Control parameter optimization method and computing device

Abstract

The invention provides a control parameter optimization method and a computing device, the method is based on initial values of a plurality of control parameters corresponding to a target process parameter in a semiconductor process device, the parameter value of each control parameter is iteratively updated by using a Tiancattle swarm algorithm, the plurality of control parameters are used for adjusting the target process parameter, and in each iteration, executing a specified number of times of position updating steps, respectively updating the positions of a plurality of long-horned beetles on different dimensions in the specified number of times of position updating steps, and updating the optimal long-horned beetle position corresponding to the preset long-horned beetle group based on the fitness of each long-horned beetle. The fitness of the longicorn is determined based on the adjustment error of the target process parameter corresponding to the latest updating result of the position of the longicorn on each dimension in the current position updating step, so that the accuracy of the optimization result of each control parameter can be effectively ensured, and the adjustment efficiency of the target process parameter is improved; and the overshoot of the target process parameters in the adjusting process is reduced.

Description

Technical Field

[0001] This application relates to the field of semiconductor manufacturing technology, and in particular to a method and computing device for optimizing control parameters. Background Technology

[0002] With the rapid development of semiconductor technology, the requirements for the precision of process parameter control during semiconductor manufacturing processes are becoming increasingly stringent. Controllers typically adjust the values ​​of process parameters based on multiple control parameters, such as PID (Proportion-Integral-Derivative) control methods, to stabilize the process parameter values ​​within a preset range, thereby improving manufacturing quality. Among these adjustments, the values ​​of each control parameter are crucial factors influencing the effectiveness of the adjustment.

[0003] Currently, the parameter values ​​of each control parameter are usually optimized based on engineering experience, which cannot guarantee the accuracy of the optimization results of each control parameter, and thus cannot guarantee the adjustment efficiency of the process parameters. In addition, the overshoot during the process parameter adjustment process is relatively large. Summary of the Invention

[0004] To address the aforementioned technical problems, this application provides a control parameter optimization method and computing device to solve the problems in the prior art that cannot guarantee the adjustment efficiency of process parameters and that the overshoot during the process parameter adjustment process is large.

[0005] To achieve the above technical objectives, the embodiments of this application provide the following technical solutions:

[0006] Firstly, the embodiments of this specification provide a method for optimizing control parameters, including:

[0007] The initial values ​​of multiple control parameters corresponding to the target process parameters in the semiconductor process equipment are obtained; the multiple control parameters are used to adjust the target process parameters;

[0008] Based on the initial values ​​of the multiple control parameters, the parameter values ​​of each control parameter are iteratively updated using the beetle swarm algorithm to obtain the target values ​​of each control parameter;

[0009] In the current iteration, a specified number of position update steps are performed, the specified number being the same as the number of control parameters. In the specified number of position update steps, the positions of multiple longhorn beetles in a preset longhorn beetle group are updated in different dimensions, and the preferred longhorn beetle position corresponding to the preset longhorn beetle group is updated based on the fitness of each longhorn beetle. The fitness of the longhorn beetle is determined based on the adjustment error of the target process parameter corresponding to the latest update result of the longhorn beetle's position in each dimension in the current position update step. The positions of the longhorn beetles in different dimensions represent different parameter values ​​of the control parameters.

[0010] In one implementation, updating the positions of multiple longhorn beetles in a preset longhorn beetle herd in different dimensions includes:

[0011] Based on the update result of the longhorn beetle's movement speed in the target dimension in the previous iteration, determine the update result of the longhorn beetle's position in the target dimension in the current iteration;

[0012] The update result of the beetle's movement speed in the target dimension in the current iteration is determined based on the update result of the beetle's position in the target dimension in the current iteration and the latest update result of the preferred beetle position in the current iteration.

[0013] In one implementation, the update result of the longhorn beetle's movement speed in the target dimension in the current iteration is determined by the following method:

[0014] Based on the position update result of the longhorn beetle in the target dimension in the current iteration and the latest update result of the preferred longhorn beetle position in the current iteration, the initial determination result of the moving speed of the longhorn beetle in the target dimension in the current iteration is obtained;

[0015] The initial determination result is corrected based on the shrinkage factor to obtain the updated result of the longhorn beetle's movement speed in the target dimension in the current iteration, wherein the shrinkage factor is greater than 0 and less than 1.

[0016] In one implementation, obtaining the initial determination result of the longhorn beetle's movement speed in the target dimension based on the position update result of the longhorn beetle in the current iteration and the latest update result of the preferred longhorn beetle position in the current iteration includes:

[0017] Based on the position update result of the longhorn beetle in the target dimension in the current iteration and the latest update result of the preferred longhorn beetle position in the current iteration, the velocity compensation value of the longhorn beetle in the target dimension is determined;

[0018] Based on the speed compensation value, the update result of the beetle's movement speed in the target dimension in the previous iteration is compensated to obtain the initial determination result of the beetle's movement speed in the target dimension in the current iteration.

[0019] In one embodiment, the preferred longhorn beetle location includes a first preferred longhorn beetle location corresponding to the preset longhorn beetle group and a second preferred longhorn beetle location corresponding to each of the longhorn beetles in the preset longhorn beetle group;

[0020] The step of determining the velocity compensation value of the longhorn beetle in the target dimension based on the position update result of the longhorn beetle in the current iteration and the latest update result of the preferred longhorn beetle position in the current iteration includes:

[0021] Based on the latest update result of the first preferred longhorn beetle position in the current iteration, the position update result of the longhorn beetle in the target dimension in the current iteration, and the first correction coefficient, the first velocity compensation value of the longhorn beetle in the target dimension is determined; and based on the latest update result of the second preferred longhorn beetle position corresponding to the longhorn beetle in the current iteration, the position update result of the longhorn beetle in the target dimension in the current iteration, and the second correction coefficient, the second velocity compensation value of the longhorn beetle in the target dimension is determined.

[0022] The speed compensation value of the longhorn beetle in the target dimension includes the first speed compensation value and the second speed compensation value.

[0023] In one embodiment, the shrinkage factor is determined based on the following method:

[0024] The shrinkage factor is determined based on the first correction coefficient and the second correction coefficient; wherein the shrinkage factor is negatively correlated with the sum of coefficients, and the sum of coefficients includes the sum of the first correction coefficient and the second correction coefficient.

[0025] In one implementation, determining the position update result of the longhorn beetle in the target dimension in the current iteration based on the update result of the longhorn beetle's movement speed in the target dimension in the previous iteration includes:

[0026] Based on the update results of the beetle's movement speed in the target dimension in the previous iteration and the update results of the beetle's position movement increment in the target dimension in the previous iteration, the position update result of the beetle in the target dimension in the current iteration is determined.

