A combined control system for a solar wing drive mechanism

Abstract

The application discloses a combined control system of a solar wing driving mechanism, which comprises a double-fuzzy PI controller and a dead-beat current predictive controller, the double-fuzzy PI controller outputs a current value based on a given angular velocity value of the driving mechanism and a difference value of an angular velocity feedback value; and the difference value and a change rate of the difference value are subjected to fuzzy processing to obtain an adjustment factor, the adjustment factor is used for online adjustment of an output result of the double-fuzzy PI controller, the control system has a certain self-adaptive ability, and can be quickly adjusted even when facing larger external interference; the dead-beat current predictive controller predicts a current value of a next period based on the output current value, and outputs a voltage value based on the current value of the next period, the voltage value acts on the driving mechanism to form a closed-loop control; the double-fuzzy PI controller avoids a relatively complicated PI parameter setting process, has a better dynamic response and smaller current harmonics, and is also beneficial to the realization of a microprocessor.

Description

Technical Field

[0001] This invention relates to the field of solar wing drive mechanism control system technology, specifically a combined control system for a solar wing drive mechanism. Background Technology

[0002] To improve the utilization rate of solar panels, modern satellites typically install a Smart Actuation and Control (SADA) mechanism to enable the solar panels to track sunlight. Due to the advantages of stepper motors, such as high positioning accuracy, no cumulative error, and ease of open-loop control, traditional SADA systems generally use a combination of stepper motors and gear reducers. However, stepper motors are affected by their own harmonic torque, frictional torque, and inherent motor pulsation, which cannot meet the high stability requirements of SADA. Furthermore, gear reduction in mechanical transmission introduces errors, delays, and vibrations. Compared to stepper motors, permanent magnet synchronous motors (PMSMs) have advantages such as high power factor, good dynamic response, and smooth operation, and are gradually replacing stepper motors as the next generation of SADA drive motors. Therefore, direct-drive PMSMs directly connected to the load can be used as SADA drive motors.

[0003] Besides the drive motor itself, the control method also affects the overall drive performance of SADA. For drive schemes using direct-drive PMSMs, the drive control of SADA is essentially the control of the direct-drive PMSM. Commonly used PMSM control strategies in engineering include vector control, constant voltage-frequency ratio control, and torque control. Vector control is widely used due to its high efficiency, stable torque, and fast dynamic response. Traditional vector control strategies often combine proportional-integral-derivative (PID) control methods to give the motor more accurate control precision. Due to the complex operating environment of outer space and low-speed or even zero-speed operating conditions, response speed, stability, and anti-interference ability are often used as the main indicators for evaluating the performance of PMSM controllers. However, simple PID controllers are difficult to meet the requirements of control indicators.

[0004] With the development of modern intelligent control methods, various new control algorithms have been applied to permanent magnet synchronous motor controllers. Fuzzy control is easy to understand and implement, and has strong fault tolerance, making it suitable for the control of nonlinear time-varying systems. Jia Dongxu et al. designed an adaptive fuzzy PI controller to address the problem of redundant torque in electric load simulators, achieving effective suppression of redundant torque. However, the controller is susceptible to strong nonlinear interference factors, and its robustness needs improvement. Predictive control is a novel control method that can effectively suppress parameter changes and uncertain disturbances. Manish et al. presented comparative experiments on finite control set model predictive control (FCS-MPC), predictive deadbeat current control (PDCC), and predictive hysteresis current control (PHCC) of permanent magnet brushless DC motors with PI control. The results show that compared with PI control, the three predictive control methods have fewer harmonic components. However, in low-speed operating environments, the control effects of FCS-MPC and PHCC are not ideal. Therefore, a combined control system for the solar wing drive mechanism is urgently needed to solve the above problems and improve the overall dynamic performance and anti-interference capability of the motor control system. Summary of the Invention

[0005] The purpose of this invention is to provide a combined control system for a solar array drive mechanism, which can effectively solve the problems existing in the prior art.

