Quartz flexible surface temperature compensation method

Abstract

The invention discloses a quartz flexibility accelerometer temperature compensation method, and belongs to the technical field of quartz flexibility accelerometer temperature compensation, and the method comprises the steps: constructing a temperature compensation model which considers the temperature, the temperature change rate and the sensitive axis direction temperature space gradient influence at the same time; identifying a coefficient of the temperature compensation model by using a super Latin square fractional inferior particle swarm algorithm, and obtaining a quartz flexibility plus surface temperature compensation model considering the temperature space gradient in the sensitive axis direction based on the coefficient; and according to real-time temperature measurement data of the quartz flexible accelerometer in the sensitive axis direction, determining a temperature compensation result of the quartz flexible accelerometer by using the temperature compensation model. According to the method, the problem that a traditional temperature compensation model and a compensation method are poor in precision is solved, the accuracy and the efficiency of determining the temperature compensation result are improved, and the measurement precision and the measurement speed of the large space carrier quartz flexibility meter are greatly improved.

Description

Technical Field

[0001] This application relates to the technical field of temperature compensation for quartz flexible accelerometers, and particularly to a temperature compensation method for quartz flexible accelerometers. Background Art

[0002] A quartz flexible accelerometer is a key sensitive device for the navigation and guidance of large space flight vehicles, mainly composed of a pendulum piece, a differential capacitance sensor, a torque motor, a permanent magnet, a yoke, and a housing. The measurement accuracy of the quartz flexible accelerometer is greatly affected by the ambient temperature. To ensure the navigation and guidance accuracy of large space flight vehicles, a temperature control system is usually configured for it during actual use to ensure that it operates at a relatively stable ambient temperature. Therefore, before the launch of a large space vehicle using a quartz flexible accelerometer, it usually needs to go through a certain heating process. Although this ensures the navigation and guidance accuracy of the large space flight vehicle, it also prolongs the launch preparation time of the large space flight vehicle and affects its rapid response ability.

[0003] In the prior art, temperature error compensation technology has been proposed and gradually popularized. This technology first establishes a temperature error model of the quartz flexible accelerometer based on test data, and then compensates the output of the accelerometer in real time according to the monitored ambient temperature to meet the navigation accuracy within its full temperature range. Compared with the traditional temperature control method, the temperature error compensation technology for accelerometers can avoid structural redundancy, save heating time, and can also meet the navigation accuracy requirements after compensation, and has received extensive attention at present.

[0004] However, most of the currently established temperature error compensation models for accelerometers only consider the time-varying influence of temperature and insufficiently consider the temperature spatial gradient in the sensitive axis direction of the quartz flexible accelerometer. Affected by the working environment characteristics of the quartz flexible accelerometer, there are often differences in the temperatures on both sides of the sensitive axis of the quartz pendulum piece of the accelerometer under some typical conditions, which results in poor accuracy of the traditional temperature compensation model and compensation method. Summary of the Invention

[0005] Aiming at the above deficiencies in the prior art, a temperature compensation method for a quartz flexible accelerometer provided by this application constructs a temperature compensation model based on the temperature spatial gradient in the sensitive axis direction of the quartz flexible accelerometer, solves the problem of poor accuracy of the traditional temperature compensation model and compensation method, and improves the measurement accuracy and measurement speed of the quartz flexible accelerometer of large space vehicles.

[0006] To achieve the above invention purpose, the technical solution adopted by this application is: A temperature compensation method for a quartz flexible accelerometer provided by this application includes: S1: Construct a temperature compensation model that simultaneously considers the influences of temperature, temperature change rate, and temperature spatial gradient in the sensitive axis direction; S2: Identify the coefficients of the temperature compensation model using the hyper-Latin square generation-elimination particle swarm optimization algorithm, and obtain a quartz flexible accelerometer temperature compensation model considering the temperature spatial gradient in the sensitive axis direction based on the coefficients. S3: Determine the compensation result of the quartz flexible accelerometer temperature according to the real-time temperature measurement data of the quartz flexible accelerometer in the sensitive axis direction and using the quartz flexible accelerometer temperature compensation model.