[0027] In this iteration, the update result of the position movement increment of the beetle in the target dimension is obtained based on the update result of the movement speed of the beetle in the target dimension in the current iteration and the target movement direction of the beetle in the target dimension in the current iteration. The target movement direction is determined based on the fitness of the beetle when it moves in different directions.

[0028] In one implementation, updating the preferred beetle position corresponding to the preset beetle herd based on the fitness of each beetle includes:

[0029] The preferred beetle position is updated based on the fitness of each beetle and the fitness corresponding to the target update result of the preferred beetle position.

[0030] Secondly, embodiments of this specification provide a computing device including at least one processor and at least one memory, wherein the memory stores a computer program that, when executed by the processor, implements the control parameter optimization method as described in any of the preceding claims.

[0031] Thirdly, embodiments of this specification provide a computer-readable storage medium storing a computer program that, when executed by a processor, implements the control parameter optimization method as described in any of the preceding claims.

[0032] Fourthly, embodiments of this specification provide a computer program product or a computer program, the computer program product including a computer program stored in a computer-readable storage medium; the processor of the computer device reads the computer program from the computer-readable storage medium, and when the processor executes the computer program, it implements the control parameter optimization method as described in any of the preceding claims.

[0033] As can be seen from the above technical solutions, the embodiments of this application provide a control parameter optimization method and computing device. The control parameter optimization method obtains initial values ​​of multiple control parameters corresponding to target process parameters in semiconductor process equipment. Based on these initial values, it uses a beetle swarm algorithm to iteratively update the parameter values ​​of each control parameter to obtain the target value of each control parameter. These multiple control parameters are used to adjust the target process parameters. The iterative optimization of the parameter values ​​using the beetle swarm algorithm effectively ensures the accuracy of the optimization results for each control parameter. In each iteration, a specified number of position update steps are executed, and the specified number is related to the control parameters. The number of beetles is the same; in the position update step of a specified number of times, the positions of multiple beetles in the preset beetle group in different dimensions are updated respectively, and the preferred beetle position corresponding to the preset beetle group is updated based on the fitness of each beetle. Thus, at the end of the iteration, the target value of each control parameter can be determined based on the final update result of the preferred beetle position. Since the fitness of the beetle is determined based on the adjustment error of the target process parameter corresponding to the latest update result of the beetle position in each dimension in the current position update step, the accuracy of the optimization result of each control parameter can be further guaranteed, thereby improving the adjustment efficiency of the target process parameter and reducing the overshoot of the target process parameter in the adjustment process. Attached Figure Description

[0034] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of this application. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0035] Figure 1 A flowchart illustrating a control parameter optimization method provided for one embodiment of this specification;

[0036] Figure 2 A flowchart illustrating another control parameter optimization method provided as one embodiment of this specification;

[0037] Figure 3 This is a schematic diagram of a control system for a target process parameter provided as one embodiment of this specification. Detailed Implementation

[0038] Unless otherwise defined, the technical or scientific terms used in the embodiments of this specification shall have the ordinary meaning understood by one of ordinary skill in the art to which this specification pertains. The terms "first," "second," and similar terms used in the embodiments of this specification do not indicate any order, quantity, or importance, but are merely used to avoid confusion of constituent elements.

[0039] Unless the context otherwise requires, throughout this specification, "a plurality of" means "at least two," and "including" is interpreted as open-ended or encompassing, that is, "including, but not limited to." In the description of this specification, terms such as "one embodiment," "some embodiments," "exemplary embodiment," "example," "specific example," or "some examples" are intended to indicate that a particular feature, structure, material, or characteristic associated with that embodiment or example is included in at least one embodiment or example of this specification. The illustrative representations of the above terms do not necessarily refer to the same embodiment or example.

[0040] The technical solutions in the embodiments of this specification will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this specification, and not all embodiments. Based on the embodiments in this specification, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this specification.

[0041] Overview

[0042] As described in the background section, with the rapid development of semiconductor technology, the requirements for the precision of process parameter control during semiconductor manufacturing processes are becoming increasingly stringent. Controllers typically adjust the values ​​of process parameters based on multiple control parameters, such as PID (Proportion Integration Differentiation) control methods, to stabilize the process parameter values ​​within a preset range, thereby improving manufacturing quality. Among these adjustments, the values ​​of each control parameter are crucial factors affecting the effectiveness of the adjustment.

[0043] Currently, the correspondence between process parameters and the operating parameters of the regulating device is typically learned through experimentation, and the process parameters are adjusted based on this correspondence. For example, when the process parameter is chamber pressure, the chamber pressure can be regulated by controlling the position of the moving parts of the pressure regulating valve through a pressure controller. The position of the moving parts of the pressure regulating valve represents the valve opening. Specifically, the correspondence between chamber pressure and the position of the moving parts can be determined experimentally. Based on the target value of the chamber pressure and the correspondence between chamber pressure and the moving parts position, the target position of the moving parts can be determined, and the position of the moving parts can be adjusted to that target position.

[0044] In determining the correspondence between chamber pressure and the position of the moving part, the moving part of the pressure regulating valve can be controlled to move from a first limit position to a second limit position under test conditions to obtain the two extreme values ​​of the actual range of chamber pressure under the current working environment. During the movement of the moving part of the pressure regulating valve, the position of the moving part and the chamber pressure corresponding to each first recording point are acquired. Then, the actual range of chamber pressure is compared with the target range of chamber pressure, and the two extreme values ​​of the smaller range are obtained. Based on the moving part position and chamber pressure corresponding to each first recording point, the moving part positions corresponding to the two extreme values ​​of the smaller range are obtained and used as the first and second extreme value positions, respectively. The moving part is controlled to move from the first extreme value position to the second extreme value position. During the movement, the position of the moving part and the chamber pressure corresponding to each second recording point are acquired, and the moving part position and chamber pressure corresponding to the first and second recording points are used as the correspondence between chamber pressure and the position of the moving part of the pressure regulating valve.

[0045] However, due to the influence of various factors such as the environment and operating conditions of semiconductor process equipment, the control accuracy of process parameters cannot be guaranteed based on this correspondence. When the control accuracy of process parameters cannot be guaranteed by the learned correspondence, the controller still needs to adjust the process parameters through methods such as PID control.

[0046] Currently, the parameter values ​​of each control parameter are usually optimized based on engineering experience, which cannot guarantee the accuracy of the optimization results of each control parameter, and thus cannot guarantee the adjustment efficiency of the process parameters. In addition, the overshoot during the process parameter adjustment process is relatively large.

