High-thrust-density and high-precision linear driving multi-objective optimization method

Abstract

The invention provides a high-thrust-density high-precision linear driving multi-objective optimization method, and relates to the technical field of motor multi-objective optimization, and the method specifically comprises the steps: carrying out the motor performance test through different design parameter combinations, obtaining the mean value of attribute parameters and control parameters, building a performance prediction model, taking different design parameters as the input, and carrying out the calculation of the performance prediction model. And carrying out training and random combination on the equalized attributes and control parameters as labels to generate an initial population, obtaining parameter values through a performance prediction model, constructing a function relationship among the parameters for comprehensively evaluating the performance of the motor, taking the maximization of a comprehensive evaluation coefficient as a target, carrying out optimization by utilizing a genetic algorithm, and selecting an optimal design parameter combination. According to the method, the influence of different design parameter combinations on the thrust density and the positioning precision can be systematically analyzed and evaluated, and a scientific basis and an evaluation standard are provided for optimization of the design parameters, so that the accuracy of optimization of the design parameters of the linear synchronous motor is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of multi-objective optimization of motors, and in particular to a multi-objective optimization method for high-thrust density and high-precision linear drive. Background Art

[0002] Linear drive systems are widely used in automation equipment, CNC machine tools, semiconductor manufacturing, medical equipment, and aerospace systems, which usually require high-precision positioning and fast dynamic response. High thrust density means that greater thrust can be output in a smaller volume or weight, which helps to design more compact systems while improving the speed and responsiveness of the equipment.

[0003] In the prior art, a method and device for multi-objective optimization of a permanent magnet linear synchronous motor provided by publication number CN116911151A comprises the following steps: using multiple structural parameters of the permanent magnet linear synchronous motor as factors and multiple performance indicators as The optimization target is simplified and preprocessed, and then the values ​​of the factors are adjusted based on iterative orthogonal tests and surrogate models to make each target as optimal as possible. This method solves the technical problems that the training sample size and model accuracy of the surrogate model used in the prior art multi-objective optimization method of permanent magnet linear synchronous motor are mutually restricted, and the experimental design method has limited accuracy and is prone to fall into local optimality. Only a small number of samples need to be calculated to obtain a high-precision optimal design from a larger factor value range.

[0004] However, there are still some deficiencies. As can be seen from the above statements, the prior art lacks analysis of the functional relationship between the attribute parameters and control parameters of the linear synchronous motor, which makes it difficult to understand the interaction between the parameters during the optimization process, and further makes it difficult to quantify the impact of different parameter combinations, limiting the accuracy of the design parameter optimization of the linear synchronous motor.

[0005] The above information disclosed in this Background section is only for enhancement of understanding of the background of the present disclosure and therefore it may contain information that does not constitute the prior art that is already known to one of ordinary skill in the art. Summary of the invention

[0006] The object of the present invention is to provide a high thrust density and high precision linear drive multi-objective optimization method to solve the problems raised in the above background technology.

