Self-adaptive control system and method for cutting transmission system of coal mining machine under complex working conditions

By using adaptive torque compensation controller and multi-layer convolutional neural network fuzzy control in the coal mining machine cutting transmission system, the problems of dynamic load impact and system parameter adjustment under complex operating conditions are solved, and the coal mining machine is realized safety and productivity improvement.

CN119087816AActive Publication Date: 2024-12-06JIANGSU ZHONGJI MINE EQUIP +1
View PDF 4 Cites 0 Cited by

Patent Information

Application Number
CN202411233733.X
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-09-04
Publication Date
2024-12-06
Estimated Expiration
2044-09-04

AI Technical Summary

Technical Problem

Under complex working conditions, it is difficult for the coal mining machine to effectively suppress the impact of the driving load, and the prior art cannot realize adaptive adjustment of the system parameters, which affects the safety and production efficiency of the coal mining machine.

Method used

Adoptable control system of the cutting transmission system of coal miner under complex working conditions, including sample construction module, adaptive torque compensation controller, superposition module and cutting motor speed adjustment module, using second-order nonlinear self-immunization The torque module, adaptive convolutional neural network fuzzy module and BP neural network are controlled by dynamic adjustment factors and multi-layer convolutional neural network to realize adaptive adjustment of system parameters.

Benefits of technology

Effectively suppress the impact of the driving load, improve the system response speed and control accuracy, reduce the impact of time-varying parameters on the controller, realize adaptive compensation control in complex working conditions, and improve the safety and production efficiency of the coal mining machine.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119087816A_ABST
    Figure CN119087816A_ABST
Patent Text Reader

Abstract

The invention discloses a self-adaptive control system and method for a cutting transmission system of a coal mining machine under a complex working condition, and the method comprises the steps: tracking a state variable of the cutting transmission system through an extended state observer in a self-adaptive torque compensation controller, and estimating the disturbance of nonlinearity, parameter time varying and the like of the system in real time; compensating the real-time action quantity of the total disturbance to a feedforward compensation disturbance term in the actual control quantity; the BP network performs adaptive adjustment on parameters of the second-order active-disturbance-rejection torque controller, and the adaptive convolutional neural network fuzzy module realizes adaptive adjustment on output compensation of the active-disturbance-rejection torque compensation controller to obtain expert control rules meeting complex impact working conditions. Calculating an initial electromagnetic torque by combining non-linear state error feedback with an adaptive convolutional neural network fuzzy module adjustment amount, generating a compensation electromagnetic torque after feedforward compensation, and superposing a torque compensation amount generated by a PI controller; the self-adaptive compensation control on the dynamic load suppression of the system and the effective suppression on the dynamic load impact energy of the cutting transmission system are realized.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention belongs to the technical field of coal mining machine control, and in particular relates to an adaptive control system and method for a coal mining machine cutting transmission system under complex working conditions. Background Art

[0002] Coal is the basic energy source in my country. The mechanization degree of coal mining has reached 85%, and the mechanization degree of tunneling has reached 65%. The contribution rate of scientific and technological innovation to the development of the industry has further increased. The informatization and intelligent construction of coal mines has made new progress, and a number of advanced and efficient smart coal mines have been built. Therefore, higher requirements are put forward for the mechanization, automation and working condition adaptability of coal mining equipment.

[0003] Due to the complexity and variability of underground coal seams, coal seams contain high-strength rock gangue, hard inclusions, rock faults, etc., which makes the external load acting on the drum have the characteristics of randomness and strong impact, which can easily lead to an increase in the dynamic load of the cutting transmission system, making the cutting part of the coal mining machine one of the weak parts of the whole machine. Due to the complex cutting environment, achieving reliable operation and efficient production of coal mining machines is an inevitable requirement for the automation and unmanned operation of coal mining machines. When the properties of the coal seam change, the adjustable motion parameters are the drum height and traction speed, but the existing technology does not realize the speed regulation of the cutting drum; under different cutting conditions, when the traction speed varies within a large range, the constant speed cutting of the drum is difficult to ensure that the coal mining machine is in the best working state with strong cutting ability, low cutting specific energy consumption, high productivity and other comprehensive performance. Therefore, under unmanned operation of the coal mining machine, it is necessary to achieve the optimal comprehensive coal mining performance under different cutting conditions through the variable speed cutting of the coal mining machine drum and the coordinated control of the cutting-traction motion parameters.

