Intelligent optimization method for control parameters of servo motor system
The radial basis function network uses the position track curve feature vector to optimize the control parameters of servo motors, which solves the problems of low efficiency and unstable optimization results in the existing technology, and realizes efficient and stable control parameters optimization, which is suitable for complex nonlinearity system.
Patent Information
- Application Number
- CN202411187765.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-28
- Publication Date
- 2024-11-12
- Estimated Expiration
- 2044-08-28
AI Technical Summary
The existing servo motor control parameter optimization algorithm is inefficient, complex model, and unstable optimization results when facing complex nonlinear systems, making it difficult to quickly obtain optimal control parameters.
A radial basis function network is adopted, and a simple network structure is designed through the position trajectory curve feature vector as input, and a Gaussian radial basis function is used for control parameters optimization to achieve multi-parameter simultaneous optimization without understanding the control structure.
It realizes efficient and stable control parameter optimization, simplifies the network structure, reduces the calculation amount, is suitable for low-computing power platforms, and quickly obtains the optimal control parameters of the servo motor system.
Smart Images

Figure CN118938686A_ABST
Abstract
Description
Technical Field
[0001] The invention belongs to the technical field of servo motor control, and in particular relates to an intelligent optimization method for control parameters of a servo motor system. Background Art
[0002] In the field of high-precision machining, servo motor control systems are usually based on magnetic field oriented control and adopt a three-loop control scheme based on a PI controller as the basic framework. Its three-loop structure includes position control, speed control and current control. The ultimate goal is to track a given position trajectory with high precision.
[0003] The control parameters of the three-loop structure are usually not universal. In the face of complex actual operating conditions, the controller parameters need to be adjusted in a targeted manner to meet actual application requirements. The classic empirical formula tuning method relies on the low-order approximate model of the motor system and the selection of some empirical coefficients. Its tuning results are difficult to use in actual high-order nonlinear systems. In practical applications, manual experience and professional knowledge are required for trial and error tuning, but the various parameters of the motor system are coupled with each other, and the selection of the optimal parameters is difficult and time-consuming. In order to reduce professional labor costs and improve setting efficiency, the industry usually considers using some automatic optimization methods to assist parameter optimization, which can be mainly divided into: adaptive control, heuristic optimization and neural network optimization algorithm; adaptive control algorithm such as literature [Xue Bing, Zhou Liming, Niu Kang, et al. Research on fuzzy adaptive PID control system for wheat planter sowing depth [J / OL]. Transactions of the Chinese Society of Agricultural Machinery, 1-12, 2024] can automatically adjust parameters online according to system state changes, but in the face of complex nonlinear systems, it may face problems such as slow convergence, high algorithm complexity and poor robustness; heuristic optimization algorithm such as literature [Gaing ZL.A particle swarm optimization approach for optimum design of PID controllerin AVR system [J]. IEEE Transactions on Energy Conversion, 2004, 19(2): 384-391. DOI: 10.1109 / TEC.2003.821821] Using genetic algorithms, particle swarm optimization algorithms and other methods to optimize motor control parameters can effectively avoid local optimal solutions. However, when faced with multi-coupled parameter optimization targets, heuristic optimization has high hardware resource usage and low convergence efficiency. Neural network optimization algorithms such as the literature [Zhang Mingjun, Zhang Huaguang. Neural network PID controller based on genetic algorithm optimization [J]. Journal of Jilin University (Engineering Edition), 2005, (01): 91-96. DOI: 10.13229 / j.cnki.jdxbgxb.200501019] use the multi-layer perceptron to fit the controller, quickly respond to system changes, and have high optimization efficiency, but need to be combined with the controller structure, and when there are too many optimization targets, it is impossible to effectively perform a global search.
[0004] It can be seen that the current control parameter optimization algorithm still exhibits problems such as low efficiency, complex model, and unstable optimization results when dealing with the needs of simultaneous optimization of multiple parameters; how to design an efficient and simple control parameter automatic optimization method so that it can quickly obtain the optimal servo motor system control parameters in a specific application context is still quite challenging. Summary of the invention
[0005] In view of the above, the present invention provides a method for intelligent optimization of control parameters of a servo motor system, which has a simple network structure, small amount of calculation, does not require understanding of the servo system control structure, and can efficiently complete simultaneous optimization of multiple parameters.