[0007] To achieve the above object, the present invention provides the following technical solutions: A multi-objective optimization method for high thrust density and high precision linear drive, the specific steps include: S1. Keep the test conditions consistent, perform performance tests of the linear synchronous motor under different combinations of design parameters of the linear synchronous motor, and obtain the attribute parameters and control parameters of the linear synchronous motor at multiple consecutive moments during the performance test, wherein the attribute parameters of the linear synchronous motor include electromagnetic thrust and motor mass, and the control parameters include drive shaft position and drive shaft speed. Based on the drive shaft position and target position at the same moment, obtain the position deviation value at the moment, and based on the drive shaft speed at the next moment and the previous moment in the consecutive moments, obtain the speed change rate, average the electromagnetic thrust, position deviation value and speed change rate, and obtain the electromagnetic thrust mean, position deviation value mean and speed change rate mean. The design parameters include air gap flux density, coil turns, current density and air gap length. S2. Construct a performance prediction model for a linear synchronous motor, use different design parameter combinations as input, and use the mean value of electromagnetic thrust, the mean value of position deviation, and the mean value of speed change rate as label training models to train the performance prediction model for the linear synchronous motor; S3. Establish constraints on design parameters, randomly combine design parameters, construct individuals of the initial population of design parameters, input individuals of the initial population of design parameters into the performance prediction model of the linear synchronous motor, and obtain the mean electromagnetic thrust, the mean position deviation value, and the mean speed change rate; S4. construct a functional relationship between the mean electromagnetic thrust, the motor mass and the thrust density coefficient, construct a functional relationship between the mean position deviation value, the mean velocity change rate and the positioning accuracy coefficient, and construct a functional relationship between the thrust density coefficient, the positioning accuracy coefficient and the comprehensive evaluation coefficient. The comprehensive evaluation coefficient is used to comprehensively evaluate the performance of the linear synchronous motor; S5. Taking the maximization of the comprehensive evaluation coefficient as the objective function, a genetic algorithm is used to select, crossover and mutate individuals in the initial population of design parameters until a predetermined number of iterations is reached, and the individual corresponding to the maximum value of the comprehensive evaluation coefficient is taken as the optimal design parameter combination.

[0008] Furthermore, the design parameters are randomly combined to construct individuals of the initial population of design parameters. The specific process is as follows: The initial population is labeled , and the initial population , is the first Individuals, is the index of the individual in the initial population, and , is the number of individuals in the initial population, ,in, , , , Respectively The air gap flux density, number of coil turns, current density and air gap length of each individual.

[0009] Furthermore, the performance prediction model of the linear synchronous motor is composed of a deep learning network based on a multilayer perceptron, wherein the deep neural network of the multilayer perceptron includes an input layer, a first hidden layer, a second hidden layer, a third hidden layer and an output layer, wherein the first hidden layer, the second hidden layer and the third hidden layer each have at least two neurons, and each uses ReLU as an activation function; The process of training the performance prediction model of the linear synchronous motor is as follows: Different design parameter combinations are used as input, and the mean position deviation value, the mean velocity change rate, and the mean electromagnetic thrust are used as output labels for training. The mean square error is used as the loss function. When the mean square error is When the performance prediction model of the linear synchronous motor is within the range, the training of the performance prediction model of the linear synchronous motor is completed.

[0010] Furthermore, a functional relationship between the mean electromagnetic thrust, motor mass and thrust density coefficient is constructed in the following form:

[0011] in, For the The thrust density coefficient of each individual is used to comprehensively evaluate the influence of attribute parameters on thrust density. For the The average electromagnetic thrust of each individual, For the The mass of each motor, is the minimum threshold of thrust density.

[0012] Furthermore, a functional relationship between the mean value of the position deviation, the mean value of the velocity change rate and the positioning accuracy coefficient is constructed in the following form:

[0013] in, For the The positioning accuracy coefficient of each individual is used to comprehensively evaluate the influence of control parameters on positioning accuracy. For the The mean position deviation of each individual is For the The average rate of change of speed of each individual, is the weight coefficient of the mean position deviation value, is the weight coefficient of the mean velocity change rate, On the basis of .

[0014] Furthermore, the functional relationship between the thrust density coefficient, positioning accuracy coefficient and comprehensive evaluation coefficient is constructed in the following form:

[0015] in, For the The comprehensive evaluation coefficient of each individual is used to combine the thrust density coefficient and the positioning accuracy coefficient to comprehensively evaluate the performance of the linear synchronous motor. and are the weights in the calculation of thrust density coefficient and positioning accuracy coefficient, respectively, and and The specific value of is determined by the hierarchical analysis method.