[0047] To address the shortcomings of traditional methods in ensuring the accuracy of optimization results for various control parameters, thereby compromising the efficiency of process parameter adjustment and leading to significant overshoot during adjustment, this application's technical solution obtains the initial values ​​of multiple control parameters corresponding to the target process parameter in the semiconductor process equipment. Based on these initial values, a beetle swarm algorithm is used to iteratively update the parameter values ​​of each control parameter, yielding the target value. These multiple control parameters are then used to adjust the target process parameter. The iterative optimization of each control parameter value using the beetle swarm algorithm effectively ensures the accuracy of the optimization results. Specifically, in each iteration, a specified number of position updates are performed. The steps are as follows: the number of times is the same as the number of control parameters; in the position update step of the specified number of times, the positions of multiple longhorn beetles in the preset longhorn beetle group are updated in different dimensions respectively, and the preferred longhorn beetle position corresponding to the preset longhorn beetle group is updated based on the fitness of each longhorn beetle. Thus, at the end of the iteration, the target value of each control parameter can be determined based on the final update result of the preferred longhorn beetle position. Since the fitness of the longhorn beetle is determined based on the adjustment error of the target process parameter corresponding to the latest update result of the longhorn beetle position in each dimension in the current position update step, the accuracy of the optimization result of each control parameter can be further guaranteed, thereby improving the adjustment efficiency of the target process parameter and reducing the overshoot of the target process parameter during the adjustment process.

[0048] Based on the above inventive concept, the control parameter optimization method provided in the embodiments of this specification will be described exemplarily below.

[0049] Exemplary methods

[0050] This specification provides a method for optimizing control parameters, such as... Figure 1 As shown, it includes:

[0051] S101. Obtain the initial values ​​of multiple control parameters corresponding to the target process parameters in the semiconductor process equipment; the multiple control parameters are used to adjust the target process parameters.

[0052] Specifically, semiconductor process equipment can include etching equipment, deposition equipment, wafer cleaning equipment, etc. The target process parameter is the process parameter that needs to be adjusted during the semiconductor process equipment's processing, such as temperature, pressure, and flow rate. Multiple control parameters corresponding to the target process parameter are used to adjust the target process parameter. For example, the controller corresponding to the target process parameter can adjust the operating parameters of the regulating device based on multiple control parameters to adjust the parameter value of the target process parameter, stabilizing the actual value of the target process parameter within its target value range.

[0053] The initial values ​​of multiple control parameters can be set according to actual needs. For example, for any control parameter, the initial value can be randomly determined based on its preset value range, or it can be determined based on experience. In implementation, one set of initial values ​​can be determined, or multiple sets of initial values ​​can be set. One set of initial values ​​can include the initial value of each control parameter.

[0054] S102. Based on the initial values ​​of the multiple control parameters, the parameter values ​​of each control parameter are iteratively updated using the beetle swarm algorithm to obtain the target values ​​of each control parameter.

[0055] In the current iteration, a specified number of position update steps are performed, the specified number being the same as the number of control parameters. In the specified number of position update steps, the positions of multiple longhorn beetles in a preset longhorn beetle group are updated in different dimensions, and the preferred longhorn beetle position corresponding to the preset longhorn beetle group is updated based on the fitness of each longhorn beetle. The fitness of the longhorn beetle is determined based on the adjustment error of the target process parameter corresponding to the latest update result of the longhorn beetle's position in each dimension in the current position update step. The positions of the longhorn beetles in different dimensions represent different parameter values ​​of the control parameters.

[0056] Specifically, based on the initial values ​​of multiple control parameters, the beetle swarm algorithm can be used to iteratively update the parameter values ​​of each control parameter to obtain the target value of each control parameter. For example, based on the initial values ​​of multiple control parameters, the initial position of each beetle in a preset beetle swarm can be determined. For any given beetle, its position can represent the parameter value of each control parameter. That is, the beetle's position can include its position in multiple dimensions, with each dimension corresponding one-to-one with a control parameter. The beetle's position in any dimension is the parameter value of the control parameter corresponding to that dimension.

[0057] It is understandable that the initial positions of different longhorn beetles in a pre-defined longhorn beetle herd can be the same or different. For example, the initial position of each longhorn beetle can be determined based on the same set of initial values ​​or based on different initial values.

[0058] In the process of iteratively updating the values ​​of each control parameter using the beetle swarm algorithm, for any iteration, a specified number of position update steps can be executed. The specified number can be the same as the number of control parameters. That is, in this iteration, each time the position update step is executed, the position of each beetle in one dimension is updated. Thus, by executing the specified number of updates, the position update of each beetle in each dimension can be completed.

[0059] For any beetle, during the current position update step, its position in the corresponding dimension can be updated, and its fitness can be determined based on its current position. The beetle's current position can include the latest update results of its position in each dimension during the current position update step; that is, for any dimension, the latest update result of the beetle's position in that dimension can include the position update result in the most recent execution of the position update step corresponding to that dimension.

[0060] In determining the fitness of a longhorn beetle based on its current position, the target process parameters can be adjusted based on the parameter values ​​of each control parameter corresponding to the beetle's current position, and the fitness of the beetle can be determined based on the adjustment error of the target process parameters. Specifically, the parameter values ​​of each control parameter corresponding to the beetle's current position can be transmitted to the controller corresponding to the target process parameters, so that the controller adjusts the target process parameters based on the parameter values ​​of each control parameter corresponding to the beetle's current position, and the fitness of the beetle can be determined based on the adjustment error of the target process parameters. For example, the fitness can be calculated as shown in equation (1):

[0061]

[0062] In the formula, |e(t)| represents the adjustment error of the target process parameter at time t, e(t) is the difference between the detected value of the target process parameter and the target value of the target process parameter at time t, and T is the adjustment time of the target process parameter. Therefore, the smaller the fitness, the better the parameter values ​​of each control parameter.

[0063] It is understandable that in any position update step, the position updates of each beetle are carried out synchronously. Therefore, the fitness of each beetle in the preset beetle group in the current position update step can be obtained.

[0064] The preferred beetle position corresponding to the preset beetle swarm can include the best position among all positions reached by each beetle in the preset beetle swarm during its movement from the initial position to the current position; that is, the best position experienced by the preset beetle swarm. During the process of updating this preferred beetle position based on the fitness of each beetle, the fitness of each beetle can be compared with the target fitness, and the preferred beetle position can be updated based on the comparison result. The target fitness can include the fitness corresponding to the update result of the preferred beetle position during the previous position update step. It can be understood that if the current position update step is the first execution of the position update step, then the fitness corresponding to the update result of the preferred beetle position during the previous position update step can be the fitness corresponding to the best beetle position among the initial positions of each beetle.