[0006] To solve the above-mentioned technical problems, the present invention adopts the following technical solution: a combined control system for a solar array drive mechanism, comprising:

[0007] A dual fuzzy PI controller outputs a current value based on the difference between the given angular velocity value of the drive mechanism and the angular velocity feedback value; and

[0008] The difference and the rate of change of the difference are fuzzy processed to obtain an adjustment factor, which is used to adjust the output of the dual fuzzy PI controller online.

[0009] The deadbeat current prediction controller predicts the current value of the next cycle based on the output current value, and outputs a voltage value based on the current value of the next cycle. The voltage value acts on the drive mechanism to form a closed-loop control.

[0010] Preferably, the dual-fuzzy PI controller includes a fuzzy controller and a PI controller; the difference and the rate of change of the difference are fuzzified by a first quantization factor and then input into the fuzzy controller for fuzzy inference, and the output value of the fuzzy inference is then defuzzified by a scaling factor to obtain the adjustment factor.

[0011] Preferably, the dual-fuzzy PI controller further includes a universe-of-discourse factor fuzzy controller. The values ​​and the rate of change of the difference are fuzzified by the second quantization factor and then input to the universe-of-discourse factor fuzzy controller for fuzzy inference and defuzzification. The output universe-of-discourse adjustment factor is then used as an input to the fuzzy controller.

[0012] Preferably, both the fuzzy controller and the universe factor fuzzy controller adopt a Mamdani-type fuzzy system.

[0013] Preferably, the fuzzy sets of the fuzzy controller and the universe factor fuzzy controller are both {NB, NM, NS, ZE, PS, PM, PB}, where NB and PB adopt Z-shaped functions and S-shaped functions respectively to ensure coverage of all input quantities, and the remaining functions are triangular functions.

[0014] Preferably, in the dq rotating coordinate system: the predicted current value of the dq axis in the (k+1)th cycle is used as the delay compensation for the deadbeat current prediction controller, wherein the deadbeat current prediction equation for delay compensation is:

[0015]

[0016] in, Predicted voltage value for the d-axis. R is the predicted voltage value along the q-axis. s For armature wire resistance, This is the d-axis compensation current value. ω is the q-axis compensation current value, where L is the stator inductance; e Let ψ be the rotor's electric angular velocity. f For rotor permanent magnet flux linkage.

[0017] Beneficial effects: In this invention, a combined control system consisting of a dual-fuzzy PI controller and a deadbeat current predictive controller is constructed to solve the problems of poor robustness and dynamic performance of motor control systems. In the dual-fuzzy PI controller, the introduction of fuzzy control concepts gives the control system a certain degree of self-adaptability, enabling it to adjust quickly even in the face of large external disturbances and achieving online optimization. Furthermore, the introduction of the universe of discourse factor accelerates the controller adjustment process and improves the control accuracy of the system. In the deadbeat current predictive controller, a discrete mathematical model of the motor is introduced to predict the current at the next moment, which avoids the relatively cumbersome PI parameter tuning process, has better dynamic response and lower current harmonics, and is also conducive to microprocessor implementation. Attached Figure Description

[0018] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used together with the embodiments of the invention to explain the invention and do not constitute a limitation thereof.

[0019] In the attached diagram:

[0020] Figure 1 This is a dq rotating coordinate system diagram of the permanent magnet synchronous motor of the present invention;

[0021] Figure 2 This is a block diagram of the dual closed-loop control structure of the present invention;

[0022] Figure 3 This is a block diagram of the dual-fuzzy PI control structure of the present invention;

[0023] Figure 4 This is a membership function setting diagram of the present invention;

[0024] Figure 5 This is a block diagram of the deadbeat current prediction control structure of the present invention;

[0025] Figure 6 This is a delay diagram of one step of the control system of the present invention;

[0026] Figure 7 This is a Simulink simulation structure block diagram of the present invention;

[0027] Figure 8 This is a comparison chart of the simulation results of the two controllers of this invention. Detailed Implementation

[0028] The embodiments of the present invention will now be described with reference to the accompanying drawings. The terminology used in the embodiments section of this invention is only for explaining specific embodiments of the invention and is not intended to limit the invention. The embodiments of this application will now be described with reference to the accompanying drawings. It will be understood by those skilled in the art that, with the development of technology and the emergence of new scenarios, the technical solutions provided in the embodiments of this application are also applicable to similar technical problems. The terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such terms can be interchanged where appropriate; this is merely a way of distinguishing objects with the same attributes in the embodiments of this application. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion, so that a process, method, system, product, or device that comprises a series of units is not necessarily limited to those units, but may include other units not explicitly listed or inherent to these processes, methods, products, or devices.