[0016] Furthermore, the process of step S5 is as follows: Iterative optimization is performed on the individuals of the initial population of design parameters. During the iterative optimization process, the constraints of the design parameters are set, that is, the maximum and minimum values ​​of the air gap flux density, the number of coil turns, the current density and the air gap length are set respectively. Within the constraints of the air gap flux density, the number of coil turns, the current density and the air gap length, the design parameters of the linear synchronous motor are iteratively optimized. Specifically, the comprehensive evaluation coefficient Sort from large to small and select the comprehensive evaluation coefficient The individuals in the front row are taken as the parents. Through crossover and mutation operations, the genes of the parent individuals are exchanged, combined and mutated to generate new individuals. The performance prediction model of the linear synchronous motor is used to obtain the attribute parameters and control parameters of the newly generated individuals, and the comprehensive evaluation coefficient is calculated. The new individuals and the parents are taken as the new population, and the selection, crossover and mutation operations are repeated until the predetermined number of iterations is reached. The individual corresponding to the maximum value of the comprehensive evaluation coefficient is taken as the optimal combination of design parameters of the linear synchronous motor, and the individual corresponding to the maximum value of the comprehensive evaluation coefficient is calibrated as , then the optimal design parameter combination of the linear synchronous motor is the air gap flux density , Coil turns , current density and air gap length .

[0017] Compared with the prior art, the present invention has the following beneficial effects: The present invention constructs a performance prediction model of a linear synchronous motor on the basis of acquiring real-time data, takes different design parameter combinations as input, and takes averaged attribute parameters and control parameters as labels for training. Through the model, it is possible to systematically analyze and evaluate the influence of different design parameter combinations on thrust density and positioning accuracy. On this basis, a functional relationship between the mean electromagnetic thrust, motor mass and thrust density coefficient is established, and a functional relationship between the mean position deviation value, the mean speed change rate and the positioning accuracy coefficient is constructed. These functional relationships make the relationship between the thrust density coefficient, the positioning accuracy coefficient and the comprehensive evaluation coefficient clearer, provide a scientific basis and evaluation criteria for the optimization of design parameters, and thus improve the accuracy of the optimization of design parameters of the linear synchronous motor. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 It is a schematic diagram of the overall method flow of the present invention. DETAILED DESCRIPTION

[0019] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with specific embodiments.

[0020] It should be noted that, unless otherwise defined, the technical terms or scientific terms used in the present invention should be understood by people with ordinary skills in the field to which the present invention belongs. The words "first", "second" and similar words used in the present invention do not indicate any order, quantity or importance, but are only used to distinguish different components. "Include" or "comprise" and similar words mean that the elements or objects appearing before the word include the elements or objects listed after the word and their equivalents, without excluding other elements or objects. "Connect" or "connected" and similar words are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. "Up", "down", "left", "right" and the like are only used to indicate relative positional relationships. When the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0021] Embodiment 1:

[0022] See also Figure 1 , the present invention provides a technical solution: A multi-objective optimization method for high thrust density and high precision linear drive, the specific steps include: S1. Keep the test conditions consistent, perform performance tests of the linear synchronous motor under different combinations of design parameters of the linear synchronous motor, and obtain the attribute parameters and control parameters of the linear synchronous motor at multiple consecutive moments during the performance test, wherein the attribute parameters of the linear synchronous motor include electromagnetic thrust and motor mass, and the control parameters include drive shaft position and drive shaft speed. Based on the drive shaft position and target position at the same moment, obtain the position deviation value at the moment, and based on the drive shaft speed at the next moment and the previous moment in the consecutive moments, obtain the speed change rate, average the electromagnetic thrust, position deviation value and speed change rate, and obtain the electromagnetic thrust mean, position deviation value mean and speed change rate mean. The design parameters include air gap flux density, coil turns, current density and air gap length. The drive shaft position is the actual position reached by the drive shaft, and the target position of the drive shaft is the set position that the drive shaft should reach at that moment; S2. Construct a performance prediction model for a linear synchronous motor, use different design parameter combinations as input, and use the mean value of electromagnetic thrust, the mean value of position deviation, and the mean value of speed change rate as label training models to train the performance prediction model for the linear synchronous motor; S3. Establish constraints on design parameters, randomly combine design parameters, construct individuals of the initial population of design parameters, input individuals of the initial population of design parameters into the performance prediction model of the linear synchronous motor, and obtain the mean electromagnetic thrust, the mean position deviation value, and the mean speed change rate; S4. construct a functional relationship between the mean electromagnetic thrust, the motor mass and the thrust density coefficient, construct a functional relationship between the mean position deviation value, the mean velocity change rate and the positioning accuracy coefficient, and construct a functional relationship between the thrust density coefficient, the positioning accuracy coefficient and the comprehensive evaluation coefficient. The comprehensive evaluation coefficient is used to comprehensively evaluate the performance of the linear synchronous motor; S5. Taking the maximization of the comprehensive evaluation coefficient as the objective function, a genetic algorithm is used to select, crossover and mutate individuals in the initial population of design parameters until a predetermined number of iterations is reached, and the individual corresponding to the maximum value of the comprehensive evaluation coefficient is taken as the optimal design parameter combination.