[0065] During implementation, at the end of the iteration, the target values ​​of each control parameter can be determined based on the latest update result of the selected beetle position in the last iteration. For example, the latest update result of the selected beetle position in each dimension can be used as the target value of the corresponding control parameter, thereby enabling the effective updating of the parameter values ​​of each control parameter, greatly improving the accuracy of the optimization results of each control parameter, thereby improving the adjustment efficiency of the target process parameter, and reducing the overshoot of the target process parameter during the adjustment process.

[0066] To accelerate the search for better locations by longhorn beetles and improve the reliability of location search results, in one embodiment of this specification, updating the positions of multiple longhorn beetles in a preset longhorn beetle swarm in different dimensions includes:

[0067] Based on the update result of the longhorn beetle's movement speed in the target dimension in the previous iteration, determine the update result of the longhorn beetle's position in the target dimension in the current iteration;

[0068] The update result of the beetle's movement speed in the target dimension in the current iteration is determined based on the update result of the beetle's position in the target dimension in the current iteration and the latest update result of the preferred beetle position in the current iteration.

[0069] Specifically, the target dimension is the dimension in which the beetle's position is updated in the current position update step. In the current iteration, for any beetle, during the update of its position in the target dimension, the position update result in the current iteration can be determined based on the update result of the beetle's velocity in the target dimension in the previous iteration. For any dimension, the beetle's movement speed in that dimension can represent the rate of change of the parameter value of the corresponding control parameter during the iteration process.

[0070] During implementation, the beetle's movement speed in the corresponding dimension can also be updated during each position update step. Specifically, in the current iteration, the beetle's movement speed in the target dimension can be updated based on the position update result of the beetle in the target dimension in the current iteration and the latest update result of the optimal beetle position in the current iteration, so as to obtain the updated result of the beetle's movement speed in the target dimension in the current iteration.

[0071] The latest update result of the preferred beetle position in the current iteration can include the update result of the preferred beetle position in the current position update step.

[0072] Specifically, the initial determination of the beetle's movement speed in the target dimension can be determined based on the position update result of the beetle in the current iteration and the latest update result of the preferred beetle position in the current iteration. The update result of the beetle's movement speed in the target dimension in the current iteration can then be determined based on this initial determination result. For example, the initial determination result can be directly used as the update result of the beetle's movement speed in the target dimension in the current iteration. Alternatively, the initial determination result can be corrected, and the result of the correction can be used as the update result of the beetle's movement speed in the target dimension in the current iteration. The correction process can include magnification, compression, etc. Thus, in each iteration, the beetle's speed can be adjusted according to the difference between the beetle's current position and the preferred beetle position, enabling the beetle to move towards a better position. This accelerates the beetle's search for a better position and improves the reliability of the position search results, thereby enabling the rapid and accurate determination of the target values ​​of each control parameter.

[0073] It is understandable that when the current iteration is the first iteration, the update result of the longhorn beetle's movement speed in the target dimension in the previous iteration can be the initial speed of the longhorn beetle in the target dimension. That is, before performing the iterative update, the initial speed of each longhorn beetle in the preset longhorn beetle group can be set.

[0074] In a feasible implementation, the update result of the longhorn beetle's movement speed in the target dimension in the current iteration is determined by the following method:

[0075] Based on the position update result of the longhorn beetle in the target dimension in the current iteration and the latest update result of the preferred longhorn beetle position in the current iteration, the initial determination result of the moving speed of the longhorn beetle in the target dimension in the current iteration is obtained;

[0076] The initial determination result is corrected based on the shrinkage factor to obtain the updated result of the longhorn beetle's movement speed in the target dimension in the current iteration, wherein the shrinkage factor is greater than 0 and less than 1.

[0077] Specifically, for any beetle, during the update of its movement speed in the target dimension in the current iteration, an initial determination of its movement speed in the target dimension can be obtained based on the position update result of the beetle in the target dimension in the current iteration and the latest update result of the preferred beetle position in the current iteration. For example, a speed compensation value for the beetle's movement speed in the target dimension can be determined based on the position update result of the beetle in the target dimension in the current iteration and the latest update result of the preferred beetle position in the current iteration, and then based on this speed compensation value, the speed compensation value for the beetle's movement speed in the target dimension in the previous iteration can be adjusted. The movement speed update result is adjusted to obtain the initial determination result of the beetle's movement speed in the target dimension in the current iteration. Alternatively, the initial determination result of the beetle's movement speed in the target dimension in the current iteration can be determined based on the position update result of the beetle in the target dimension in the current iteration, the latest update result of the preferred beetle position in the current iteration, and a first preset correspondence. The first preset correspondence can include the correspondence between the position update result of the beetle in the target dimension in the current iteration, the latest update result of the preferred beetle position in the current iteration, and the initial determination result of the beetle's movement speed in the target dimension in the current iteration. Therefore, based on the position update result of the beetle in the target dimension in the current iteration and the latest update result of the preferred beetle position in the current iteration, the initial determination result of the beetle's movement speed in the target dimension in the current iteration can be determined quickly and accurately.

[0078] In practice, the initial determination results can be corrected based on the shrinkage factor, and the updated result of the beetle's movement speed in the target dimension in the current iteration can be determined based on the result of the correction. The shrinkage factor can be greater than 0 and less than 1. That is, the correction can effectively reduce the beetle's movement speed in the target dimension, thereby reducing the risk of the beetle missing the search for the optimal position due to excessive movement speed, and further improving the accuracy of the determination results of the target values ​​of each control parameter.

[0079] In one feasible implementation, obtaining the initial determination result of the longhorn beetle's movement speed in the target dimension based on the position update result of the longhorn beetle in the current iteration and the latest update result of the preferred longhorn beetle position in the current iteration includes:

[0080] Based on the position update result of the longhorn beetle in the target dimension in the current iteration and the latest update result of the preferred longhorn beetle position in the current iteration, the velocity compensation value of the longhorn beetle in the target dimension is determined;

[0081] Based on the speed compensation value, the update result of the beetle's movement speed in the target dimension in the previous iteration is compensated to obtain the initial determination result of the beetle's movement speed in the target dimension in the current iteration.

[0082] Specifically, for any beetle, in the process of determining the initial determination result of the beetle's movement speed in the target dimension in the current iteration, the speed compensation value of the beetle in the target dimension can be determined based on the position update result of the beetle in the target dimension in the current iteration and the latest update result of the preferred beetle position in the current iteration. For example, the difference between the latest update result of the preferred beetle position in the target dimension and the position update result of the beetle in the target dimension in the current iteration can be obtained, and the speed compensation value of the beetle in the target dimension can be determined based on the difference. Alternatively, the speed compensation value of the beetle in the target dimension can be determined based on the latest update result of the preferred beetle position in the target dimension, the position update result of the beetle in the target dimension in the current iteration, and a second preset correspondence. The second preset correspondence can include the correspondence between the latest update result of the preferred beetle position in the target dimension, the position update result of the beetle in the target dimension, and the speed compensation value of the beetle in the target dimension.

