf a=\\mathbf A+\\mathbf a' +2\\boldsymbol\\Omega\\times\\mathbf v' +\\dot{\\boldsymbol\\Omega}\\times\\mathbf r' +\\boldsymbol\\Omega\\times (\\boldsymbol\\Omega\\times\\mathbf r').\\]\n\nThe terms represent origin acceleration, relative acceleration, Coriolis acceleration, angular acceleration of the frame, and the rotational contribution associated with centripetal acceleration. Moving the frame-dependent terms to the force side introduces the corresponding apparent forces.\n\nThe centrifugal apparent force points outward from the rotation axis. The Coriolis effect depends on the object’s velocity relative to the rotating frame and acts perpendicular to that velocity. It influences large-scale atmospheric and oceanic circulation on the rotating Earth, while being negligible in many small laboratory motions. The Euler apparent force arises when the frame’s angular velocity changes.\n\nA person stationary on Earth has a small centripetal acceleration relative to a suitable nonrotating frame because Earth rotates. At the equator its magnitude is about \\(0.034\\ \\mathrm{m\\,s^{-2}}\\). This rotational effect, together with Earth’s shape and mass distribution, contributes to geographical variation in effective gravity. Consequently, “at rest” and “zero acceleration” are not interchangeable statements unless the reference frame is specified.\n\nApparent forces are useful descriptions rather than additional fundamental interactions. Their magnitudes depend on the chosen frame. Nevertheless, the stresses and contact forces required to keep a person or instrument stationary in an accelerating frame are physically measurable."
    },
    {
      "heading": "Relativity, proper acceleration, and gravitation",
      "text": "In special relativity, coordinate acceleration remains \\(d\\mathbf v/dt\\), but different inertial observers generally assign different values to it. Time and space coordinates transform together, and acceleration no longer has the Galilean invariance of Newtonian mechanics. An additional quantity, proper acceleration, describes acceleration measured in an object’s instantaneous rest frame and corresponds to what an ideal onboard accelerometer registers.\n\nLet \\(\\tau\\) denote proper time along an object’s path through spacetime. Its four-velocity is \\(U^\\mu=dx^\\mu/d\\tau\\), and in flat spacetime its four-acceleration is \\(A^\\mu=dU^\\mu/d\\tau\\). With metric signature \\((-+++)\\), the magnitude of proper acceleration is\n\n\\[\\alpha=\\sqrt{A^\\mu A_\\mu}.\\]\n\nFour-acceleration is orthogonal to four-velocity. This geometrical relationship encodes a distinction from ordinary three-dimensional motion: a massive object’s four-velocity has fixed spacetime norm even while its spatial velocity changes.\n\nFor acceleration parallel to velocity,\n\n\\[\\alpha=\\gamma^3 a_{\\parallel}, \\qquad \\gamma=\\frac{1}{\\sqrt{1-v^2/c^2}},\\]\n\nwhere \\(c\\) is the speed of light. For a purely perpendicular coordinate acceleration, \\(\\alpha=\\gamma^2a_{\\perp}\\). Consequently, constant proper acceleration does not imply constant coordinate acceleration. An object starting from rest and maintaining constant proper acceleration \\(\\alpha\\) along a straight line has\n\n\\[v(t)=\\frac{\\alpha t}{\\sqrt{1+(\\alpha t/c)^2}}.\\]\n\nIts speed approaches \\(c\\) without reaching it. Newtonian constant-acceleration motion is recovered when speeds are much smaller than \\(c\\).\n\nIn general relativity, freely falling test bodies follow geodesics when nongravitational forces are absent. Their four-acceleration, defined using a covariant derivative, is zero. A dropped accelerometer therefore reads approximately zero during free fall, even though an Earth-fixed coordinate system assigns it a downward acceleration. An accelerometer supported on a table instead registers an upward proper acceleration, because the support prevents it from following a free-fall path.\n\nThe equivalence principle relates this local behavior to acceleration in the absence of gravity. A sufficiently small freely falling laboratory can eliminate the local effects of a uniform gravitational field, but not tidal effects across an extended region. Nearby freely falling bodies can accelerate relative to one another because spacetime is curved; this is described by geodesic deviation.\n\nThe word acceleration also appears in cosmology, where accelerated expansion means that the cosmological scale factor has a positive second derivative with respect to cosmic time. This is a statement about the evolution of spatial distances in a cosmological model, not a claim that every galaxy registers a nonzero onboard accelerometer reading."