[0004] At present, some scholars have used active control to reduce the fatigue load of electromechanical transmission systems caused by sudden changes in external loads in the fields of wind power and rolling mills. That is, by designing a damping controller to superimpose a compensating torque on the original generator / motor torque to increase the electrical damping of the transmission chain. Some scholars have established a torsional vibration damping controller based on a bandpass filter (BPF) to suppress the impact load of the wind power gear transmission system caused by sudden changes in external loads and avoid the resonance of the transmission chain caused by torsional vibration. They have proposed an active control strategy for system fatigue loads based on BPF with the generator speed as the feedback signal to suppress multiple resonant frequencies of the system. However, when there are model parameter disturbances and uncertainty factors in the controlled object, the stability of the damping controller based on the BPF and its suppression performance of the fatigue load of the transmission system will be affected. In view of the shortcomings of the above controllers, some scholars have established a damping controller based on the controlled object model with the generator speed as the feedback signal, and used the Kalman filter to estimate the aerodynamic torque load of the system; however, this method relies on relatively accurate transmission system parameters. In fact, the electromechanical transmission system has a complex structure and time-varying parameters, and it is difficult to obtain accurate system parameters. In order to address the shortcomings of the above methods, some scholars have applied the anti-disturbance control technology that does not rely on the precise mathematical model of the controlled object and can automatically compensate for system parameter disturbances to the field of torsional vibration suppression of rolling mills and wind power transmission systems. However, these studies simplified the complex transmission system into an inertial mass model during modeling, only considered the torsional vibration of the shaft system, and did not establish a dynamic model including the gears. It is impossible to consider the time-varying meshing stiffness of the gears and the dynamic load of the gear transmission system, and it is impossible to analyze the influence of time-varying parameters on the performance of the anti-disturbance control controller. Summary of the invention

[0005] In response to the problems existing in the prior art, the present invention provides an adaptive control system and method for the cutting transmission system of a coal mining machine under complex working conditions, which can effectively suppress the impact energy of dynamic loads under complex cutting conditions, and realize a control method and control system for adaptively adjusting system parameters, so as to improve the safety and production efficiency of the coal mining machine under complex cutting conditions.

[0006] To solve the above technical problems, the present invention provides the following technical solutions: an adaptive control system for a coal mining machine cutting transmission system under complex working conditions, comprising: a sample construction module, an adaptive torque compensation controller, a superposition module, and a cutting motor speed adjustment module;

[0007] The sample construction module is used to calculate the difference between the actual rotation speed of the cutting motor and the rotation speed of the high-speed gear of the cutting transmission system; and to construct a training sample data set based on the rotation speed difference and the corresponding controller output compensation torque;

[0008] The adaptive torque compensation controller is used to output the compensation torque of the controller, and uses the sample data set to train the adaptive torque compensation control neural network. During the training process, a dynamic adjustment factor is introduced to dynamically adjust the controller output; the adaptive torque compensation controller includes a second-order nonlinear anti-disturbance torque module, an adaptive convolutional neural network fuzzy module, and a BP neural network;

[0009] The second-order nonlinear anti-disturbance torque module includes a tracking differentiator, an extended state observer, and a nonlinear error feedback law module; the second-order nonlinear anti-disturbance torque module introduces a dynamic adjustment factor to enhance the adaptability and robustness of the anti-disturbance controller, and the dynamic adjustment factor dynamically adjusts the control output according to the operating state of the system and external disturbances; the adaptive convolutional neural network fuzzy module is used to improve the working condition adaptability of the second-order nonlinear anti-disturbance torque controller, and obtain the expert control rules of the cutting transmission system under different coal-rock impacts, and generate an adaptive adjustment amount ΔT according to the working condition changes e ; BP neural network is used to optimize the parameters of NLSEF and ESO in the second-order nonlinear active disturbance rejection torque controller;

[0010] The superposition module superimposes the controller output compensation torque and the PI controller output torque to obtain the sum of the torques;

[0011] The cutting motor speed adjustment module uses the sum of the torques as the motor controller input, obtains the motor controller output, and adjusts the cutting motor speed in real time according to the output.

[0012] Furthermore, the aforementioned tracking differentiator is configured to perform the following actions to arrange a transition process for the input signal:

[0013] Where T is the sampling step; r is the speed factor; d = rh0; h0 is the filter factor, ω 11 is the input signal after transition, ω 22 ω 11 The derivative of , fst is the fastest control comprehensive function;

[0014] The extended state observer is configured to perform the following actions:

[0015]

[0016] Where y(k) is the system output; e(k) is the difference between the output of the extended state observer and the system output; β1, β2, β3 are constants; α1, α2, α3 are nonlinear parameters, all of which are positive values.

[0017] The nonlinear error feedback control law module is configured to perform the following actions:

[0018]

[0019] Where β4 and β5 are the proportional factor and differential factor respectively.

[0020] Furthermore, the aforementioned dynamic adjustment factors include a change rate factor, α, a disturbance factor, β, and an error factor γ; wherein γ=|e|, e is the control error, the change rate factor α, disturbance factor β, and error factor γ are weighted and fused to obtain the comprehensive adjustment factor λ, λ=s1α+s2β+s3γ, where s1, s2, s3 are weighted coefficients.