[0083] In practice, the speed compensation value can be used to compensate for the update result of the beetle's movement speed in the target dimension in the previous iteration, so as to obtain the initial determination result of the beetle's movement speed in the target dimension in the current iteration, thereby enabling the initial determination result of the beetle's movement speed in the target dimension to be determined quickly and accurately.

[0084] In one feasible implementation, the preferred longhorn beetle location includes a first preferred longhorn beetle location corresponding to the preset longhorn beetle group and a second preferred longhorn beetle location corresponding to each of the longhorn beetles in the preset longhorn beetle group;

[0085] The step of determining the velocity compensation value of the longhorn beetle in the target dimension based on the position update result of the longhorn beetle in the current iteration and the latest update result of the preferred longhorn beetle position in the current iteration includes:

[0086] Based on the latest update result of the first preferred longhorn beetle position in the current iteration, the position update result of the longhorn beetle in the target dimension in the current iteration, and the first correction coefficient, the first velocity compensation value of the longhorn beetle in the target dimension is determined; and based on the latest update result of the second preferred longhorn beetle position corresponding to the longhorn beetle in the current iteration, the position update result of the longhorn beetle in the target dimension in the current iteration, and the second correction coefficient, the second velocity compensation value of the longhorn beetle in the target dimension is determined.

[0087] The speed compensation value of the longhorn beetle in the target dimension includes the first speed compensation value and the second speed compensation value.

[0088] Specifically, the preferred locations for a pre-defined longhorn beetle herd can include a first preferred location for the herd and a second preferred location for each individual longhorn beetle within the herd. The first preferred location can include the optimal location among all positions reached by each longhorn beetle in the herd during its movement from its initial position to its current position; that is, the optimal location experienced by the herd as a whole. For any single longhorn beetle, the second preferred location can include the optimal location among all positions reached by that longhorn beetle during its movement from its initial position to its current position; that is, the optimal location experienced by that longhorn beetle.

[0089] In practice, for any beetle, in determining the speed compensation value of the beetle in the target dimension, the first speed compensation value of the beetle in the target dimension can be determined based on the latest update result of the first preferred beetle position in the current iteration, the update result of the beetle's position in the target dimension in the current iteration, and the first correction coefficient. The first correction coefficient can represent the importance of the first speed compensation value, that is, the first correction coefficient can represent the degree of influence of the first preferred beetle position on the beetle's moving speed in the target dimension.

[0090] For example, the difference between the latest update result of the first preferred longhorn beetle position in the current iteration and the position update result of the longhorn beetle in the target dimension can be obtained as the first difference. This first difference is then corrected based on a first correction coefficient to obtain the first velocity compensation value of the longhorn beetle in the target dimension. The first correction coefficient can be a preset value, or it can be determined based on the first difference and a preset correspondence between the first difference and the first correction coefficient, thereby enabling the rapid and accurate determination of the first velocity compensation value of the longhorn beetle in the target dimension.

[0091] Meanwhile, based on the latest update result of the second preferred beetle position corresponding to the beetle in the current iteration, the update result of the beetle's position in the target dimension in the current iteration, and the second correction coefficient, the second velocity compensation value of the beetle in the target dimension can be determined. The second correction coefficient can represent the importance of the second velocity compensation value, that is, the second correction coefficient can represent the degree of influence of the second preferred beetle position corresponding to the beetle on the beetle's movement speed in the target dimension.

[0092] For example, the difference between the latest update result of the second preferred beetle position in the target dimension and the position update result of the beetle in the target dimension in the current iteration can be obtained as the second difference. This second difference is then corrected based on a second correction coefficient to obtain the second velocity compensation value of the beetle in the target dimension. The second correction coefficient can be a preset value, or it can be determined based on the second difference and a preset correspondence between the second difference and the second correction coefficient, thus enabling the rapid and accurate determination of the second velocity compensation value of the beetle in the target dimension.

[0093] The speed compensation value of the longhorn beetle in the target dimension can include a first speed compensation value and a second speed compensation value in the target dimension. For example, it can include the sum of the first and second speed compensation values ​​in the target dimension, thereby effectively ensuring the reliability of the speed compensation value of the longhorn beetle in the target dimension. In turn, the speed compensation value of the longhorn beetle in the target dimension can effectively ensure the accuracy of the speed update result of the longhorn beetle in the target dimension, speed up the search for a better position by the longhorn beetle and improve the reliability of the position search result.

[0094] As a preferred implementation, the update result of the movement speed of the i-th longhorn beetle in dimension j in the k-th iteration It can be shown in equation (2):

[0095]

[0096] In the formula, This represents the update result of the movement speed of the i-th longhorn beetle in dimension j during the (k-1)-th iteration; This represents the latest update result of the first preferred longhorn beetle position in the k-th iteration, in dimension j. Let j be the latest update result of the second preferred longhorn beetle position corresponding to the ith longhorn beetle in the kth iteration; This represents the position update result of the i-th longhorn beetle in dimension j during the k-th iteration; r1 and r2 are the compensation coefficients for the first and second velocity compensation values, respectively, and are random numbers that can be generated by the random function rand(); c1 is the first correction coefficient, and c2 is the second correction coefficient; It is a contraction factor.

[0097] In one feasible implementation, the shrinkage factor is determined based on the following method:

[0098] The shrinkage factor is determined based on the first correction coefficient and the second correction coefficient; wherein the shrinkage factor is negatively correlated with the sum of coefficients, and the sum of coefficients includes the sum of the first correction coefficient and the second correction coefficient.

[0099] Specifically, the sum of the first correction coefficient and the second correction coefficient can be obtained, and the sum of the first correction coefficient and the second correction coefficient can be used as the coefficient sum to determine the shrinkage factor. The shrinkage factor can be negatively correlated with the coefficient sum. In practice, the coefficient sum can be input into a preset function expression to calculate the shrinkage factor through the preset function expression. Alternatively, the shrinkage factor can be determined based on the coefficient sum and the preset correspondence between the coefficient sum and the shrinkage factor, thereby enabling the size of the shrinkage factor to be determined quickly and effectively.

[0100] The first and second correction coefficients not only affect the longhorn beetle's trajectory but also the information transmission between different beetles in the preset beetle swarm. The first correction coefficient affects the beetle's convergence speed during the position search process. If the first correction coefficient is set too large, the beetle is prone to getting trapped in local optima, i.e., converging prematurely to a local optimum and missing the optimal position search, thus compromising the accuracy of the target values ​​for each control parameter. The second correction coefficient affects the beetle's optimization range. If the second correction coefficient is set too large, the beetle will move excessively within a local range, failing to guarantee a convergence speed. This embodiment determines the contraction factor based on the first and second correction coefficients, keeping them within a reasonable range. This effectively balances the convergence speed and reliability of the optimization results during the beetle's optimization process, avoiding the negative impact of improperly set first and second correction coefficients on the convergence speed and reliability of the optimization results. This enhances the beetle's ability to approach the optimal position, greatly improving the optimization efficiency of multiple control parameters and the effectiveness of the optimization results.