[0006] A servo motor system control parameter intelligent optimization method comprises the following steps:
[0007] (1) Determine the input and output of the radial basis function network;
[0008] (2) Design the structure of the radial basis function network;
[0009] (3) Train the radial basis function network and use it to predict and optimize the control parameters of the servo motor system.
[0010] Furthermore, in the step (1), a position trajectory curve feature vector is generated according to a given reference position trajectory as an input of a radial basis function network, and the position trajectory curve feature vector is composed of four features: the movement time length, maximum speed, maximum acceleration, and maximum jerk of the motor rotor.
[0011] In the traditional neural network optimization architecture, the network relies on dynamically changing information such as errors as input signals, which requires a large amount of data for effective training. Especially when dealing with complex situations involving multiple links and multiple parameters, the optimization process of the model often shows instability, resulting in training results that are difficult to achieve expected results. In contrast, the feature vector of the position trajectory curve is extracted from a given reference trajectory and is a set of parameters used to characterize the characteristics of the curve. It remains constant during the motor control process. Therefore, the motor control parameters can be optimized by simply adjusting the network's pending parameters based on the response characteristics of the motor system, thereby ensuring stability during the multi-link, multi-parameter optimization process. In addition, the dimension of the feature vector of the position trajectory curve is much lower than the actual number of points in the trajectory curve, which significantly reduces the burden of network input data.
[0012] Furthermore, the output of the radial basis function network in step (1) is a servo motor system control parameter. A three-loop (position loop, speed loop and current loop) control structure is adopted for the servo motor system. The servo motor system control parameters include position control gain, velocity feedforward, velocity control gain, velocity control integral coefficient, acceleration feedforward, position command low-pass filter time constant, velocity feedforward low-pass filter time constant, velocity command low-pass filter time constant, velocity feedback command low-pass filter time constant, and current command low-pass filter time constant.
[0013] Furthermore, in step (2), the radial basis function network is composed of an input layer, a hidden layer and an output layer, wherein the input layer is composed of 4 neurons, which respectively correspond to the four features in the position trajectory curve feature vector; the hidden layer is composed of multiple neurons, which use radial basis functions to perform nonlinear feature mapping on the input layer neurons; the output layer is composed of multiple neurons, which are used to obtain the servo motor system control parameters after linearly combining the hidden layer neurons.
[0014] Furthermore, the radial basis function adopts a Gaussian function, and its expression is as follows:
[0015]
[0016] Where: b j is the value of the jth neuron in the hidden layer, x pos is the position trajectory curve feature vector, c j is the j-th row vector in the center point parameter matrix, σ j is the j-th element value in the width parameter vector, i.e., the width parameter of the j-th neuron in the hidden layer. exp() represents the natural exponential function. j is a natural number and 1≤j≤M. M is the number of neurons in the hidden layer.
[0017] Traditional neural network optimization frameworks usually use a multi-layer perceptron network structure, which consists of multiple layers of linear transformations and nonlinear activation functions. When dealing with complex optimization problems, it is usually necessary to configure more layers and neurons, which increases the difficulty of network training and the complexity of hyperparameter selection. In contrast, the Gaussian radial basis function effectively captures the nonlinear relationship in the data by mapping the input curve feature vector to a high-dimensional space. After the radial basis function, the network does not need to explicitly calculate the high-dimensional feature vector, but instead achieves mapping through the calculation of the inner product of the high-dimensional space. Therefore, the network based on the radial basis function can more efficiently complete the optimization task of multiple control parameters in multiple links while maintaining a small scale.
[0018] Furthermore, the expression of the linear combination of the output layer neurons in the hidden layer is as follows:
[0019]
[0020] Where: w ij is the value of the element in the i-th row and j-th column of the weight parameter matrix, x para,i is the value of the i-th neuron in the output layer, i.e., the i-th servo motor system control parameter. i is a natural number and 1≤i≤k. k is the number of neurons in the output layer, i.e., the number of servo motor system control parameters.
[0021] Furthermore, in the step (3), the radial basis function network is trained and predicted simultaneously. The position trajectory curve feature vector generated by the reference position trajectory is input into the radial basis function network in each control cycle. Since the radial basis function network parameters are also iteratively updated once in each control cycle, the servo motor system control parameters output by the radial basis function network are continuously optimized each time.