[0023] Based on the above embodiment, the test conditions include: Environmental conditions: The test should be carried out under the same ambient temperature, humidity and air pressure to avoid the influence of the external environment on the motor performance.

[0024] Power supply conditions: The voltage, frequency and phase of the power supply should be kept consistent to ensure that the motor operates under the same power supply conditions to avoid performance differences caused by power supply fluctuations.

[0025] Test load: When performing performance testing, the applied load should remain consistent to ensure that the output performance of each test is performed under the same operating conditions.

[0026] Connection method: The wiring method of the linear synchronous motor and the configuration of the sensor should be consistent to ensure the accuracy of signal transmission.

[0027] Test equipment and measuring tools: Use test equipment and measuring tools of the same model and specifications to avoid measurement errors caused by equipment differences.

[0028] On the basis of the above embodiment, the electromagnetic thrust, motor mass, drive shaft position and drive shaft speed are collected by the following equipment and method: Use force sensors or load cells, which can be mounted directly on the motor or mechanical structure to measure the thrust generated by the electromagnetic.

[0029] During the motor design and manufacturing process, a precise electronic balance is used to weigh the motor and its components to ensure that the measurement is performed in a static environment in order to obtain accurate motor quality data.

[0030] The encoder installed on the motor shaft monitors the rotation angle in real time to calculate the drive shaft position.

[0031] When the motor is running, the drive shaft speed is measured by the speed sensor.

[0032] On the basis of the above embodiment, after collecting the electromagnetic thrust, motor mass, drive shaft position and drive shaft speed, the above parameters are respectively subjected to maximum-minimum normalization processing, and then the normalized data is used for subsequent analysis and processing, so that in the subsequent analysis and processing process, various data can be analyzed and processed under the same dimension, avoiding the problem of some data being neglected due to different dimensions.

[0033] Among them, the weighing sensor is used to collect electromagnetic thrust, the electronic balance is used to collect motor mass, the encoder is used to collect drive shaft position, and the speed sensor is used to collect drive shaft speed. The weighing sensor, electronic balance, encoder, and speed sensor can all use models in existing equipment and are not limited here.

[0034] Weighing sensors, electronic balances, encoders, and speed sensors are all in multiple groups (such as 5 groups). The same data detected multiple times are averaged, and the final average is used as the corresponding data in the attribute parameters and control parameters to avoid accidental errors in single point sampling.

[0035] On the basis of the above embodiment, the design parameters are randomly combined to construct individuals of the initial population of design parameters. The specific process is as follows: The initial population is labeled , and the initial population , is the first Individuals, is the index of the individual in the initial population, and , is the number of individuals in the initial population, ,in, , , , Respectively The air gap flux density, number of coil turns, current density and air gap length of each individual.

[0036] On the basis of the above embodiment, the performance prediction model of the linear synchronous motor is composed of a deep learning network based on a multilayer perceptron, and the deep neural network of the multilayer perceptron includes an input layer, a first hidden layer, a second hidden layer, a third hidden layer and an output layer, and the first hidden layer, the second hidden layer and the third hidden layer each have at least two neurons, and each uses ReLU as an activation function; In this embodiment, the input features of the deep learning network of the multilayer perceptron include four features: air gap magnetic flux density, number of coil turns, current density and air gap length.