[0021] Furthermore, the aforementioned second-order nonlinear anti-disturbance torque module compensates the output torque as follows:

[0022] T e2 =(1+λ)[β4fal(e1(t),a4,δ)+β5fal(e2(t),a5,δ)-z3 / b0].

[0023] Furthermore, the aforementioned BP neural network includes an input layer, a hidden layer and an output layer; the input layer includes 4 neural nodes to track the differentiator output signal e1, and its differential e2, the control system total output y and 1 as neuron input; the hidden layer is determined to be 6 in combination with the cutting transmission system; the output layer has 6 nodes, corresponding to the second-order nonlinear self-disturbance rejection torque controller parameters β1, β2, β3, β4, β5 respectively.

[0024] Furthermore, the aforementioned adaptive convolutional neural network fuzzy module adopts an adaptive neural-fuzzy control system ANFIS combined with a multi-layer convolutional neural network structure, specifically, a multi-layer convolutional neural network CNN is introduced before the input layer of the adaptive neural-fuzzy control system ANFIS. The adaptive convolutional neural network fuzzy module is used to improve the working condition adaptability of the second-order nonlinear self-disturbance rejection torque controller, and obtain the expert control rules of the cutting transmission system under different coal-rock impacts, and generate an adaptive adjustment amount ΔT according to the working condition changes. e ;

[0025] The adaptive convolutional neural network fuzzy module includes: a data input module, which converts the output signal of the tracking differentiator into a format suitable for CNN processing and performs normalization and standardization processing to form CNN time series slices;

[0026] The multi-layer convolutional neural network CNN, including multiple convolutional layers, activation function layers, and pooling layers, processes the output signal of the tracking differentiator through multiple layers, extracts the high-level features of the input data, and finally outputs the feature vector through the fully connected layer; the feature fusion module fuses the features extracted by the multi-layer convolutional neural network CNN with other input features of the adaptive neural-fuzzy control system ANFIS, and inputs them into the adaptive neural-fuzzy control system ANFIS for adaptive fuzzy reasoning and decision-making.

[0027] Furthermore, the aforementioned adaptive neural-fuzzy control system ANFIS includes an input layer, a fuzzification layer, a fuzzy rule layer, a fuzzy reasoning layer, and a defuzzification layer;

[0028] The input layer receives the fused feature vector and defines the fused feature vector of ANFIS input as x i =[e1(k),e2(k)] T ;

[0029] The fuzzification layer assigns a linguistic variable value to each node through fuzzy inference rules, and calculates the membership function of each input component belonging to the fuzzy set of each linguistic variable value. The membership function is represented by a Gaussian function, and its formula is: Where i = 1, 2; j = 1, 2, ..., m j , m j is x i The fuzzy segmentation number of ij and δ ij are the center and width of the membership function respectively;

[0030] The fuzzy rule layer is used to define fuzzy rules, namely Where, i1∈{1,2,...,m1},i1∈{1,2,...,m1},…,i1∈{1,2,...,m1},j=1,2,...,m, Perform fuzzy reasoning;

[0031] The fuzzy reasoning layer combines fuzzy rules, calculates the applicability of each rule, and performs normalization calculation, that is,

[0032]

[0033] The defuzzification layer converts the fuzzy inference results into clear control outputs In the formula, w ij is the connection weight between the fourth layer nodes and the output layer nodes.

[0034] Furthermore, when the aforementioned adaptive neural-fuzzy control system ANFIS performs adaptive fuzzy reasoning and decision-making, a performance index function is defined: Where, t i is the target output; calculated using the gradient descent method Then use the first-order gradient optimization algorithm to adjust w ij , c ij , δ ij , the first-order gradient is:

[0035]

[0036] The adaptive fuzzy inference parameter learning algorithm is:

[0037]

[0038] 9. The method for adaptive control of a cutting transmission system of a coal mining machine under complex working conditions according to claim 8, characterized in that the adaptive convolutional neural network fuzzy module is configured to perform the following actions: T tra =-1×(T me -T ste ), where T me is the electromagnetic torque output by the motor itself, T ste is the torque in the opposite direction to the electromagnetic torque output by the motor, T tra is the steady-state target value of the motor electromagnetic torque; combined with the controller input feature vector, continuous training can obtain the adaptive convolutional neural network for the control system to compensate for the electromagnetic torque adjustment ΔT e .

[0039] The final output torque of the adaptive torque compensation controller is:

[0040] T e2 =(1+λ)[β4fal(e1(t),a4,δ)+β5fal(e2(t),a5,δ)+ΔT e ]-z3 / b0

[0041] Superposition module, superimposition PI control output torque Te1;

[0042] The cutting motor speed adjustment module outputs the total output compensation torque of the coal mining machine adaptive control system motor control:

[0043] T eM =Te1+(1+λ)[β4fal(e1(t),a4,δ)+β5fal(e2(t),a5,δ)+ΔT e ]-z3 / b0.