[0022] Furthermore, the specific process of training the radial basis function network in step (3) is as follows:
[0023] 3.1 Initialize radial basis function network parameters;
[0024] 3.2 Input the position trajectory curve feature vector generated by the reference position trajectory into the radial basis function network to obtain the servo motor system control parameters;
[0025] 3.3 Control the servo motor to track the reference position trajectory under the control parameters of the servo motor system to obtain the actual position trajectory of the servo motor;
[0026] 3.4 Calculate the loss function of the radial basis function network based on the error between the actual position trajectory and the reference position trajectory;
[0027] 3.5 Use the loss function to back-propagate the radial basis function network through the gradient descent method and update the radial basis function network parameters;
[0028] 3.6 Repeat steps 3.2 to 3.5 to iteratively update the radial basis function network parameters until the loss function or the maximum number of iterations is reached.
[0029] Furthermore, the loss function is expressed as follows:
[0030]
[0031] Among them: Loss represents the loss function, p ref and p real are the reference position trajectory and the actual position trajectory respectively, and T represents transposition.
[0032] Furthermore, the radial basis function network parameters include a center point parameter matrix (M×4 dimensions), a width parameter vector (M×1 dimensions) and a weight parameter matrix (M×k dimensions), and the iterative update formula is as follows:
[0033]
[0034]
[0035]
[0036] w'=w+η1Δw / L
[0037] c'=c+η2Δc / L
[0038] σ'=σ+η3Δσ / L
[0039] Where: w and w' are the weight parameter matrices before and after the update, Δw is the error gradient of the weight parameter matrix, c and c' are the center point parameter matrices before and after the update, Δc is the error gradient of the center point parameter matrix, σ and σ' are the width parameter vectors before and after the update, Δσ is the error gradient of the width parameter vector, x pos is the position trajectory curve feature vector, x para is the control parameter vector of the servo motor system, L is the total number of points of the reference position trajectory, η1~η3 are the given learning rate coefficients, and b represents the hidden layer.
[0040] Based on the above technical solution, the present invention has the following beneficial technical effects:
[0041] 1. The parameter optimization method based on the radial basis function network of the present invention has the advantages of simple structure, small number of unknown parameters of the optimization network, high calculation efficiency, etc. It is easy to be deployed on various low-computing power control platforms, thereby realizing online optimization of the control parameters of the servo motor system.
[0042] 2. The optimization method of the present invention does not need to consider the internal details of the control structure, does not rely on expert knowledge in the field of servo motor control, and reduces the manpower burden.
[0043] 2. The radial basis function network of the present invention can be regarded as a black box model, and multiple control parameters of the servo motor system can be optimized simultaneously based on the position error, and the optimization process is fast and efficient. BRIEF DESCRIPTION OF THE DRAWINGS
[0044] Figure 1 Schematic diagram of the structure of a servo motor control system in an embodiment of the present invention.
[0045] Figure 2 Schematic diagram of a reference position trajectory curve given in an embodiment of the present invention.
[0046] Figure 3 Schematic diagram of radial basis function network parameter optimization process in an embodiment of the present invention.
[0047] Figure 4 Schematic diagram of the loss iteration process curve of network training in an embodiment of the present invention. DETAILED DESCRIPTION
[0048] In order to describe the present invention more specifically, the technical solution of the present invention is described in detail below in conjunction with the accompanying drawings and specific implementation methods.
[0049] This implementation method takes a small servo motor with a rated power of 0.4 kW, a rated torque of 1.27 Nm, and a rated speed of 3000 r / min as an example to optimize multiple control parameters in the control structure.