[0037] The structure of the deep learning network of multi-layer perceptron is: Input layer: receives input of 4 features; The first hidden layer has 64 neurons and uses ReLU as the activation function. The second hidden layer has 32 neurons and also uses the ReLU activation function. The third hidden layer has 16 neurons and uses the ReLU activation function. Output layer: has 3 neurons, the mean electromagnetic thrust, the mean position deviation, and the mean velocity change rate.

[0038] The process of training the performance prediction model of the linear synchronous motor is as follows: Different design parameter combinations are used as input, and the mean electromagnetic thrust, the mean position deviation, and the mean velocity change rate are used as output labels for training. The mean square error is used as the loss function. When the mean square error is When the performance prediction model of the linear synchronous motor is within the range, the training of the performance prediction model of the linear synchronous motor is completed.

[0039] Based on the above embodiment, a functional relationship between electromagnetic thrust, motor mass and thrust density coefficient is constructed in the following form:

[0040] in, For the The thrust density coefficient of each individual is used to comprehensively evaluate the influence of attribute parameters on thrust density. The larger the thrust density coefficient, the better the motor acceleration performance. In the formula, For the The average electromagnetic thrust of each individual, For the The mass of each motor, is the minimum threshold of thrust density, i.e., the thrust density under ideal or reference conditions; when , , indicating that the thrust density of the individual is better than the minimum threshold, showing better performance. On the contrary, when ,Right now , indicating that the thrust density of this individual is below the minimum threshold, showing substandard performance.

[0041] The reason for setting the above function form to express the functional relationship between the thrust density coefficient and the mean electromagnetic thrust and the motor mass is as follows: First, The mean electromagnetic thrust of each individual The larger it is, the more load it can theoretically push or the faster it can accelerate, so it is set on the numerator of the fractional row number, which is in line with the positive correlation between the mean electromagnetic thrust and the thrust density coefficient.

[0042] Second, The mass of the individual motors It is larger, which reduces energy consumption and maneuverability, so it is set in the denominator of the fractional row number, which is in line with the negative correlation between motor mass and thrust density coefficient.

[0043] third, is the minimum thrust density threshold, which can be used to and Compare to judge the performance of the motor.

[0044] Based on the above embodiment, the correlation between the mean value of the position deviation value, the mean value of the speed change rate and the positioning accuracy is as follows: The larger the mean position deviation value, the greater the gap between the system positioning and the target position, and the lower the positioning accuracy. Therefore, the positioning accuracy is negatively correlated with the mean position deviation value.

[0045] The higher the mean value of the speed change rate, the more frequent or drastic the speed change, which may cause the system to respond lagging or unstable, thereby reducing the positioning accuracy. Therefore, there is also a negative correlation between positioning accuracy and the mean value of the speed change rate.

[0046] According to the correlation between the mean value of position deviation, the mean value of velocity change rate and positioning accuracy, a functional relationship between the mean value of position deviation, the mean value of velocity change rate and positioning accuracy coefficient is constructed in the following form:

[0047] in, For the The positioning accuracy coefficient of each individual is used to comprehensively evaluate the influence of control parameters on positioning accuracy. The larger the positioning accuracy coefficient, the higher the positioning accuracy. For the The mean position deviation of each individual is For the The average rate of change of speed of each individual, is the weight coefficient of the mean position deviation value, is the weight coefficient of the mean value of the velocity change rate; The reason for setting the above function form to express the functional relationship between the positioning accuracy coefficient and the mean value of the position deviation value and the mean value of the speed change rate is as follows: First, the mean of position deviation The smaller the positioning accuracy coefficient The higher the value, the smaller the position deviation value can improve the positioning accuracy and stability of the system. As the denominator, it can express the mean of the position deviation value And positioning accuracy coefficient There is a negative correlation between them, which is consistent with the normal logic of system positioning performance.