[0044] The present invention also provides a control method based on an adaptive control system of a coal mining machine cutting transmission system under complex working conditions, comprising the following steps:

[0045] S1, determine whether the cutting impedance of the shearer drum changes, if yes, determine it is a full coal working condition, and execute step S2, otherwise continue to determine whether it is a gangue-interposed working condition, if yes, execute step S2, otherwise determine it is a fault working condition, and execute step S2;

[0046] S2, determine the motor speed and impedance, input the signal to the adaptive torque compensation controller, the controller compensates the electromagnetic torque in real time according to the signal, and outputs the electromagnetic torque to adjust the cutting motor speed, traction speed and drum speed;

[0047] S3, determine whether the cutting impedance changes, if yes, execute step S2, otherwise end.

[0048] Compared with the prior art, the beneficial technical effects of the present invention using the above technical solution are as follows:

[0049] 1. The present invention realizes the application of self-disturbance rejection technology in the cutting transmission system of coal mining machines, overcoming the problem that the cutting working conditions of coal mining machines in deep coal seams are harsh and the cutting motor adopts traditional direct torque control to be difficult to quickly and effectively suppress the dynamic load of the cutting transmission system of coal mining machines caused by impact loads. By introducing a dynamic adjustment factor to improve the controller in the second-order nonlinear self-disturbance rejection torque compensation controller, the compensation torque can more accurately adapt to the actual working condition changes. The dynamic adjustment factor can automatically adjust the compensation strength according to the real-time working condition, improve the response speed and control accuracy of the system, and thus more effectively suppress the dynamic load impact energy. This control method does not rely on the mathematical model of the controlled object. Compared with the traditional direct torque control method, this control method not only has good tracking and estimation performance, but also is insensitive to the time-varying parameters of the system, and can reduce the influence of time-varying parameters on the second-order nonlinear self-disturbance rejection torque controller.

[0050] 2. The present invention introduces a multi-layer convolutional neural network (CNN) before the adaptive neural-fuzzy control system (ANFIS) to perform hierarchical feature extraction on the input data. CNN is good at processing high-dimensional data, can extract multi-level features and more accurately capture complex nonlinear relationships. By introducing a multi-layer convolutional neural network (CNN) after the input layer of ANFIS, the learning ability and generalization ability of the system can be improved, the feature extraction ability of the system can be enhanced, and the processing effect of complex input data can be improved.

[0051] 3. The present invention adopts a control method combining an adaptive convolutional neural network fuzzy controller with a second-order nonlinear anti-disturbance torque compensation controller, which solves the problem of the adaptive adjustment capability of the output torque compensation of the second-order nonlinear anti-disturbance torque compensation controller under complex working conditions. Combining the advantages of multi-layer convolutional neural network control and fuzzy control, the adaptive adjustment of the output compensation of the anti-disturbance torque compensation controller is realized, and the expert control rules that meet the complex impact working conditions are obtained. The adaptive adjustment amount can be generated according to the changes in the working conditions, thereby realizing the adaptive compensation control of the system dynamic load suppression. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] Figure 1 It is the schematic diagram of the adaptive control system of the coal mining machine under complex working conditions.

[0053] Figure 2 It is the structure diagram of adaptive torque compensation controller.

[0054] Figure 3 It is the structure diagram of the second-order nonlinear anti-disturbance torque compensation controller.

[0055] Figure 4 This is the structure diagram of the BP neural network.

[0056] Figure 5 It is the structure diagram of the adaptive convolutional neural network fuzzy controller.

[0057] Figure 6 It is the structural diagram of the adaptive neural network-fuzzy control module.

[0058] Figure 7 It is the flow chart of the adaptive control system of coal mining machine under complex working conditions. DETAILED DESCRIPTION

[0059] In order to better understand the technical content of the present invention, specific embodiments are described below in conjunction with the accompanying drawings.

[0060] Various aspects of the invention are described herein with reference to the accompanying drawings, in which many illustrative embodiments are shown. The embodiments of the invention are not limited to those described in the accompanying drawings. It should be understood that the invention is implemented by any of the various concepts and embodiments described above, as well as the concepts and embodiments described in detail below, because the concepts and embodiments disclosed in the invention are not limited to any implementation. In addition, some aspects disclosed in the invention may be used alone or in any appropriate combination with other aspects disclosed in the invention.