[0029] To facilitate the design of the PMSM control system, a two-phase direct-axis to quadrature-axis (dq) rotating coordinate system is used for PMSM modeling and analysis. The principle is as follows: Figure 1As shown in the figure; the magnets represent the rotor poles of the permanent magnet synchronous motor. A stationary orthogonal coordinate system α-β and a two-phase rotating coordinate system dq are established with the pole center as the origin o. The axis parallel to the pole direction is the direct axis (d), and the axis perpendicular to the pole direction is the quadrature axis (q). Given a rotor mechanical angular velocity ω... m With the coordinate system remaining unchanged, the dq coordinate system moves at an angular velocity ω m Rotate counterclockwise at a constant speed around the origin o;

[0030] The stator flux linkage equation of the permanent magnet synchronous motor in the dq rotating coordinate system is:

[0031]

[0032] Where, ψ d ψ q These are the dq-axis components of the stator flux linkage, L d L q These are the dq axis inductance components, i d i q These are the dq-axis components of the stator current, ψ f For rotor permanent magnet flux linkage;

[0033] The stator voltage equation is:

[0034] Among them, u d u q These are the dq-axis components of the stator voltage, R s ω is the armature resistance. e The rotor's electrical angular velocity is ω, which is related to the rotor's mechanical angular velocity. m There is ω e =p n .ω m ;

[0035] The electromagnetic torque equation is:

[0036]

[0037] T e p is the electromagnetic torque. n This represents the number of pole pairs of the motor.

[0038] The mechanical motion equation of the motor rotor is:

[0039] Where J is the rotor moment of inertia, T m Where is the load torque and B is the damping coefficient.

[0040] In the vector control strategy of permanent magnet synchronous motor, the d-axis reference current This method is widely used because of its good control effect and ease of implementation.

[0041] Example: Figure 2 As shown, a combined control system for a solar array drive mechanism includes a dual fuzzy PI controller and a deadbeat-free current predictive controller. The dual fuzzy PI controller is based on a given angular velocity value of the drive mechanism. and angular velocity feedback value ω m The difference is output as the current value, which is used as the q-axis current setpoint. The d-axis current is a given value. In this process, the difference and the rate of change of the difference are fuzzy processed to obtain an adjustment factor, which is used to adjust the output of the dual fuzzy PI controller online; the deadbeat current prediction controller predicts the current value of the next cycle based on the output current value, and outputs a voltage value based on the current value of the next cycle, which acts on the drive mechanism to form a closed-loop control.

[0042] In high vacuum environments, to prevent problems such as solar panel resonance caused by excessively fast SADA drive speed, the motor usually operates at extremely low speed or even zero speed. To ensure smooth start of the drive motor, traditional motor speed loops often use PI controllers to eliminate start-up overshoot. However, PI controllers have a relatively simple structure and poor robustness when facing external interference, which cannot meet the stability requirements of SADA drive systems.

[0043] For dual-fuzzy PI controllers:

[0044] The system includes a fuzzy controller and a PI controller. The difference and the rate of change of the difference are fuzzified by a first quantization factor and then input into the fuzzy controller for fuzzy inference. The output value of the fuzzy inference is then defuzzified by a proportional factor to obtain the adjustment factor. The dual-fuzzy PI controller also includes a universe-of-discourse factor fuzzy controller. The value and the rate of change of the difference are fuzzified by a second quantization factor and then input into the universe-of-discourse factor fuzzy controller for fuzzy inference and defuzzification. The output is a universe-of-discourse adjustment factor, which is then input into the fuzzy controller. Through multiple adjustment factors, the PI controller has a certain degree of adaptability, optimizing the speed loop control effect.