[0048] Second, the mean rate of change of speed The smaller the positioning accuracy coefficient The higher the speed, the smaller the speed change rate is, which is beneficial for the system to maintain a stable positioning state. As the denominator, it can express the mean rate of change of speed And positioning accuracy coefficient There is a negative correlation between them, which is consistent with the normal logic of system positioning performance.

[0049] third, are the important weights of the mean position deviation value and the mean velocity change rate. Their settings reflect the relative importance of these two factors in the positioning accuracy evaluation. By adjusting these coefficients, different system requirements and characteristics can be better adapted.

[0050] Since in many precision positioning systems, position accuracy is usually more important than responding to dynamic changes, a small positioning error may have a significant impact on the accuracy of the entire system and fail to achieve the expected target position. Therefore, giving a higher weight coefficient to the mean of the position deviation value can ensure that the system has a higher priority in position accuracy. In contrast, the mean velocity change rate usually affects the dynamic response characteristics, but has relatively little direct impact on short-term positioning accuracy. On the basis of .

[0051] As an implementation method, The value range is an open interval of 0.4-0.5. The value range is an open interval of 0.5-0.6. The specific value is set by the technicians according to the actual situation and is not limited here.

[0052] Based on the above embodiment, a functional relationship between the thrust density coefficient, the positioning accuracy coefficient and the comprehensive evaluation coefficient is constructed in the following form:

[0053] in, For the The comprehensive evaluation coefficient of each individual is used to combine the thrust density coefficient and the positioning accuracy coefficient to comprehensively evaluate the performance of the linear synchronous motor. The performance of the linear synchronous motor includes the motor acceleration performance and positioning accuracy. The larger the comprehensive evaluation coefficient, the better the performance of the linear synchronous motor. It should be noted that, as can be seen from the above description, the thrust density coefficient The larger the value, the better the motor acceleration performance and the positioning accuracy coefficient. The larger the value, the higher the positioning accuracy, so the comprehensive evaluation coefficient and thrust density coefficient , Positioning accuracy coefficient All are positively correlated, so the calculation formula of the comprehensive evaluation coefficient in the form of weighted summation is set; In the formula, and are the weights in the calculation of thrust density coefficient and positioning accuracy coefficient, respectively, and and The specific value of is determined by the hierarchical analysis method, and the specific logic is as follows: The thrust density coefficient and positioning accuracy coefficient are marked, and the relative importance between the two indicators is determined by the nine-scale method to construct a judgment matrix, in which the index of the thrust density coefficient is marked as 1 and the index of the positioning accuracy coefficient is marked as 2. The constructed judgment matrix for:

[0054] in, , denotes the index of the coefficient, and , , indicating that the index is The importance of the index of relative to the index of v to the comprehensive evaluation coefficient, The specific value is determined by relevant experts using a 1-9 scoring method. Indicates that the index is The index of is more important to the comprehensive evaluation coefficient than the index of v. Indicates that the index is The coefficient of is extremely unimportant to the comprehensive evaluation coefficient compared to the coefficient with index v; Each element value in the judgment matrix is ​​divided by the sum of its columns to obtain a normalized judgment matrix. The mean of the element values ​​in each row of the normalized judgment matrix is ​​calculated, and the mean of the element values ​​in the first row is used as the proportional coefficient of the thrust density coefficient, and the mean of the element values ​​in the second row is used as the proportional coefficient of the positioning accuracy coefficient. With the constraint that the sum of the scaled values ​​is equal to 1, the two proportional coefficients are scaled in equal proportion, and the values ​​obtained after scaling are used as the weights of the corresponding coefficients.