[0061] like Figure 1 As shown in the figure, it is the principle diagram of the adaptive control system of the coal mining machine under complex working conditions. This control method collects the speed ω of one end of the flexible shaft of the cutting motor mThe difference Δω1 between the flexible shaft end speed ω1 of the gear transmission system is used as the input of the adaptive torque compensation controller. The output Te2 of the adaptive torque compensation controller is superimposed on the output Te1 of the PI controller as the total compensation torque Te of the motor control. m The cutting motor speed is adjusted in real time. This control method does not rely on the mathematical model of the controlled object. Compared with the traditional direct torque control method, this control method not only has good tracking and estimation performance, but is also insensitive to the time-varying parameters of the system. It can more effectively suppress the impact energy of external dynamic loads in complex cutting environments.

[0062] The present invention provides an adaptive control system for a coal mining machine cutting transmission system under complex working conditions, comprising: a sample construction module, an adaptive torque compensation controller, a superposition module, and a cutting motor speed adjustment module;

[0063] A sample construction module is used to calculate the difference between the actual speed of the cutting motor and the speed of the high-speed gear of the cutting transmission system; and to construct a training sample data set based on the speed difference and the corresponding controller output compensation torque;

[0064] like Figure 2 The figure shows the structure of the adaptive torque compensation controller, which includes a second-order nonlinear anti-disturbance torque module, an adaptive convolutional neural network fuzzy module, and a BP neural network. The controller is used to output the compensation torque, and the sample data set is used to train the adaptive torque compensation control neural network. During the training process, a dynamic adjustment factor is introduced to dynamically adjust the controller output.

[0065] The second-order nonlinear anti-disturbance torque module includes a tracking differentiator, an extended state observer, and a nonlinear error feedback law module; the second-order nonlinear anti-disturbance torque module introduces a dynamic adjustment factor to enhance the adaptability and robustness of the anti-disturbance controller, and the dynamic adjustment factor dynamically adjusts the control output according to the operating state of the system and external disturbances; the adaptive convolutional neural network fuzzy module is used to improve the working condition adaptability of the second-order nonlinear anti-disturbance torque controller, and obtain the expert control rules of the cutting transmission system under different coal-rock impacts, and generate an adaptive adjustment amount ΔT according to the working condition changes e ; BP neural network is used to optimize the parameters of NLSEF and ESO in the second-order nonlinear active disturbance rejection torque controller;

[0066] The superposition module superimposes the controller output compensation torque and the PI controller output torque to obtain the sum of the torques;

[0067] The cutting motor speed adjustment module uses the sum of the torques as the motor controller input, obtains the motor controller output, and adjusts the cutting motor speed in real time according to the output.

[0068] like Figure 3As shown in the figure, the structure of the second-order nonlinear active disturbance rejection torque compensation controller is shown, in which the tracking differential controller (TD) arranges the transition process for the input signal and is configured to perform the following actions:

[0069]

[0070] Where T is the sampling step; r is the speed factor; d = rh0; h0 is the filter factor. 11 is the input signal after transition, ω 22 ω 11 The derivative of . fst is the fastest control synthesis function.

[0071] The Extended State Observer (ESO) is configured to perform the following actions:

[0072]

[0073] Where y(k) is the system output; e(k) is the difference between the observed output of the extended state observer and the system output; β1, β2, β3 are constants; α1, α2, α3 are nonlinear parameters, all of which are positive values; the extended state observer is established based on the measured output and control input of the cutting transmission system, estimates the system and external disturbances, and compensates the torque in real time through the disturbance estimation value.

[0074] The nonlinear error feedback law module (NLSEF) is configured to perform the following actions:

[0075]

[0076] Where β4 and β5 are the proportional factor and differential factor respectively.

[0077] The dynamic adjustment factors introduced include the change rate factor α, the disturbance factor β, and the error factor γ; γ=|e|, e is the control error. The above factors are weighted and fused to form a comprehensive adjustment factor λ, λ=s1α+s2β+s3γ, where s1, s2, s3 are weighted coefficients.

[0078] The second-order nonlinear active disturbance rejection torque controller adopts disturbance compensation to compensate the real-time action of the total disturbance to the actual control quantity. After disturbance compensation, the compensation output torque of the second-order nonlinear active disturbance rejection torque controller is:

[0079] T e2 =(1+λ)[β4fal(e1(t),a4,δ)+β5fal(e2(t),a5,δ)-z3 / b0]

[0080] like Figure 4As shown in the figure, it is the structure diagram of BP neural network. Since there are too many and fixed parameters in the second-order nonlinear anti-disturbance torque controller, the torque compensation output under different cutting conditions is limited. The purpose of introducing BP neural network is to optimize the parameters of NLSEF and ESO in the anti-disturbance controller. This BP neural network adopts a 3-layer BP neural network, including input layer, hidden layer and output layer. The input layer uses 4 neural nodes to track the output signal e1 of the differentiator, its differential e2, the total output y of the control system and 1 as neuron input. The hidden layer is determined to be 6 in combination with the cutting transmission system. There are 5 nodes in the output layer, corresponding to the parameters β1, β2, β3, β4, and β5 of the second-order nonlinear anti-disturbance torque controller respectively. The BP neural network controller adjusts the parameters of NLSEF and ESO according to the changes of the controlled object and external disturbances, which helps to improve the accuracy of state estimation, improve the robustness of the system, and improve the control effect.