[0050] like Figure 1 As shown in the figure, the servo motor control system adopts a three-loop control framework based on the magnetic field oriented control strategy, which generally includes a position loop, a speed loop and a current loop control structure. Given a reference position trajectory L is the total number of points in the trajectory. In this embodiment, L=8000 is set. Under appropriate control parameters, the actual position trajectory of the controlled motor is With a given reference position trajectory In the position loop structure, the given position trajectory first passes through the position command low-pass filter link C1 to obtain a relatively smooth position given θ0, and then passes through θ0 to compare with the actual trajectory feedback, and passes through the position ratio link C2 to obtain the speed given s1, which is used to improve the steady-state accuracy of the system; at the same time, the speed feedforward filter link C3 and the speed feedforward link C4 generate a direct speed given s2 to improve the dynamic response of the system and improve the tracking accuracy; s1 and s2 are added as the speed given s ref It acts on the subsequent speed loop. In the speed loop structure, given s ref After a speed command low-pass filter link C5, a smooth speed reference s0 is obtained, which is then compared with the speed feedback and passed through a speed proportional-integral link C6 to obtain the q-axis current reference i q1 At the same time, given s0 generates a direct current given i through the acceleration feedforward link C7 q2 , used to improve dynamic performance and enhance system stability; the speed feedback is derived from the position feedback, and it is necessary to add a speed feedback low-pass filter link C8 to smooth the feedback speed signal, i q1 and i q2 After superposition, the q-axis current given by i is generated qref , and acts on the subsequent q-axis current loop. The current loop is compared with the current feedback after passing through the current command low-pass filter link C9 and C10, and the voltage vector of the motor is obtained after passing through the proportional integral link C11 and C12, thereby controlling the motor to run according to the desired trajectory. In addition, in order to improve the stability and anti-interference ability of the system and reduce the impact of noise on control accuracy, a series of filters will be set in the three-loop control structure of the servo motor control system. The parameters of the filter may affect the response speed of the system. To ensure the overall performance of the system, it is necessary to balance noise suppression and response speed.
[0051] In order to achieve the purpose of optimizing control parameters, this embodiment establishes an optimization network with radial basis function as the core, given the position trajectory curve p refInput to the optimization network, output the control parameters to be optimized, and then these parameters are used to control the motor to track the given trajectory with high precision. The initial network parameters cannot output the optimal control parameters. It is necessary to adjust the network parameters in multiple rounds according to the given reference position trajectory curve and the current tracking error until the optimal control parameters are output. The specific implementation process is as follows:
[0052] (1) Determine the input of the radial basis function network.
[0053] In order to unify the input of the network and reduce the scale of the network, parameters are generated according to the curve of the given reference position trajectory. In this embodiment, the reference position trajectory adopts an S-shaped trajectory curve, which can smoothly control the changes in speed and acceleration, reduce the impact and vibration of the system, and improve the smoothness and accuracy of the movement; the trajectory is usually divided into three stages, namely the acceleration stage, the uniform speed stage and the deceleration stage, which can be described by the four characteristic parameters of the motor rotor's movement time length, maximum speed, maximum acceleration, and maximum jerk. Therefore, these four characteristic parameters can be combined into a characteristic vector As the input of the network. In this embodiment, the motor rotor movement time length is set to 2s, the maximum speed is 8r / s, and the maximum acceleration is 2000r / s 2 , the maximum jerk is 1000r / s, and the reference position trajectory curve is as follows Figure 2 shown.
[0054] (2) Determine the output of the radial basis function network.
[0055] Output of the network The k control parameters to be optimized need to be selected based on the three-loop control system structure. The current loop is the basic link in the control structure, responsible for adjusting the current to achieve the required torque output, but it mainly affects the system's rapid response and anti-interference capabilities; in the trajectory tracking scenario, the current loop has a relatively small direct impact on trajectory tracking, and usually pays more attention to the position loop control parameters and speed loop control parameters.
[0056] In this embodiment, the parameters to be optimized include position control gain K pp , speed feedforward K vf , speed control gain K vp , speed control integral coefficient K vi , acceleration feedforward K af , Position command low-pass filter time constant T p , speed feedforward low-pass filter time constant T pvf , speed command low-pass filter time constant T v , speed feedback command low-pass filter time constant T vf , Current command low-pass filter time constant T c, a total of 10 parameters, that is, the radial basis function network has 10 outputs, and can optimize 10 key control parameters at the same time.
[0057] (3) Determine the radial basis function network structure.
[0058] In this embodiment, the radial basis function network has three layers, namely, the input layer, the hidden layer and the output layer. The input layer transmits the original input to the hidden layer, which is composed of d neurons corresponding to x. pos ; The output layer is a linear combination of the hidden layer results to generate optimized parameters, consisting of k neurons, corresponding to x para The hidden layer performs nonlinear feature mapping, which greatly affects the final output result and is composed of M neurons. In this embodiment, d=4, k=10, and considering the high-dimensional feature processing capability and computational complexity of the network, the number of neurons in the hidden layer M is set to 20.