[0055] Based on the above embodiment, the process of step S5 is as follows: Iterative optimization is performed on the individuals of the initial population of design parameters. During the iterative optimization process, the constraints of the design parameters are set, that is, the maximum and minimum values ​​of the air gap flux density, the number of coil turns, the current density and the air gap length are set respectively. Within the constraints of the air gap flux density, the number of coil turns, the current density and the air gap length, the design parameters of the linear synchronous motor are iteratively optimized. Specifically, the comprehensive evaluation coefficient Sort from large to small and select the comprehensive evaluation coefficient The individuals in the front row are regarded as the parents, and the front row refers to the individuals in the comprehensive evaluation coefficient. For the first 50% of individuals, the genes of the parent individuals are exchanged, combined and mutated through crossover and mutation operations to generate new individuals. The performance prediction model of the linear synchronous motor is used to obtain the attribute parameters and control parameters of the newly generated individuals, and their comprehensive evaluation coefficients are calculated. The new individuals and the parent generation are used as the new population, and the selection, crossover and mutation operations are repeated until the predetermined number of iterations is reached. The individual corresponding to the maximum value of the comprehensive evaluation coefficient is used as the optimal combination of design parameters of the linear synchronous motor, and the individual corresponding to the maximum value of the comprehensive evaluation coefficient is calibrated as , then the optimal design parameter combination of the linear synchronous motor is the air gap flux density , Coil turns , current density and air gap length .

[0056] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters in the formula are set by technicians in this field according to actual conditions.

[0057] The above embodiments may be implemented in whole or in part by software, hardware, firmware or any other combination thereof. When implemented using software, the above embodiments may be implemented in whole or in part in the form of a computer program product. Those skilled in the art may appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein may be implemented by computer software, electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed by hardware or software depends on the specific application and design constraints of the technical solution.

[0058] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, and may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0059] The above description is only a specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application.

Claims

1. A multi-objective optimization method for high thrust density and high precision linear drive, characterized in that: The specific steps include: S1. Keep the test conditions consistent, perform performance tests of the linear synchronous motor under different combinations of design parameters of the linear synchronous motor, and obtain the attribute parameters and control parameters of the linear synchronous motor at multiple consecutive moments during the performance test, wherein the attribute parameters of the linear synchronous motor include electromagnetic thrust and motor mass, and the control parameters include drive shaft position and drive shaft speed. Based on the drive shaft position and target position at the same moment, obtain the position deviation value at the moment, and based on the drive shaft speed at the next moment and the previous moment in the consecutive moments, obtain the speed change rate, average the electromagnetic thrust, position deviation value and speed change rate, and obtain the electromagnetic thrust mean, position deviation value mean and speed change rate mean. The design parameters include air gap flux density, coil turns, current density and air gap length. S2. Construct a performance prediction model for a linear synchronous motor, use different design parameter combinations as input, and use the mean value of electromagnetic thrust, the mean value of position deviation, and the mean value of speed change rate as label training models to train the performance prediction model for the linear synchronous motor; S3. Establish constraints on design parameters, randomly combine design parameters, construct individuals of the initial population of design parameters, input individuals of the initial population of design parameters into the performance prediction model of the linear synchronous motor, and obtain the mean electromagnetic thrust, the mean position deviation value, and the mean speed change rate; S4. construct a functional relationship between the mean electromagnetic thrust, the motor mass and the thrust density coefficient, construct a functional relationship between the mean position deviation value, the mean velocity change rate and the positioning accuracy coefficient, and construct a functional relationship between the thrust density coefficient, the positioning accuracy coefficient and the comprehensive evaluation coefficient. The comprehensive evaluation coefficient is used to comprehensively evaluate the performance of the linear synchronous motor; S5. Taking the maximization of the comprehensive evaluation coefficient as the objective function, a genetic algorithm is used to select, crossover and mutate individuals in the initial population of design parameters until a predetermined number of iterations is reached, and the individual corresponding to the maximum value of the comprehensive evaluation coefficient is taken as the optimal design parameter combination.

2. The high thrust density and high precision linear drive multi-objective optimization method according to claim 1 is characterized in that: The design parameters are randomly combined to construct individuals of the initial population of design parameters. The specific process is as follows: The initial population is labeled , and the initial population , is the first Individuals, is the index of the individual in the initial population, and , is the number of individuals in the initial population, ,in, , , , Respectively The air gap flux density, number of coil turns, current density and air gap length of each individual.