[0007] Further, in the S1, the temperature compensation model is:

[0008] where is the error caused by the temperature change on the A side of the quartz flexible accelerometer; is the measured temperature value on the A side inside the quartz flexible accelerometer; is the fifth-order coefficient of the temperature on the A side; is the fourth-order coefficient of the temperature on the A side; is the third-order coefficient of the temperature on the A side; is the second-order coefficient of the temperature on the A side; is the first-order coefficient of the temperature on the A side; is the first-order coefficient of the temperature change rate on the A side; is the second-order coefficient of the temperature change rate on the A side, is the constant coefficient; is the error caused by the temperature change on the B side of the quartz flexible accelerometer; is the measured temperature value on the B side inside the quartz flexible accelerometer; is the fifth-order coefficient of the temperature on the B side; is the fourth-order coefficient of the temperature on the B side; is the third-order coefficient of the temperature on the B side; is the second-order coefficient of the temperature on the B side; is the first-order coefficient of the temperature on the B side; is the first-order coefficient of the temperature change rate on the B side; is the second-order coefficient of the temperature change rate on the B side, is the constant coefficient; is a function used to describe the temperature change on the A side of the quartz flexible accelerometer; is a function used to describe the temperature change on the B side of the quartz flexible accelerometer; is the temperature measurement time; 、 represent the temperature change rates of the temperatures on the A side and the B side, 、 represent the temperature change acceleration rates of the temperatures on the A side and the B side, is the total error.

[0009] Further, the S2 specifically includes: S201: Initialize the population; S202: Construct the fitness function:

[0010] where is to minimize the cumulative sum of squares of temperature errors, is the sum of squares of temperature errors, E(t) is the temperature error compensation term, is the true temperature, is the temperature obtained by testing, is the th coefficient to be solved in the compensation model, , is the lower and upper limits of the search range of the th parameter, is the time; S203: Calculate the fitness values of all random particles in the population using the fitness function, and take the individual fitness extremum of each particle in the population as the individual optimal solution, and determine the current global optimal solution from all the individual optimal solutions; S204: Compare the current global optimal solution with the historical global optimal solution to obtain a comparison result; if the current global optimal solution is better than the historical global optimal solution, update the particle position and velocity, calculate the fitness of the updated particle, and determine the individual optimal solution and the current global optimal solution, where the update of the particle position and velocity is:

[0011]

[0012] where is the inertia weight, , are the acceleration constants, , are random numbers in [0, 1], is the particle at the th generation, the th dimension of the individual extremum, is the particle at the th generation, the th dimension of the global extremum, is the particle at the th generation, the th dimension of the position, is the particle at the th generation, the th dimension of the velocity, is the particle at the generation, the -dimensional position, is the particle at the generation, the -dimensional velocity; S205: When the minimum fitness value of all random particles in the population satisfies: , reduce the population size until the termination condition is met, where is the minimum fitness value of the fitness function of the initial generation particles, is the y -th generation of the minimum fitness value of the fitness function of the particles, is a constant; S206: Use the particle parameters corresponding to the global optimal solution as the coefficients of the temperature compensation model; S207: According to the coefficients, obtain a quartz flexure accelerometer temperature compensation model considering the temperature space gradient in the sensitive axis direction.

[0013] Furthermore, the initialization of the population in S201 specifically includes: S2011: Initialize the population parameters, including: initialize the number of iterations, population size, particle dimension, the search interval of each dimension variable , fitness function, variable value range, particle swarm; S2012: Use the Latin hypercube method to sample the initial sample population to obtain uniformly distributed random initial particles.

[0014] Furthermore, the use of the Latin hypercube method to sample the initial sample population in S2012 to obtain uniformly distributed random initial particles specifically includes: S20121: Determine the size H of the initial particle swarm; S20122: Divide the search interval of each dimension variable into H equally spaced intervals to form H hypercubes; S20123: Based on the initial particle swarm size H and the equally spaced intervals, generate a matrix A of H×16, where each column of the matrix A is a random permutation of the sequence {1, 2,..., H}; S20124: When only one small hypercube is selected in each row of the matrix A, generate a particle in each small hypercube. The coordinates of the j-th particle in the 16-dimensional space are ( ), where the -th dimension of the -th particle is:

[0015] where 1 ≤ ≤ 16, 1 ≤ ≤ , is the initial subgroup size, , are the interval endpoints of each dimension variable , is the length of each sub - interval, is the th random offset of the particle within the sub - interval, is matrix, is the th dimension of the th particle; S20125: Each particle forms an initial particle.

[0016] Furthermore, the termination conditions are: the current global optimal solution of the particle fitness in the population reaches a preset value; the number of algorithm loops reaches a preset value.