[0101] As a preferred implementation, the preset function expression can be as shown in equation (3):

[0102]

[0103] In the formula, Let c be the contraction factor, c be the sum of coefficients, c = c1 + c2, c1 is the first correction coefficient, c2 is the second correction coefficient, and c > 4.

[0104] In one feasible implementation, determining the position update result of the longhorn beetle in the target dimension in the current iteration based on the update result of the longhorn beetle's movement speed in the target dimension in the previous iteration includes:

[0105] Based on the update results of the beetle's movement speed in the target dimension in the previous iteration and the update results of the beetle's position movement increment in the target dimension in the previous iteration, the position update result of the beetle in the target dimension in the current iteration is determined.

[0106] In this iteration, the update result of the position movement increment of the beetle in the target dimension is obtained based on the update result of the movement speed of the beetle in the target dimension in the current iteration and the target movement direction of the beetle in the target dimension in the current iteration. The target movement direction is determined based on the fitness of the beetle when it moves in different directions.

[0107] Specifically, for any beetle, in determining its position update in the target dimension in the current iteration, the position update in the target dimension can be determined based on the update results of its movement speed and position movement increment in the target dimension from the previous iteration. The position movement increment in the target dimension represents the required movement distance of the beetle starting from its position update result in the previous iteration.

[0108] In practice, the update results of the beetle's movement speed in the target dimension and the update results of its position movement increment in the target dimension in the previous iteration can be weighted and summed to obtain the beetle's optimal movement distance in the target dimension. Based on the beetle's position update results in the target dimension in the previous iteration and the optimal movement distance, the position update results of the beetle in the target dimension in the current iteration can be determined. Thus, based on the update results of the beetle's movement speed and position movement increment in the target dimension in the previous iteration, the beetle's position can be updated quickly and accurately.

[0109] In a preferred embodiment, the position update result of the i-th longhorn beetle in dimension j in the k-th iteration It can be shown in equation (4):

[0110]

[0111] In the formula, Update the position of the i-th longhorn beetle in dimension j during the (k-1)-th iteration; This represents the update result of the movement speed of the i-th longhorn beetle in dimension j during the (k-1)-th iteration; λ is the update result of the positional movement increment of the i-th longhorn beetle in dimension j during the (k-1)-th iteration; λ is... The weighting coefficients, where λ is a constant.

[0112] During implementation, in each position update step, the beetle's position movement increment in the corresponding dimension can also be updated. Specifically, the update result of the beetle's position movement increment in the target dimension in the current iteration can be determined based on the updated result of the beetle's movement speed in the target dimension and the target movement direction in the target dimension in the current iteration. The target movement direction in the target dimension in the current iteration can be determined based on the beetle's left and right whisker fitness in the target dimension. The left whisker fitness in the target dimension can be the fitness of the beetle's left whisker position, and the right whisker fitness in the target dimension can be the fitness of the beetle's right whisker position. The beetle's left whisker position is the position obtained by shifting left in the target dimension from its current position, while maintaining its position in all other dimensions. Similarly, the beetle's right whisker position is the position obtained by shifting right in the target dimension from its current position, while maintaining its position in all other dimensions. The current position of the longhorn beetle is the position update result of the longhorn beetle in the target dimension in the current iteration.

[0113] Specifically, the target process parameters can be adjusted based on the parameter values ​​of each control parameter corresponding to the left antenna position of the beetle, and the left antenna fitness of the beetle in the target dimension can be determined based on the adjustment error of the target process parameters; in addition, the target process parameters can be adjusted based on the parameter values ​​of each control parameter corresponding to the right antenna position of the beetle, and the right antenna fitness of the beetle in the target dimension can be determined based on the adjustment error of the target process parameters.

[0114] For example, the update result of the positional movement increment of the i-th longhorn beetle in dimension j in the k-th iteration. It can be shown in equation (5):

[0115]

[0116] In the formula, δ k This is the step size factor in the k-th iteration, which can be a constant; This represents the update result of the movement speed of the i-th longhorn beetle in dimension j during the k-th iteration; Let be the position of the right whisker of the i-th longhorn beetle in the k-th iteration. Let be the right whisker fitness of the i-th longhorn beetle in dimension j during the k-th iteration; Let be the position of the left whisker of the i-th longhorn beetle in the k-th iteration. Let be the left whisker fitness of the i-th longhorn beetle in dimension j during the k-th iteration; sign() is the sign function, when When, sign() returns 1, When, sign() returns -1, When this happens, sign() returns 0.

[0117] in, Value in dimension j It can be as shown in equation (6), Value in dimension j It can be shown in equation (7):

[0118]

[0119] In the formula, d represents the position update result of the i-th longhorn beetle in dimension j during the k-th iteration; d is the distance between the left and right whiskers of the longhorn beetle, which is a constant.

[0120] It is understandable that when the current iteration is the first iteration, the position update result of the beetle in the target dimension in the previous iteration can be the initial position of the beetle in the target dimension, and the update result of the position movement increment of the beetle in the target dimension in the previous iteration can be determined based on the initial position and initial velocity of the beetle in the target dimension.

[0121] In one feasible implementation, updating the preferred beetle location corresponding to the preset beetle herd based on the fitness of each beetle includes:

[0122] The preferred beetle position is updated based on the fitness of each beetle and the fitness corresponding to the target update result of the preferred beetle position.

[0123] Specifically, when updating the preferred longhorn beetle position corresponding to the preset longhorn beetle group, the first preferred longhorn beetle position and the second preferred longhorn beetle position corresponding to each longhorn beetle in the preset longhorn beetle group can be updated simultaneously.

[0124] Specifically, when updating the first preferred longhorn beetle position, the update result of the first preferred longhorn beetle position in the current position update step can be determined based on the comparison between the fitness of each longhorn beetle and the fitness corresponding to the target update result of the first preferred longhorn beetle position. The target update result of the first preferred longhorn beetle position can be the latest update result of the first preferred longhorn beetle position before the current position update step is executed. For example, the longhorn beetle position corresponding to the best fitness among the fitness of each longhorn beetle and the fitness corresponding to the target update result of the first preferred longhorn beetle position can be used as the update result of the first preferred longhorn beetle position in the current position update step. Thus, by comparing fitness, the first preferred longhorn beetle position can be updated quickly and accurately.