3. The high thrust density and high precision linear drive multi-objective optimization method according to claim 1 is characterized in that: The performance prediction model of the linear synchronous motor is composed of a deep learning network based on a multilayer perceptron, wherein the deep neural network of the multilayer perceptron includes an input layer, a first hidden layer, a second hidden layer, a third hidden layer and an output layer, wherein the first hidden layer, the second hidden layer and the third hidden layer each have at least two neurons, and each uses ReLU as an activation function; The process of training the performance prediction model of the linear synchronous motor is as follows: Different design parameter combinations are used as input, and the mean position deviation value, the mean velocity change rate, and the mean electromagnetic thrust are used as output labels for training. The mean square error is used as the loss function. When the mean square error is When the performance prediction model of the linear synchronous motor is within the range, the training of the performance prediction model of the linear synchronous motor is completed.

4. The high thrust density and high precision linear drive multi-objective optimization method according to claim 2 is characterized in that: The functional relationship between the mean electromagnetic thrust, motor mass and thrust density coefficient is constructed in the following form: ， in, For the The thrust density coefficient of each individual is used to comprehensively evaluate the influence of attribute parameters on thrust density. For the The average electromagnetic thrust of each individual, For the The mass of each motor, is the minimum threshold of thrust density.

5. The high thrust density and high precision linear drive multi-objective optimization method according to claim 4 is characterized in that: Construct a functional relationship between the mean position deviation value, the mean velocity change rate and the positioning accuracy coefficient in the following form: ， in, For the The positioning accuracy coefficient of each individual is used to comprehensively evaluate the influence of control parameters on positioning accuracy. For the The mean position deviation of each individual is For the The average rate of change of speed of each individual, is the weight coefficient of the mean position deviation value, is the weight coefficient of the mean velocity change rate, On the basis of .

6. The high thrust density and high precision linear drive multi-objective optimization method according to claim 5 is characterized in that: The functional relationship between thrust density coefficient, positioning accuracy coefficient and comprehensive evaluation coefficient is constructed in the following form: ， in, For the The comprehensive evaluation coefficient of each individual is used to combine the thrust density coefficient and the positioning accuracy coefficient to comprehensively evaluate the performance of the linear synchronous motor. and are the weights in the calculation of thrust density coefficient and positioning accuracy coefficient, respectively, and and The specific value of is determined by the hierarchical analysis method.

7. The high thrust density and high precision linear drive multi-objective optimization method according to claim 6 is characterized in that: The process of step S5 is as follows: Iterative optimization is performed on the individuals of the initial population of design parameters. During the iterative optimization process, the constraints of the design parameters are set, that is, the maximum and minimum values ​​of the air gap flux density, the number of coil turns, the current density and the air gap length are set respectively. Within the constraints of the air gap flux density, the number of coil turns, the current density and the air gap length, the design parameters of the linear synchronous motor are iteratively optimized. Specifically, the comprehensive evaluation coefficient Sort from large to small and select the comprehensive evaluation coefficient The individuals in the front row are taken as the parents. Through crossover and mutation operations, the genes of the parent individuals are exchanged, combined and mutated to generate new individuals. The performance prediction model of the linear synchronous motor is used to obtain the attribute parameters and control parameters of the newly generated individuals, and the comprehensive evaluation coefficient is calculated. The new individuals and the parents are taken as the new population, and the selection, crossover and mutation operations are repeated until the predetermined number of iterations is reached. The individual corresponding to the maximum value of the comprehensive evaluation coefficient is taken as the optimal combination of design parameters of the linear synchronous motor, and the individual corresponding to the maximum value of the comprehensive evaluation coefficient is calibrated as , then the optimal design parameter combination of the linear synchronous motor is the air gap flux density , Coil turns , current density and air gap length .

Images