[0017] The beneficial effects of this application are: Based on the constructed temperature compensation model considering the influence of the temperature spatial gradient in the sensitive axis direction, this application takes into account the temperature differences on both sides of the sensitive axis direction of the quartz pendulum chip under some typical states, improving the accuracy and precision of the model. And by using the Latin - hypercube - based generation - eliminating particle swarm algorithm to identify the coefficients of the temperature compensation model, it avoids the calculation falling into local optimum during the coefficient identification process and improves the calculation efficiency. In addition, combined with the test data of the temperature cyclic variation in the sensitive axis direction of the quartz flexure accelerometer, a high - precision temperature compensation model of the quartz flexure accelerometer for large - scale space vehicles can be obtained, which can greatly improve the measurement accuracy and measurement speed of the quartz flexure accelerometer for large - scale space vehicles and promote the improvement of the flight trajectory and attitude control capabilities of large - scale space vehicles. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present application, and those of ordinary skill in the art can also obtain other embodiments based on these drawings.

[0019] Figure 1 is a flowchart of a temperature compensation method for a quartz flexure accelerometer provided by an embodiment of the present application.

[0020] Figure 2 is a layout diagram of temperature - measuring sensors in the sensitive axis direction of a quartz flexure accelerometer for a large - scale space vehicle provided by an embodiment of the present application.

[0021] Figure 3 Schematic diagram for realizing temperature compensation of a quartz flexible accelerometer provided by an embodiment of the present application for a large space vehicle. Detailed implementation manners

[0022] Next, the technical solutions in the embodiments of the present application will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art based on the present application belong to the scope of protection of the present application.

[0023] An embodiment of the present application provides a method for temperature compensation of a quartz flexible accelerometer, which can be applied to any quartz flexible accelerometer of a large space vehicle. The method can be referred to Figure 1 , Figure 1 The flowchart of a method for temperature compensation of a quartz flexible accelerometer provided by an embodiment of the present application is shown as follows, including: S1: Construct a temperature compensation model that simultaneously considers the influences of temperature, temperature change rate, and temperature spatial gradient in the sensitive axis direction.

[0024] Further, the temperature compensation model is:

[0025] where, is the error caused by the temperature change on the A side of the quartz flexible accelerometer; is the measured value of the temperature on the A side inside the quartz flexible accelerometer; is the fifth-order coefficient of the temperature on the A side; is the fourth-order coefficient of the temperature on the A side; is the third-order coefficient of the temperature on the A side; is the second-order coefficient of the temperature on the A side; is the first-order coefficient of the temperature on the A side; is the first-order coefficient of the temperature change rate on the A side; is the second-order coefficient of the temperature change rate on the A side, is the constant coefficient; is the error caused by the temperature change on the B side of the quartz flexible accelerometer; is the measured value of the temperature on the B side inside the quartz flexible accelerometer; is the fifth-order coefficient of the temperature on the B side; is the fourth-order coefficient of the temperature on the B side; is the third-order coefficient of the temperature on the B side; is the second-order coefficient of the temperature on the B side; is the first-order coefficient of the temperature on the B side; is the first-order coefficient of the temperature change rate on the B side; is the second-order coefficient of the temperature change rate on the B side, is the constant coefficient; is a function used to describe the temperature change on the A side of the quartz flexible accelerometer; is a function used to describe the temperature change on the B side of the quartz flexible accelerometer; is the temperature measurement time; and represent the temperature change rates of the temperatures on the A side and the B side, and represent the temperature change acceleration rates of the temperatures on the A side and the B side, is the total error.

[0026] In a possible embodiment, the model simultaneously considers the influences of temperature, temperature change rate, and the influence of the temperature spatial gradient in the sensitive axis direction on the test error of the quartz flexible accelerometer. Among them, for the temperature parameter on the A side in the sensitive axis direction of the quartz flexible accelerometer, a fifth-order model is constructed to finely depict the influence of the temperature on the A side on the test error of the quartz flexible accelerometer; for the temperature change rate on the A side in the sensitive axis direction of the quartz flexible accelerometer, a second-order model is constructed to characterize the influence of the temperature change rate on the A side in the sensitive axis direction of the quartz flexible accelerometer on the test error of the quartz flexible accelerometer. Similarly, for the temperature parameter on the B side in the sensitive axis direction of the quartz flexible accelerometer, a fifth-order model is constructed to finely depict the influence of the temperature on the B side on the test error of the quartz flexible accelerometer; for the temperature change rate on the B side in the sensitive axis direction of the quartz flexible accelerometer, a second-order model is constructed to characterize the influence of the temperature change rate on the B side in the sensitive axis direction of the quartz flexible accelerometer on the test error of the quartz flexible accelerometer.