[0125] For any longhorn beetle, when updating the second-preferred beetle position corresponding to that beetle, the beetle's fitness can be compared with the fitness corresponding to the target update result of the second-preferred beetle position. Based on the comparison result, the update result of the second-preferred beetle position corresponding to that beetle in the current position update step is determined. The target update result of the second-preferred beetle position can be the latest update result of the second-preferred beetle position corresponding to that beetle before the current position update step is executed. For example, the beetle position corresponding to the better fitness between the beetle's fitness and the fitness corresponding to the target update result of the second-preferred beetle position can be used as the update result of the second-preferred beetle position corresponding to that beetle in the current position update step. Thus, by comparing fitness, the second-preferred beetle positions corresponding to each beetle can be updated quickly and accurately.

[0126] It is understandable that when the current position update step is the first position update step executed in the first iteration, the target update result of the first preferred beetle position can be the initial position with the best fitness among the initial positions of each beetle, and the target update result of the second preferred beetle position can be the initial position of that beetle.

[0127] refer to Figure 2 The following uses PID control as an example to illustrate the implementation process of the control parameter optimization method in this embodiment. In the PID control process, the control parameters may include the proportional coefficient K. p Integral coefficient K i and differential coefficient K d The position, speed, and positional increment of the longhorn beetle all include the proportionality coefficient K. p Integral coefficient K i and differential coefficient K d The process corresponds to three dimensions, where dimension j = 1 represents the scaling factor K. p The corresponding dimension, where dimension j = 2 represents the integral coefficient K. i The corresponding dimension, where dimension j = 3 represents the integral coefficient K.d The corresponding dimension. During PID control, the output control signal u(t) can be expressed as shown in equation (8):

[0128]

[0129] The output control signal is used to adjust the target process parameters, such as the process chamber pressure.

[0130] S201. Input the number of longhorn beetles in the preset longhorn beetle swarm and the maximum number of iterations I. max ;

[0131] S202. Set the initial position and initial velocity of each longhorn beetle in the preset longhorn beetle group;

[0132] S203. Let the current iteration number k = 1;

[0133] S204. Let the position update step be executed n times = 1;

[0134] S205. Determine the position update result of each beetle in dimension j = n; wherein, for any beetle, the position update result of the beetle in dimension j in the k-th iteration can be determined based on the position update result, the movement speed update result, and the position movement increment update result of the beetle in dimension j in the (k-1)-th iteration.

[0135] S206. Based on the position update results of each longhorn beetle in dimension j=n, update the first preferred longhorn beetle position corresponding to the preset longhorn beetle group and the second preferred longhorn beetle position corresponding to each longhorn beetle.

[0136] S207. Determine the update result of the moving speed of each beetle in dimension j = n; wherein, for any beetle, the update result of the moving speed of the beetle in dimension j in the k-1th iteration can be determined based on the shrinkage factor, the update result of the moving speed of the beetle in dimension j in the k-1th iteration, the update result of the position of the beetle in dimension j in the k-th iteration, and the latest update result of the first preferred beetle position and the second preferred beetle position corresponding to the beetle in the k-th iteration.

[0137] S208. Determine the update result of the position movement increment of each beetle in dimension j = n; wherein, for any beetle, the left and right antenna positions of the beetle in dimension j can be determined based on the position update result of the beetle in dimension j in the k-th iteration and the update result of the beetle's movement speed in dimension j in the k-th iteration, and the left and right antenna fitness of the beetle in dimension j can be determined according to the left and right antenna positions, respectively. Based on the left and right antenna fitness of the beetle in dimension j, the target movement direction of the beetle in dimension j in the k-th iteration can be determined, and the update result of the position movement increment of the beetle in dimension j in the k-th iteration can be determined according to the update result of the beetle's movement speed in dimension j and the target movement direction in the k-th iteration.

[0138] S209. Determine if n is less than 3. If so, it indicates that the scaling factor K has not been adjusted in the k-th iteration. p Integral coefficient K i and differential coefficient K d If all three control parameters are updated, proceed to step S210; otherwise, it indicates that the scaling factor K has been updated in the k-th iteration. p Integral coefficient K i and differential coefficient K d Update all three control parameters and proceed to step S211;

[0139] S210. Let n = n + 1, and execute step S205;

[0140] S211. Determine if k is less than I. max If yes, proceed to step S212; otherwise, proceed to step S213.

[0141] S212. Let k = k + 1, and execute step S204;

[0142] S213. Determine the proportional coefficient K based on the latest update result of the first preferred longhorn beetle position. p Integral coefficient K i and differential coefficient K d The target value.

[0143] The aforementioned control parameter optimization method can be applied to the optimization of PID control parameters for process chamber pressure in semiconductor process equipment. The PID control parameters for process chamber pressure include the proportional coefficient K. p Integral coefficient K i and differential coefficient K dIn the process of optimizing the PID control parameters of the process chamber pressure, parameter initialization can be performed according to steps S201 to S204, and then iterative optimization of the PID control parameters of the process chamber pressure can be performed according to steps S205 to S213 based on the parameter initialization. In each iteration, the parameter values ​​of each control parameter can be updated sequentially through steps S205 to S208 (where the proportional coefficient K is updated). p Integral coefficient K i and differential coefficient K d (The update order is not specifically limited). For any control parameter, after updating its value, the latest update result of each control parameter in this iteration can be output to the pressure controller. The deviation between the target value and the actual detected value of the chamber pressure of the process chamber under the updated parameter result is obtained from the pressure controller. The fitness is determined based on this deviation, and the control parameter is iterated again based on the fitness. The pressure controller can output a control signal based on the received parameter values ​​of each control parameter to adjust the position of the moving parts in the pressure regulating valve, thereby regulating the chamber pressure of the process chamber. After the chamber pressure is regulated, the deviation between the target value and the actual detected value of the chamber pressure is obtained.

[0144] Exemplary device

[0145] This specification also provides a computing device, including at least one processor and at least one memory, wherein the memory stores a computer program that, when executed by the processor, implements the control parameter optimization method as described in any of the above embodiments.

[0146] The computing device can be connected to the control system for the target process parameters. (Reference) Figure 3 Taking the target process parameter as the chamber pressure of a semiconductor process equipment as an example, the input terminal of the computing device can be connected to the first input terminal of the pressure controller, and the output terminal of the computing device can be connected to the second input terminal of the pressure controller. The output terminal of the pressure controller is connected to the pressure regulating valve, which is connected to the process chamber of the semiconductor process equipment. During the optimization of the control parameters of the pressure controller, the computing device outputs the parameter values ​​of each control parameter found to the pressure controller. The pressure controller outputs control signals based on the parameter values ​​of each control parameter to adjust the position of the moving parts in the pressure regulating valve, thereby regulating the chamber pressure of the process chamber. The deviation between the target value and the actual detected value of the chamber pressure is input to the computing device. The computing device determines the fitness based on this deviation and searches for the parameter values ​​of each control parameter based on the fitness.