[0081] like Figure 5 The figure shows the structure of the adaptive convolutional neural network fuzzy controller. In order to further improve the working condition adaptability of the second-order nonlinear anti-disturbance torque controller, this adaptive convolutional neural network fuzzy module is designed. This controller adopts the structure of adaptive neural-fuzzy control system (ANFIS) combined with multi-layer convolutional neural network (CNN). CNN is good at processing high-dimensional data, can extract multi-level features and more accurately capture complex nonlinear relationships. By introducing a multi-layer convolutional neural network (CNN) after the input layer of ANFIS, the learning ability and generalization ability of the system can be improved. Figure 5 The specific structure of the adaptive convolutional neural network fuzzy controller includes: a data input module, which converts the output signal of the tracking differentiator into a format suitable for CNN processing and normalizes and standardizes it to form CNN time slices. A multi-layer convolutional neural network (CNN) module, including multiple convolutional layers, activation function layers, and pooling layers. After multi-layer processing, it extracts high-level features of the input data and finally outputs the feature vector through the fully connected layer. A feature fusion module, which fuses the features extracted by CNN with other input features of ANFIS and inputs them into ANFIS for fuzzy reasoning and decision-making. An adaptive neural-fuzzy reasoning module.

[0082] like Figure 6 As shown in the figure, it is the structure diagram of the adaptive neural network-fuzzy control module, including the input layer, which receives the fused feature vector and defines the fused feature vector x input by ANFIS. i =[e1(k),e2(k)] T The fuzzification layer assigns a linguistic variable value to each node through fuzzy inference rules, and calculates the membership function of each input component belonging to the fuzzy set of each linguistic variable value. The membership function is represented by a Gaussian function, and its formula is: Where i = 1, 2; j = 1, 2, ..., m j , m j is x i The fuzzy segmentation number of ij and δ ij are the center and width of the membership function respectively. Fuzzy rule layer, define fuzzy rules, namely Where, i1∈{1,2,...,m1},i1∈{1,2,...,m1},…,i1∈{1,2,...,m1},j=1,2,...,m, Perform fuzzy reasoning; the fuzzy reasoning layer combines fuzzy rules, calculates the applicability of each rule, and performs normalization calculation, that is, (j=1,2,...,m). Defuzzification layer, converting fuzzy inference results into clear control outputs In the formula, w ij is the connection weight between the fourth layer nodes and the output layer nodes.

[0083] In the adaptive neuro-fuzzy inference module, define the performance index function Where, t i is the target output. Calculated using the gradient descent method Then use the first-order gradient optimization algorithm to adjust w ij , c ij , δ ij , the first-order gradient is:

[0084]

[0085] The parameter learning algorithm of the adaptive neural-fuzzy inference module is obtained as follows:

[0086]

[0087] In order to obtain expert control rules under different working conditions, the training algorithm of adaptive convolutional neural network fuzzy controller is defined as: tra =-1×(T me -T ste ), where T me is the electromagnetic torque output by the motor itself, T ste is the torque in the opposite direction to the electromagnetic torque output by the motor, T tra is the steady-state target value of the motor electromagnetic torque. Combined with the controller input feature vector, continuous training can be used to obtain the adaptive convolutional neural network fuzzy controller for the control system to compensate for the electromagnetic torque adjustment ΔT e .

[0088] The final output torque of the adaptive torque compensation controller is:

[0089] T e2=(1+λ)[β4fal(e1(t),a4,δ)+β5fal(e2(t),a5,δ)+ΔT e ]-z3 / b0

[0090] Superimposing the PI control output torque Te1, the total output compensation torque of the coal mining machine adaptive control system is:

[0091] T eM =Te1+(1+λ)[β4fal(e1(t),a4,δ)+β5fal(e2(t),a5,δ)+ΔT e ]-z3 / b0

[0092] refer to Figure 7 On the other hand, the present invention also provides a control method based on an adaptive control system of a coal mining machine cutting transmission system under complex working conditions, comprising the following steps:

[0093] S1, determine whether the cutting impedance of the shearer drum changes, if yes, determine it is a full coal working condition, and execute step S2, otherwise continue to determine whether it is a gangue-interposed working condition, if yes, execute step S2, otherwise determine it is a fault working condition, and execute step S2;

[0094] S2, determine the motor speed and impedance, input the signal to the adaptive torque compensation controller, the controller compensates the electromagnetic torque in real time according to the signal, and outputs the electromagnetic torque to adjust the cutting motor speed, traction speed and drum speed;

[0095] S3, determine whether the cutting impedance changes, if yes, execute step S2, otherwise end.