[0027] S2: Identify the coefficients of the temperature compensation model using the super Latin square generation elimination particle swarm algorithm, and obtain the temperature compensation model of the quartz flexible accelerometer considering the temperature spatial gradient in the sensitive axis direction based on the coefficients.

[0028] Since the temperature error compensation model constructed above contains 16 unknown parameter values and the parameter variation range is relatively wide, in the process of using the PSO algorithm for parameter identification, if the initial population is large, it can better avoid the identification process falling into the local optimal value, but it may lead to a slower parameter identification process; if the initial population is too small, it may lead to the parameter identification process falling into the local optimal solution. Based on this, in order to effectively improve the speed and accuracy of the parameter identification of the temperature error compensation model, the super Latin square generation elimination particle swarm algorithm is used to identify the coefficients of the temperature compensation model.

[0029] Furthermore, the S2 specifically includes: S201: Initialize the population.

[0030] Further, the initialization of the population in S201 specifically includes: S2011: Initialize the population parameters, including: initialize the number of iterations, population size, particle dimension, search interval of each dimension variable , fitness function, variable value range, and particle swarm; S2012: Sample the initial sample population using the Latin hypercube method to obtain uniformly distributed random initial particles.

[0031] Further, the sampling of the initial sample population using the Latin hypercube method in S2012 to obtain uniformly distributed random initial particles specifically includes: S20121: Determine the scale H of the initial particle swarm; S20122: Divide the search interval of each dimension variable into H equally spaced intervals to form H hypercubes; In a possible embodiment, divide the search interval of each dimension variable into H equally spaced intervals to form H hypercubes, thus dividing the entire 16-dimensional search space into search interval of each dimension variable hypercubes.

[0032] S20123: Generate a matrix A of H×16 based on the scale H of the initial particle swarm and the equally spaced intervals, where each column of the matrix A is a random full permutation of the sequence {1, 2,..., H}; S20124: When only one small hypercube is selected in each row of the matrix A, generate a particle within each small hypercube. The coordinates of the th particle in the 16-dimensional space are ( ), where the th dimension of the th particle is:

[0033] where 1 ≤ ≤ 16, 1 ≤ ≤ , is the scale of the initial subgroup, , are the interval endpoints of each dimension variable , is the length of each sub-interval, is the random offset of the th particle within the sub-interval, is matrix, is the The th particle of dimension; S20125: Each particle is formed into an initial particle.

[0034] In a possible embodiment, the initial sample population is sampled by using the Latin hypercube method to obtain uniformly distributed random initial particles. H samples can be uniformly selected within the entire search space, and the samples are randomly distributed within each hypercube, which not only ensures the uniformity of the samples but also ensures the randomness of the samples, avoiding the parameter identification process from falling into local optimality.

[0035] S202: Construct a fitness function:

[0036] Wherein, is the cumulative sum of squares for minimizing the temperature error, is the sum of squares of the temperature error, E(t) is the temperature error compensation term, is the true temperature, is the temperature obtained by testing, is the th coefficient to be solved in the compensation model, 、 is the lower and upper limits of the search range of the th parameter, is the time.

[0037] S203: Calculate the fitness values of all random particles in the population by using the fitness function, and take the individual fitness extreme value of each particle in the population as the individual optimal solution, and determine the current global optimal solution from all the individual optimal solutions.

[0038] In a possible embodiment, substitute the 16 random parameters of the generated random particles into the fitness function determined in step S202 to calculate the fitness values of all random particles in the population, and determine the individual fitness extreme value of each particle in the population as the individual optimal solution, and determine the fitness extreme value of all particles as the global optimal solution.

[0039] S204: Compare the current global optimal solution with the historical global optimal solution to obtain a comparison result; if the current global optimal solution is better than the historical global optimal solution, update the particle position and velocity, calculate the fitness of the updated particle, and determine the individual optimal solution and the current global optimal solution, where the update of the particle position and velocity is:

[0040]

[0041] Wherein, is the inertia weight, , is the acceleration constant, , is a random number within [0, 1], is the particle at the th generation, the individual extremum of the th dimension, is the particle at the th generation, the global extremum of the th dimension, is the particle at the th generation, the position of the th dimension, is the particle at the th generation, the velocity of the th dimension, is the particle at the th generation, the position of the th dimension, is the particle at the th generation, the velocity of the th dimension.