[0059] Input x pos After the radial basis function activation function is used, M activation states are obtained, and then k optimization parameters are obtained by linear combination; the radial basis function uses a Gaussian function, and its expression is:
[0060]
[0061] Where: b j For hidden layer The jth neuron element in ; is the center point parameter matrix The j-th row vector in represents the center point of each j-hidden layer neuron; σ j is the width parameter vector The j-th element in represents the width parameter of each j neuron.
[0062] The output of the network is x para The sum of the direct products of the basis function and the connection weight is used, that is, a linear combination is performed, which is in the form of:
[0063]
[0064] Where: w ij is the weight parameter matrix The element in the i-th row and j-th column represents the linear combination weight.
[0065] (4) Network training and control parameter optimization.
[0066] The weight parameter w, center point parameter c and width parameter σ of the radial basis function network are yet to be determined. They need to be trained through data. After the stable convergence of network parameters is obtained, the network output is the optimized control parameter. The radial basis function network optimization process is as follows Figure 3As shown, the corresponding network update process may include the following detailed steps:
[0067] Step 1: Initialize the radial basis function network parameters and control parameters x para .
[0068] Step 2: Under the current control parameter x para Under this condition, the motor is controlled to track the given reference position trajectory p ref , get the actual position trajectory p real .
[0069] Step 3: Using the reference position trajectory p ref and the actual position trajectory p real The error between them is used to calculate the loss function of the radial basis function network. The loss function adopts the minimum mean square error, and its form satisfies:
[0070]
[0071] Step 4: Using the back propagation algorithm, the gradient of the error is calculated and propagated backward layer by layer, and the network parameters are updated in the average gradient direction. The update formulas for the weight parameter w, the center point parameter c, and the width parameter σ are:
[0072]
[0073]
[0074]
[0075] w'=w+η1Δw / L
[0076] c'=c+η2Δc / L
[0077] σ'=σ+η3Δσ / L
[0078] The radial basis function network is used to optimize the control parameters. The loss change curve during the iteration is as follows: Figure 4 As shown in the figure, after 60 rounds of training, the parameters to be optimized have converged to stable values. It should be noted that the position read back is a discrete integer value. Figure 4 The initial loss is 4.30×10 9 , the loss value at the end of training is 1.67×10 7 From the change in loss, it can be seen that after the control parameters of the servo motor system are optimized by the radial basis function network, the position tracking error is significantly reduced, about 258 times.
[0079] The following table compares the position tracking error results of the servo motor system under three conditions: the optimization parameters obtained in this implementation, the initial training parameters, and the manual adjustment parameters.
[0080] Table 1
[0081]
[0082] ISE represents the sum of squares of deviations between a given position trajectory and an actual position trajectory. The results show that the ISE index is significantly reduced after the optimization parameters obtained by the present implementation are adopted. It is 867 times smaller than the manually adjusted parameters and 3933 times smaller than the initial training parameters. Obviously, after optimization, the position tracking accuracy is significantly improved.
[0083] The above description of the embodiments is to facilitate the understanding and application of the present invention by those skilled in the art. It is obvious that those skilled in the art can easily make various modifications to the above embodiments and apply the general principles described herein to other embodiments without creative work. Therefore, the present invention is not limited to the above embodiments. Improvements and modifications made by those skilled in the art to the present invention based on the disclosure of the present invention should be within the protection scope of the present invention.
Claims
1. A servo motor system control parameter intelligent optimization method, comprising the following steps: (1) Determine the input and output of the radial basis function network; (2) Design the structure of the radial basis function network; (3) Train the radial basis function network and use it to predict and optimize the control parameters of the servo motor system.
2. The servo motor system control parameter intelligent optimization method according to claim 1, Features: In the step (1), a position trajectory curve feature vector is generated according to a given reference position trajectory as an input of a radial basis function network, and the position trajectory curve feature vector is composed of four features: the movement time length, maximum speed, maximum acceleration, and maximum jerk of the motor rotor.
3. The servo motor system control parameter intelligent optimization method according to claim 1, Features: The output of the radial basis function network in step (1) is the servo motor system control parameter. A three-loop control structure is adopted for the servo motor system. The servo motor system control parameters include position control gain, velocity feedforward, velocity control gain, velocity control integral coefficient, acceleration feedforward, position command low-pass filter time constant, velocity feedforward low-pass filter time constant, velocity command low-pass filter time constant, velocity feedback command low-pass filter time constant, and current command low-pass filter time constant.