[0096] Although the present invention has been described above with preferred embodiments, it is not intended to limit the present invention. A person skilled in the art of the present invention may make various modifications and improvements without departing from the spirit and scope of the present invention. Therefore, the scope of protection of the present invention shall be determined by the claims.

Claims

1. An adaptive control system for a coal mining machine cutting transmission system under complex working conditions, characterized in that: include: A sample construction module, an adaptive torque compensation controller, a superposition module, and a cutting motor speed adjustment module; The sample construction module is used to calculate the difference between the actual rotation speed of the cutting motor and the rotation speed of the high-speed gear of the cutting transmission system; and to construct a training sample data set based on the rotation speed difference and the corresponding controller output compensation torque; The adaptive torque compensation controller is used to output the compensation torque of the controller, use the sample data set to train the adaptive torque compensation control neural network, introduce a dynamic adjustment factor during the training process, and dynamically adjust the controller output; The adaptive torque compensation controller includes a second-order nonlinear anti-disturbance torque module, an adaptive convolutional neural network fuzzy module, and a BP neural network; The second-order nonlinear anti-disturbance torque module includes a tracking differentiator, an extended state observer, and a nonlinear error feedback law module; the second-order nonlinear anti-disturbance torque module introduces a dynamic adjustment factor to enhance the adaptability and robustness of the anti-disturbance controller, and the dynamic adjustment factor dynamically adjusts the control output according to the operating state of the system and external disturbances; the adaptive convolutional neural network fuzzy module is used to improve the working condition adaptability of the second-order nonlinear anti-disturbance torque controller, and obtain the expert control rules of the cutting transmission system under different coal-rock impacts, and generate an adaptive adjustment amount ΔT according to the working condition changes e ; BP neural network is used to optimize the parameters of NLSEF and ESO in the second-order nonlinear active disturbance rejection torque controller; The superposition module superimposes the controller output compensation torque and the PI controller output torque to obtain the sum of the torques; The cutting motor speed adjustment module uses the sum of the torques as the motor controller input, obtains the motor controller output, and adjusts the cutting motor speed in real time according to the output.

2. The adaptive control system for the cutting transmission system of a coal mining machine under complex working conditions according to claim 1 is characterized in that: The tracking differentiator is configured to perform the following actions to arrange a transition process for the input signal: Where T is the sampling step; r is the speed factor; d = rh0; h0 is the filter factor, ω 11 is the input signal after transition, ω 22 ω 11 The derivative of , fst is the fastest control comprehensive function; The extended state observer is configured to perform the following actions: In the formula, y(k) is the system output; e(k) is the difference between the output of the extended state observer and the system output; β1, β2, β3 are constants; α1, α2, α3 are nonlinear parameters, all of which are positive values. The nonlinear error feedback control law module is configured to perform the following actions: Where β4 and β5 are the proportional factor and differential factor respectively.

3. The adaptive control system for the cutting transmission system of a coal mining machine under complex working conditions according to claim 2 is characterized in that: The dynamic adjustment factors include the change rate factor, α, the disturbance factor β, and the error factor γ; e is the control error. The rate of change factor α, disturbance factor β, and error factor γ are weighted and fused to obtain the comprehensive adjustment factor λ, λ=s1α+s2β+s3γ, where s1, s2, s3 are weighted coefficients.

4. The adaptive control method of the cutting transmission system of a coal mining machine under complex working conditions according to claim 3 is characterized in that: The second-order nonlinear self-disturbance rejection torque module compensates the output torque as follows: T e2 =(1+λ)[β4fal(e1(t),a4,δ)+β5fal(e2(t),a5,δ)-z3 / b0].

5. The adaptive control method of the cutting transmission system of a coal mining machine under complex working conditions according to claim 4 is characterized in that: The BP neural network includes an input layer, a hidden layer and an output layer; the input layer includes 4 neural nodes to track the differentiator output signal e1, its differential e2, the control system total output y and 1 as neuron input; the hidden layer is determined to be 6 in combination with the cutting transmission system; the output layer has 6 nodes, corresponding to the second-order nonlinear self-disturbance rejection torque controller parameters β1, β2, β3, β4, β5 respectively.