[0042] S205: When the minimum fitness value of all random particles in the population satisfies: , reduce the population size until the termination condition is met, where is the minimum fitness value of the fitness function of the initial generation of particles, is the minimum fitness value of the fitness function of the y th generation of particles, is a constant.

[0043] Furthermore, the termination condition is: the current global optimal solution of the particle fitness in the population reaches a preset value; the number of algorithm loops reaches a preset value.

[0044] In a possible embodiment, to improve the parameter identification speed and avoid the particles being too dense in the later stage of the search due to a too large population, when the fitness function satisfies the requirement, eliminate the particles with the lowest 5% fitness in the population, and stop the elimination when the global optimal solution of the particle fitness in the population reaches a preset value or the number of algorithm loops reaches a preset value.

[0045] S206: Use the particle parameters corresponding to the global optimal solution as the coefficients of the temperature compensation model.

[0046] S207: Obtain a temperature compensation model for the quartz flexible accelerometer considering the temperature spatial gradient in the sensitive axis direction according to the said coefficient.

[0047] S3: Determine the compensation result of the temperature of the quartz flexible accelerometer by using the temperature compensation model of the quartz flexible accelerometer based on the real-time temperature measurement data of the quartz flexible accelerometer in the sensitive axis direction.

[0048] In a possible embodiment, import the established temperature compensation model of the quartz flexible accelerometer considering the temperature spatial gradient in the sensitive axis direction into the temperature compensation model operation unit. When in use, in the sensitive axis direction of the quartz flexible accelerometer, assemble temperature sensors according to the test layout scheme during parameter identification to measure the real-time temperatures of the A side and B side of the quartz flexible accelerometer in the sensitive axis direction respectively. and , for the temperature sensor layout scheme, reference can be made to Figure 2 , Figure 2 , which is a layout diagram of the temperature measurement sensors in the sensitive axis direction of the quartz flexible accelerometer of a large space vehicle provided by the embodiment of the present application. And synchronously output the temperature measurement data and the output data of the quartz flexible accelerometer to the built-in operation unit. After being compensated by the built-in operation unit, high-precision acceleration information is obtained and output to the flight control computer of the large space vehicle for navigation calculation and attitude control, as shown in the appendix Figure 3 shown, Figure 3 , which is a schematic diagram of the temperature compensation implementation of the quartz flexible accelerometer of a large space vehicle provided by the embodiment of the present application.

[0049] Based on the constructed temperature compensation model of the quartz flexible accelerometer considering the temperature spatial gradient in the sensitive axis direction and the method for identifying the coefficients of the temperature compensation model based on the super Latin square generation elimination particle swarm, combined with the test data of the cyclic temperature change in the sensitive axis direction of the quartz flexible accelerometer, a high-precision temperature compensation model of the quartz flexible accelerometer of a large space vehicle can be obtained. Based on this, combined with the introduction of the built-in operation unit, a scientific layout of the temperature measurement sensors in the sensitive axis direction and a high-precision temperature measurement circuit, the measurement accuracy and measurement speed of the quartz flexible accelerometer of a large space vehicle can be greatly improved, promoting the improvement of the flight trajectory and attitude control capabilities of the large space vehicle.

[0050] Each embodiment in this specification is described in a related manner. For the same or similar parts among the embodiments, reference can be made to each other. The key point of each embodiment is to illustrate the differences from other embodiments. The above description is only a preferred embodiment of the present application and is not used to limit the protection scope of the present application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application are all included in the protection scope of the present application.

Claims

1. A quartz flexible surface temperature compensation method, characterized in that: include: S1: Construct a temperature compensation model that considers the influence of temperature, temperature change rate and temperature spatial gradient in the sensitive axis direction; S2: using a super Latin square generational elimination particle swarm algorithm to identify the coefficients of the temperature compensation model, and based on the coefficients, obtaining a quartz flexible surface temperature compensation model that takes into account the temperature spatial gradient in the sensitive axis direction; S3: According to the real-time temperature measurement data of the quartz flexure meter in the sensitive axis direction, the compensation result of the quartz flexure meter temperature is determined by using the quartz flexure meter temperature compensation model.