[0045] Among them, reference Figure 3 As shown, the angular velocity is given. With angular velocity ω m The difference e and the rate of change ec of e are respectively processed by the first quantization factor K. e1 and K c1 After fuzzification, E1 and EC1 serve as the two inputs to the fuzzy controller; E1 and EC1 are then processed through fuzzy inference to output U1 and U2, which are then processed by a scaling factor K. U1 and K U2 After defuzzification, the change in the proportional coefficient ΔK of the PI controller is obtained.p and the change in integral coefficient ΔK i As an adjustment factor, it enables online adjustment of the PI controller parameters; the principle of the universe-of-discourse factor fuzzy controller is roughly the same, also using e and ec through a second quantization factor K. e2 and K c2 The fuzzified E2 and EC2 are used as inputs, and after fuzzy inference and defuzzification, the output domain adjustment factor λ is generated. λ is a quantity that varies with the input, and it also serves as an input to the fuzzy controller to adjust the controller's output.

[0046] In this system, both the fuzzy controller and the universe-of-discourse factor fuzzy controller adopt the Mamdani type fuzzy system. The universes of discourse for the inputs E2 and EC2, and the output U3 of the universe-of-discourse factor fuzzy controller are set to [-0.6, 0.6], [-0.6, 0.6], and [0, 2], respectively. The universes of discourse for the inputs E1 and EC1 of the fuzzy controller are both [-0.6λ, 0.6λ], while the universes of discourse for the outputs U1 and U2 are [-120, 120] and [-6, 6], respectively. The fuzzy sets for both fuzzy controllers are {NB, NM, NS, ZE, PS, PM, PB}, and all membership functions adopt the following... Figure 4 The functions shown are in the form of Z-shaped and S-shaped functions, respectively, to ensure that all input quantities are covered and to avoid control misalignment. The remaining functions are triangular functions.

[0047] For the fuzzy controller rules, at the start of motor receiving angular velocity commands, a larger proportional coefficient is set to improve response speed; when the motor angular velocity reaches near the predetermined angular velocity, the integral coefficient is increased and the proportional coefficient is appropriately decreased to reduce overshoot. For the universe of discourse factor fuzzy controller rules, at the start of motor receiving angular velocity commands, the universe of discourse of the fuzzy controller is appropriately widened to improve response speed; when the motor angular velocity reaches near the predetermined angular velocity, the universe of discourse is appropriately compressed to improve control accuracy. The fuzzy rules determined based on the above analysis are shown in the table below; the defuzzification method for both fuzzy controllers is the area centroid method.

[0048]

[0049]

[0050] Compared to traditional PI control, the introduction of fuzzy control gives the control system a certain degree of self-adaptability, enabling it to adjust quickly even in the face of significant external disturbances and achieving online optimization. Furthermore, the introduction of the universe of discourse factor accelerates the controller adjustment process and improves the system's control accuracy. Finally, the lookup table method of fuzzy control is also beneficial for microprocessor implementation, enhancing its engineering application value.

[0051] For deadbeat current prediction controllers:

[0052] To avoid chattering issues caused by overshoot during startup, traditional PMSM vector control systems typically use PI controllers for the d-axis and q-axis of the current loop. However, PI controller parameter tuning requires separate tuning for each axis, which is relatively cumbersome. Therefore, a deadbeat controller is used to replace the PI controller for closed-loop control of the current loop, avoiding the complex parameter tuning process. The specific principle is as follows: Figure 5 As shown in the figure, k is the kth sampling period.

[0053] The principle of the deadbeat control algorithm is as follows:

[0054]

[0055] To obtain the discretized stator voltage equations of the system, a sampling time of T = 10 is chosen. -6 s is small enough that the above equation can be discretized using the forward Euler method:

[0056]

[0057] In the deadbeat algorithm, in order to make the stator current i d and i q Tracking the current setpoint and The dq-axis current in the (k+1)th sampling period is replaced by the current setpoint value of the corresponding kth sampling period, i.e. Furthermore, for SPMSM, we have approximately L d =L q =L, simplified to:

[0058]

[0059] in, Predicted voltage value for the d-axis. Predict the voltage value for the q-axis.