6. The adaptive control system for the cutting transmission system of a coal mining machine under complex working conditions according to claim 5 is characterized in that: The adaptive convolutional neural network fuzzy module adopts the adaptive neural-fuzzy control system ANFIS combined with a multi-layer convolutional neural network structure. Specifically, a multi-layer convolutional neural network CNN is introduced before the input layer of the adaptive neural-fuzzy control system ANFIS. The adaptive convolutional neural network fuzzy module is used to improve the working condition adaptability of the second-order nonlinear anti-disturbance torque controller, and obtain the expert control rules of the cutting transmission system under different coal-rock impacts, and generate an adaptive adjustment amount ΔT according to the working condition changes. e ; The adaptive convolutional neural network fuzzy module includes: a data input module, which converts the output signal of the tracking differentiator into a format suitable for CNN processing and performs normalization and standardization processing to form CNN time series slices; The multi-layer convolutional neural network CNN, including multiple convolutional layers, activation function layers, and pooling layers, processes the output signal of the tracking differentiator through multiple layers, extracts the high-level features of the input data, and finally outputs the feature vector through the fully connected layer; the feature fusion module fuses the features extracted by the multi-layer convolutional neural network CNN with other input features of the adaptive neural-fuzzy control system ANFIS, and inputs them into the adaptive neural-fuzzy control system ANFIS for adaptive fuzzy reasoning and decision-making.

7. The adaptive control system for the cutting transmission system of a coal mining machine under complex working conditions according to claim 6 is characterized in that: The adaptive neuro-fuzzy control system ANFIS includes input layer, fuzzification layer, fuzzy rule layer, fuzzy reasoning layer and defuzzification layer; The input layer receives the fused feature vector and defines the fused feature vector of ANFIS input as x i =[e1(k),e2(k)] T ; The fuzzification layer assigns a linguistic variable value to each node through fuzzy inference rules, and calculates the membership function of each input component belonging to the fuzzy set of each linguistic variable value. The membership function is represented by a Gaussian function, and its formula is: Where i = 1, 2; j = 1, 2, ..., m j , m j is x i The fuzzy segmentation number of ij and δ ij are the center and width of the membership function respectively; The fuzzy rule layer is used to define fuzzy rules, namely Where, i1∈{1,2,...,m1},i1∈{1,2,...,m1},…,i1∈{1,2,...,m1},j=1,2,...,m, Perform fuzzy reasoning; The fuzzy reasoning layer combines fuzzy rules, calculates the applicability of each rule, and performs normalization calculation, that is, The defuzzification layer converts the fuzzy inference results into clear control outputs In the formula, w ij is the connection weight between the fourth layer nodes and the output layer nodes.

8. The adaptive control method of the cutting transmission system of a coal mining machine under complex working conditions according to claim 7 is characterized in that: When the adaptive neural-fuzzy control system ANFIS performs adaptive fuzzy reasoning and decision-making, it defines the performance index function In the formula, t i is the target output; calculated using the gradient descent method Then use the first-order gradient optimization algorithm to adjust w ij , c ij , δ ij , the first-order gradient is: The adaptive fuzzy inference parameter learning algorithm is:

9. The adaptive control method of the cutting transmission system of a coal mining machine under complex working conditions according to claim 8 is characterized in that: The adaptive convolutional neural network fuzzy module is configured to perform the following actions: T tra = -1×(T me -T ste ), where T me is the electromagnetic torque output by the motor itself, T ste is the torque in the opposite direction to the electromagnetic torque output by the motor, T tra is the steady-state target value of the motor electromagnetic torque; combined with the controller input feature vector, continuous training can obtain the adaptive convolutional neural network for the control system to compensate for the electromagnetic torque adjustment ΔT e . The final output torque of the adaptive torque compensation controller is: T e2 =(1+λ)[β4fal(e1(t),a4,δ)+β5fal(e2(t),a5,δ)+ΔT e ]-z3 / b0 Superposition module, superimposition PI control output torque Te1; The cutting motor speed adjustment module outputs the total output compensation torque of the coal mining machine adaptive control system motor control: T eM =Te1+(1+λ)[β4fal(e1(t),a4,δ)+β5fal(e2(t),a5,δ)+ΔT e ]-z3 / b0。 10. A control method based on an adaptive control system of a coal mining machine cutting transmission system under complex working conditions, characterized in that: The following steps are involved: S1, determine whether the cutting impedance of the shearer drum changes, if yes, determine it is a full coal working condition, and execute step S2, otherwise continue to determine whether it is a gangue-interposed working condition, if yes, execute step S2, otherwise determine it is a fault working condition, and execute step S2; S2, determine the motor speed and impedance, input the signal to the adaptive torque compensation controller, the controller compensates the electromagnetic torque in real time according to the signal, and outputs the electromagnetic torque to adjust the cutting motor speed, traction speed and drum speed; S3, determine whether the cutting impedance changes, if yes, execute step S2, otherwise end.

Citation Information

Patent Citations

  • Non-uniformly distributed energy-saving scraper conveyor and control method

    CN116101726A

  • Moment optimal control method for cutting part of multi-servo drive coal mining machine

    CN117335689A

  • Active-disturbance-rejection control method and system for permanent magnet synchronous motor

    CN118100724A

  • AGV tracing active-disturbance-rejection controller based on adaptive neural network and method thereof

    CN118348794A