[0147] Exemplary computer program products and storage media

[0148] In addition to the methods and devices described above, the control parameter optimization methods provided in the embodiments of this specification can also be computer program products, which include computer program instructions that, when executed by a processor, cause the processor to perform the steps in the control parameter optimization methods according to various embodiments of this specification as described in the "Exemplary Methods" section above.

[0149] The computer program product described herein can be written in any combination of one or more programming languages ​​to perform the operations of the embodiments described herein. These programming languages ​​include object-oriented programming languages ​​such as Java and C++, as well as conventional procedural programming languages ​​such as C or similar languages. The program code can be executed entirely on the user's computing device, partially on the user's computing device, as a standalone software package, partially on the user's computing device and partially on a remote computing device, or entirely on a remote computing device or server.

[0150] Furthermore, embodiments of this specification also provide a computer-readable storage medium having a computer program stored thereon, the computer program being executed by a processor of the steps in the control parameter optimization methods according to various embodiments of this specification as described in the "Exemplary Methods" section above.

[0151] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the methods described above. Any references to memory, storage, databases, or other media used in the embodiments provided in this specification can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and RAMbus dynamic RAM (RDRAM), etc.

[0152] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0153] The embodiments described above are merely illustrative of several implementation methods outlined in this specification. While the descriptions are specific and detailed, they should not be construed as limiting the scope of the solutions provided in this specification. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this specification, and these all fall within the scope of protection of this specification. Therefore, the scope of protection for this patent should be determined by the appended claims.

Claims

1. A method of control parameter optimization, characterized by, The method comprises: obtaining initial values of a plurality of control parameters corresponding to a target process parameter in a semiconductor process equipment; the plurality of control parameters are used to adjust the target process parameter; based on the initial values of the plurality of control parameters, iteratively updating parameter values of each of the control parameters by using a firefly algorithm to obtain target values of the control parameters; wherein, in the current iteration, a specified number of position updating steps are performed, the specified number being the same as the number of the control parameters; in the specified number of position updating steps, the positions of a plurality of fireflies in a preset firefly group in different dimensions are updated respectively, and the optimal firefly position corresponding to the preset firefly group is updated based on the fitness of each firefly; the fitness of the firefly is determined based on the adjustment error of the target process parameter corresponding to the latest update result of the position of the firefly in each dimension in the current position updating step; and the positions of the firefly in different dimensions respectively represent the parameter values of different control parameters.

2. The method of claim 1, wherein, The method of updating the positions of the plurality of fireflies in the preset firefly group in different dimensions comprises: determining the position update result of the firefly in the target dimension in the current iteration based on the update result of the moving speed of the firefly in the target dimension in the last iteration; wherein, the update result of the moving speed of the firefly in the target dimension in the current iteration is determined based on the position update result of the firefly in the target dimension in the current iteration and the latest update result of the optimal firefly position in the current iteration.

3. The method of claim 2, wherein, The update result of the moving speed of the firefly in the target dimension in the current iteration is determined by the following method: based on the position update result of the firefly in the target dimension in the current iteration and the latest update result of the optimal firefly position in the current iteration, obtaining an initial determination result of the moving speed of the firefly in the target dimension in the current iteration; based on a contraction factor, correcting the initial determination result to obtain the update result of the moving speed of the firefly in the target dimension in the current iteration, the contraction factor being greater than 0 and less than 1.

4. The method of claim 3, wherein, The method of obtaining the initial determination result of the moving speed of the firefly in the target dimension in the current iteration based on the position update result of the firefly in the target dimension in the current iteration and the latest update result of the optimal firefly position in the current iteration comprises: based on the position update result of the firefly in the target dimension in the current iteration and the latest update result of the optimal firefly position in the current iteration, determining a speed compensation value of the firefly in the target dimension; based on the speed compensation value, compensating the update result of the moving speed of the firefly in the target dimension in the last iteration to obtain the initial determination result of the moving speed of the firefly in the target dimension in the current iteration.

5. The method of claim 4, wherein, The optimal firefly position comprises a first optimal firefly position corresponding to the preset firefly group and a second optimal firefly position corresponding to each firefly in the preset firefly group; The velocity compensation value of the beetle in the target dimension is determined based on the position update result of the beetle in the target dimension in the current iteration and the latest update result of the optimal beetle position in the current iteration, and the velocity compensation value of the beetle in the target dimension is determined based on the latest update result of the first optimal beetle position in the current iteration, the position update result of the beetle in the target dimension in the current iteration, and the first correction coefficient, and the second velocity compensation value of the beetle in the target dimension is determined based on the latest update result of the corresponding second optimal beetle position of the beetle in the current iteration, the position update result of the beetle in the target dimension in the current iteration, and the second correction coefficient. The velocity compensation value of the beetle in the target dimension includes the first velocity compensation value and the second velocity compensation value. The contraction factor is determined based on the following method:

6. The method of claim 5, wherein, The contraction factor is determined based on the first correction coefficient and the second correction coefficient, and the contraction factor is negatively correlated with the sum of the coefficients, and the sum of the coefficients includes the sum of the first correction coefficient and the second correction coefficient. The position update result of the beetle in the target dimension in the current iteration is determined based on the update result of the moving speed of the beetle in the target dimension in the last iteration, and the position update result of the beetle in the target dimension in the current iteration is determined based on the update result of the moving speed of the beetle in the target dimension in the last iteration and the update result of the position movement increment of the beetle in the target dimension in the last iteration.

7. The method of claim 2, wherein, The position update result of the beetle in the target dimension in the current iteration is determined based on the update result of the moving speed of the beetle in the target dimension in the current iteration and the target moving direction of the beetle in the target dimension in the current iteration, and the target moving direction is determined based on the fitness of the beetle moving in different directions. The optimal beetle position corresponding to the preset beetle group is updated based on the fitness of each beetle and the fitness corresponding to the target update result of the optimal beetle position. The optimal beetle position is updated based on the fitness of each beetle and the fitness corresponding to the target update result of the optimal beetle position.

8. The method according to any one of claims 1 to 7, characterized in that, The computer readable storage medium stores a computer program, and the computer program is executed by the processor to implement the control parameter optimization method in any one of claims 1 to 8. The computer readable storage medium stores a computer program, and the computer program is executed by the processor to implement the control parameter optimization method in any one of claims 1 to 8.

9. A computing device, comprising: ​ 10. A computer-readable storage medium, characterized in that, ​

Images