4. The servo motor system control parameter intelligent optimization method according to claim 1, Features: In the step (2), the radial basis function network is composed of an input layer, a hidden layer and an output layer, wherein the input layer is composed of 4 neurons, which respectively correspond to the four features in the feature vector of the position trajectory curve; The hidden layer is composed of multiple neurons, which use radial basis functions to perform nonlinear feature mapping on the input layer neurons; the output layer is composed of multiple neurons, which are used to obtain the servo motor system control parameters after linearly combining the hidden layer neurons.
5. The servo motor system control parameter intelligent optimization method according to claim 4, Features: The radial basis function adopts a Gaussian function, and its expression is as follows: Where: b j is the value of the jth neuron in the hidden layer, x pos is the position trajectory curve feature vector, c j is the jth row vector in the center point parameter matrix, σ j is the j-th element value in the width parameter vector, i.e., the width parameter of the j-th neuron in the hidden layer. exp() represents the natural exponential function. j is a natural number and 1≤j≤M. M is the number of neurons in the hidden layer.
6. The servo motor system control parameter intelligent optimization method according to claim 5, Features: The expression of the linear combination of the output layer and the hidden layer neurons is as follows: Where: w ij is the value of the element in the i-th row and j-th column of the weight parameter matrix, x para,i is the value of the i-th neuron in the output layer, i.e., the i-th servo motor system control parameter. i is a natural number and 1≤i≤k. k is the number of neurons in the output layer, i.e., the number of servo motor system control parameters.
7. The servo motor system control parameter intelligent optimization method according to claim 1, Features: In the step (3), the radial basis function network is trained and predicted simultaneously. The position trajectory curve feature vector generated by the reference position trajectory is input into the radial basis function network in each control cycle. Since the radial basis function network parameters are also iteratively updated once in each control cycle, the servo motor system control parameters output by the radial basis function network are continuously optimized each time.
8. The servo motor system control parameter intelligent optimization method according to claim 1, Features: The specific process of training the radial basis function network in step (3) is as follows: 3.1 Initialize radial basis function network parameters; 3.2 Input the position trajectory curve feature vector generated by the reference position trajectory into the radial basis function network to obtain the servo motor system control parameters; 3.3 Control the servo motor to track the reference position trajectory under the control parameters of the servo motor system to obtain the actual position trajectory of the servo motor; 3.4 Calculate the loss function of the radial basis function network based on the error between the actual position trajectory and the reference position trajectory; 3.5 Use the loss function to back-propagate the radial basis function network through the gradient descent method and update the radial basis function network parameters; 3.6 Repeat steps 3.2 to 3.5 to iteratively update the radial basis function network parameters until the loss function or the maximum number of iterations is reached.
9. The servo motor system control parameter intelligent optimization method according to claim 8, Features: The expression of the loss function is as follows: Among them: Loss represents the loss function, p ref and p real are the reference position trajectory and the actual position trajectory respectively, and T represents transposition.
10. The servo motor system control parameter intelligent optimization method according to claim 8, Features: The radial basis function network parameters include a center point parameter matrix, a width parameter vector, and a weight parameter matrix, and the formula for iterative update is as follows: w'=w+η 1 Δw / L c'=c+η 2 Δc / L σ'=σ+η 3 Ds / L Where: w and w' are the weight parameter matrices before and after the update, Δw is the error gradient of the weight parameter matrix, c and c' are the center point parameter matrices before and after the update, Δc is the error gradient of the center point parameter matrix, σ and σ' are the width parameter vectors before and after the update, Δσ is the error gradient of the width parameter vector, p ref and p real are the reference position trajectory and the actual position trajectory respectively, x pos is the position trajectory curve feature vector, x para is the control parameter vector of the servo motor system, L is the total number of points of the reference position trajectory, η 1 ~η 3 is a given learning rate coefficient, and b represents the hidden layer.
Citation Information
Patent Citations
A method of demand side response potential mining based on radial basis function neural network
CN109242286A
Dual-neural network self-learning IPMSM active disturbance rejection control method
CN113176731A
Sliding mode controller design method based on multi-parameter self-adaptive neural network
CN113589689A
Motor PID control method for optimizing RBF neural network based on PSO algorithm
CN115016248A
Indoor positioning method based on 5G space-time big data collaboration
CN116095600A
Cited By
Feed-forward compensation device of unknown servo system and feed-forward coefficient self-tuning method
CN119758707A