2. The quartz flexible surface temperature compensation method according to claim 1, characterized in that: In S1, the temperature compensation model is: in, The error is caused by the temperature change on the A side of the quartz flexure. It is the temperature measurement of the internal side A of the quartz flex gauge; is the fifth-order coefficient of the temperature on the A side; is the fourth-order coefficient of the temperature on the A side; is the cubic coefficient of the temperature on side A; is the quadratic coefficient of the temperature on side A; is the first-order coefficient of the temperature on the A side; is the linear coefficient of the temperature change rate on side A; is the quadratic coefficient of the temperature change rate on side A, is a constant coefficient; It is the error caused by the quartz flexibility plus the temperature change on the B side of the table; It is the temperature measurement of the B side inside the quartz flex gauge; is the fifth-order coefficient of the temperature on the B side; is the fourth-order coefficient of the temperature on the B side; is the cubic coefficient of the temperature on the B side; is the quadratic coefficient of the temperature on the B side; is the first-order coefficient of the temperature on the B side; is the linear coefficient of the temperature change rate on the B side; is the quadratic coefficient of the temperature change rate on the B side, is a constant coefficient; It is a function used to describe the temperature change on the A side of the quartz flexure gauge; It is a function used to describe the temperature change on the B side of the quartz flexure adder; is the temperature measurement time; , Indicates the temperature change rate of the temperature on the A side and the temperature on the B side, , Indicates the temperature change acceleration rate of the A side temperature and the B side temperature, is the total error.

3. The quartz flexible surface temperature compensation method according to claim 1, characterized in that: The S2 specifically includes: S201: Initialize the population; S202: Construct fitness function: in, To minimize the cumulative sum of squares of temperature errors, is the sum of squares of temperature errors, E(t) is the temperature error compensation term, is the true temperature, is the temperature obtained by the test, is the first problem to be solved in the compensation model. coefficients, and For the The lower and upper limits of the search range for the parameters, For time; S203: Calculate the fitness values ​​of all random particles in the population using the fitness function, and take the individual fitness extreme value of each particle in the population as the individual optimal solution, and determine the current global optimal solution from all the individual optimal solutions; S204: Compare the current global optimal solution with the historical global optimal solution to obtain a comparison result; if the current global optimal solution is better than the historical global optimal solution, update the particle position and speed, calculate the fitness of the updated particle, determine the individual optimal solution and the current global optimal solution, wherein the updated particle position and speed are: in, is the inertia weight, , is the acceleration constant, , is a random number in [0,1], For particles In the Generation, The individual extreme value of dimension, For particles In the Generation, The global extremum of dimension, For particles In the Generation, The location of the dimension, For particles In the Generation, The speed of the dimension, For particles In the Generation, The location of the dimension, For particles In the Generation, The speed of the dimension; S205: When the minimum fitness value of all random particles in the population satisfies: , reduce the population size until the termination condition is met, where is the minimum fitness value of the fitness function of the first generation of particles, For the y The fitness minimum value of the fitness function of the generation particle, is a constant; S206: Using the particle parameters corresponding to the global optimal solution as coefficients of the temperature compensation model; S207: According to the coefficients, a quartz flexibility plus surface temperature compensation model taking into account the temperature spatial gradient in the sensitive axis direction is obtained.

4. The quartz flexible surface temperature compensation method according to claim 3, characterized in that: Initializing the population in S201 specifically includes: S2011: Initialize population parameters, including: initialization iteration number, population size, particle dimension, variables per dimension The search interval, fitness function, variable value range and particle swarm; S2012: The initial sample population is sampled using the Super Latin Square method to obtain uniformly distributed random initial particles.

5. The quartz flexible surface temperature compensation method according to claim 4, characterized in that: In S2012, the initial sample population is sampled using the super Latin square method to obtain uniformly distributed random initial particles, which specifically includes: S20121: Determine the initial particle swarm size H; S20122: Each dimension variable The search interval Divide into H equally spaced intervals, forming H hypercubes; S20123: Based on the initial particle swarm size H and the equally spaced intervals, generate an H×16 matrix A, wherein each column of the matrix A is a random full permutation of the sequence {1, 2, …, H}; S20124: When only one small hypercube is selected in each row of the matrix A, a particle is generated in each small hypercube, and the coordinates of the jth particle in the 16-dimensional space are ( ), among which The The particles are: Among them, 1≤ ≤16, 1≤ ≤ , is the initial subgroup size, , For each dimension variable The endpoints of the interval, For each subinterval length, For the The random displacement of particles in the subinterval, for The matrix of For the The Particles; S20125: Each particle constitutes an initial particle.

6. The quartz flexible surface temperature compensation method according to claim 3, characterized in that: The termination conditions are: the current global optimal solution of the particle fitness in the population reaches a preset value; the number of algorithm cycles reaches a preset value.

Images