    },
    {
      "heading": "Measurement, instruments, and uncertainty",
      "text": "Acceleration can be determined from observations of position or velocity, or measured through an instrument’s response to force. Tracking systems use cameras, radar, laser interferometry, or satellite navigation to estimate motion. Numerical differentiation then produces acceleration. For equally spaced position samples, a central approximation is\n\n\\[a(t)\\approx \\frac{x(t+\\Delta t)-2x(t)+x(t-\\Delta t)} {\\Delta t^2}.\\]\n\nAlthough simple, this operation can strongly amplify measurement noise. Reducing the sampling interval does not automatically improve the result, because position errors are divided by the square of that interval.\n\nAn accelerometer commonly contains a proof mass coupled to its housing by an elastic or controlled suspension. When the housing accelerates, the proof mass’s displacement or the force required to hold it in place provides a signal. In a simple spring model, displacement is approximately proportional to acceleration at frequencies well below the instrument’s mechanical resonance. Capacitive sensors detect changes in electrode spacing, piezoresistive sensors detect stress-dependent electrical resistance, and piezoelectric sensors generate charge under applied stress.\n\nMicroelectromechanical accelerometers are widely used in phones, vehicles, and inertial instruments. Three sensing axes allow the measured acceleration-related signal to be resolved in three dimensions. Piezoelectric devices are particularly useful for vibration and shock, but their electrical readout generally does not retain a truly static acceleration signal indefinitely. Instrument selection therefore depends on frequency range as well as sensitivity and maximum measurable acceleration.\n\nA crucial distinction is that an accelerometer ordinarily measures specific force rather than gravitational coordinate acceleration directly. In a Newtonian navigation model, its ideal signal can be written schematically as\n\n\\[\\mathbf f=\\mathbf a-\\mathbf g,\\]\n\nafter expressing the vectors in a common frame and accounting separately for frame rotation where necessary. A stationary supported instrument has a nonzero signal; an ideal freely falling one has approximately zero. Navigation systems combine this signal with orientation estimates and gravity models to reconstruct motion.\n\nCalibration establishes scale factors, offsets, axis misalignment, and other characteristics. Gravity can provide a convenient reference through controlled changes of orientation, while centrifuges and vibration exciters supply other acceleration inputs. Optical measurements can relate a vibrating platform’s displacement and frequency to acceleration. For sinusoidal displacement \\(x=X\\sin(2\\pi ft)\\), peak acceleration is \\((2\\pi f)^2X\\).\n\nMeasurement uncertainty includes sensor noise, bias drift, temperature dependence, cross-axis sensitivity, mounting effects, and timing errors. Integrating acceleration twice to estimate position makes persistent errors particularly serious: a constant uncorrected acceleration bias \\(b\\) produces a position error of \\(\\tfrac12bt^2\\). Inertial navigation therefore often incorporates independent observations to constrain accumulated drift."
    },
    {
      "heading": "Engineering and scientific applications",
      "text": "Transportation engineering uses acceleration to characterize propulsion, braking, cornering, and ride quality. Longitudinal acceleration determines how rapidly a vehicle changes speed, while lateral acceleration constrains its passage through curves. For a level road and a simplified friction limit, the available lateral acceleration is of order \\(\\mu g\\), where \\(\\mu\\) is the relevant tyre–road friction coefficient. Actual limits depend on tyre behavior, load transfer, road condition, suspension, and combined braking and steering demands.\n\nAcceleration is often expressed in multiples of the conventional standard gravity \\(g_{\\mathrm n}=9.80665\\ \\mathrm{m\\,s^{-2}}\\), an exact reference value rather than a universal local gravitational acceleration. A quoted load of “several g” generally concerns specific force or load factor, so its meaning depends on the measurement convention. In aerospace structures and equipment qualification, the direction, duration, and frequency content of an acceleration can be as important as its peak magnitude. Human responses likewise depend on direction, duration, posture, and restraint, so a peak value alone does not characterize an exposure.\n\nStructural engineering examines how buildings, bridges, and machines respond to accelerating supports and varying forces. A common single-degree-of-freedom model is\n\n\\[m\\ddot x+c\\dot x+kx=-m\\ddot y,\\]\n\nwhere \\(x\\) is displacement relative to a moving base, \\(y\\) is base displacement, \\(c\\) is damping, and \\(k\\) is stiffness. The base acceleration \\(\\ddot y\\) acts as the excitation. Earthquake recordings and vibration measurements help determine whether the excitation contains frequencies capable of producing large structural responses; peak ground acceleration alone cannot describe the full loading history.\n\nIn fluids, acceleration must account for motion through a spatially varying velocity field. If \\(\\mathbf u(\\mathbf x,t)\\) is fluid velocity, the acceleration of a moving fluid parcel is the material derivative\n\n\\[\\frac{D\\mathbf u}{Dt} =\\frac{\\partial\\mathbf u}{\\partial t} +(\\mathbf u\\cdot\\nabla)\\mathbf u.\\]\n\nThe first term describes local time variation, while the second describes changes encountered as the parcel moves between locations. Thus steady flow can still be accelerated flow: water speeding through a narrowing channel or changing direction around a bend has nonzero material acceleration.\n\nIn a particle accelerator, electric fields change charged particles’ energy, while magnetic fields commonly guide and focus their paths. The Lorentz force is \\(q(\\mathbf E+\\mathbf v\\times\\mathbf B)\\). Its magnetic part is perpendicular to velocity and does no work, yet can produce large changes in direction. At relativistic speeds, substantial energy increases produce increasingly small changes in speed.\n\nCharged particles undergoing suitable accelerated motion also emit electromagnetic radiation. In circular electron accelerators, synchrotron radiation creates an important energy loss and is also exploited as an intense research light source. Accelerator design must therefore treat acceleration as more than a change in speed: trajectory curvature, beam focusing, radiative losses, and relativistic momentum all enter the equations of motion."
    }
  ],
  "links": [
    "velocity",
    "international-system-of-units",
    "classical-mechanics",
    "newtons-laws-of-motion",
    "derivative",
    "jerk",
    "aristotle",
    "galileo-galilei",
    "isaac-newton",
    "albert-einstein",
    "integral",
    "inertial-reference-frame",
    "force",
    "mass",
    "newton-unit",
    "kinetic-energy",
    "work-physics",
    "center-of-mass",
    "momentum",
    "coriolis-effect",
    "earth",
    "special-relativity",
    "spacetime",
    "speed-of-light",
    "general-relativity",
    "geodesic",
    "equivalence-principle",
    "cosmology",
    "calibration",
    "measurement-uncertainty",
    "particle-accelerator",
    "lorentz-force",
    "electromagnetic-radiation",
    "synchrotron-radiation"
  ],
  "sources": [],
  "license": "https://creativecommons.org/licenses/by/4.0/"
}