[0060] refer to Figure 6 As shown: Due to limitations in microprocessor hardware, the voltage applied to the motor after PWM triggering is not the control quantity for the current cycle, but rather the calculation result from the previous cycle. This results in a delay; the delay leads to inaccurate predicted voltage results for the current cycle, thus reducing the dynamic performance of the system; the delay is compensated by predicting the dq-axis current value for the (k+1)th cycle.

[0061]

[0062] in, This is the d-axis compensation current value. This is the q-axis compensation current value;

[0063] Therefore, the deadbeat current prediction equation with delay compensation is:

[0064]

[0065] in, Predicted voltage value for the d-axis. R is the predicted voltage value along the q-axis. s For armature wire resistance, This is the d-axis compensation current value. ω is the q-axis compensation current value, where L is the stator inductance; e Let ψ be the rotor's electric angular velocity. f For rotor permanent magnet flux linkage.

[0066] Compared to traditional PI control methods, deadbeat current predictive control introduces a discrete mathematical model of the motor to predict the current at the next moment. This avoids the relatively cumbersome PI parameter tuning process, while also exhibiting better dynamic response and lower current harmonics. Furthermore, it is easier to implement with a microprocessor.

[0067] Based on the above, a simulation experiment was conducted:

[0068] A combination where both the velocity loop and current loop are PI controllers is defined as a dual PI controller; a combination where the velocity loop is a dual fuzzy PI controller and the current loop is a deadbeat controller is defined as a combined controller. Based on the theories of deadbeat controllers and dual fuzzy PI controllers, a Simulink simulation model is built to experimentally verify the combined controller. The simulation structure is as follows: Figure 7 As shown in the figure; the left area is for inputting angular velocity and load values, the middle area is for comparing the dual PI controller and the combined controller, and the right area is for outputting the actual angular velocity value;

[0069] To ensure the authenticity of the experiment, 0.065° / s and 0.03° / s, as required by the project, were selected as the system step response input values. A sine curve with an amplitude of 0.03° / s and a frequency of 1kHz was selected as the system sinusoidal response input value. A sudden change in load was applied to the system to verify its anti-interference capability. The initial proportional coefficient of the speed loop controller of both the dual PI controller and the combined controller was K. p =75, and the integral coefficients are all K. i =1, the proportional coefficient K of the current loop controller of the dual PI controller p =1000, integral coefficient K i=1; The total simulation test duration is 1.5ms, with loads of 1.5N, 15N, and -1.5N applied to the system at 0.5ms, 0.7ms, and 1.0ms respectively; The parameters of the dq-PMSM motor used in the simulation are shown in the table below;

[0070]

[0071] The final simulation results are as follows Figure 8 As shown, by Figure 8 (a) and Figure 8 (b) It is known that during the step signal application phase and the recovery phase after the external load is applied to the system, the average settling times of the dual-PI controller and the combined controller are 0.076 ms and 0.053 ms, respectively; during the external load disturbance phase, the average angular velocity errors of the dual-PI controller and the combined controller are 0.034° / s and 0.026° / s, respectively; Figure 8 (c) and Figure 8 (d) It is known that, in the stage without external load, the average angular velocity errors of the dual PI controller and the combined controller are 0.027° / s and 0.015° / s, respectively; in the stage with external load interference, the average angular velocity errors of the dual PI controller and the combined controller are 0.051° / s and 0.037° / s, respectively. These data indicate that, compared to the dual PI controller, the combined controller has smaller average settling time and smaller average angular velocity error.

[0072] To further analyze the overall control performance of the controller, this experiment introduces the standard deviation η of the angular velocity as the system's control performance factor. The control performance factor reflects the degree of deviation between the actual angular velocity value and the desired angular velocity value. The smaller the control performance factor, the better the tracking effect of the control system and the better the overall control effect of the controller. The formula for the control performance factor is as follows:

[0073]

[0074] Where, Δω m denoted as angular velocity error, and N as the number of sampling points.

[0075] The calculated control performance factors of the dual PI controller and the combined controller under different input signals are shown in the table below:

[0076]

[0077] As shown in the table, compared with the dual PI controller, the combined controller has a smaller control performance factor for both step and sinusoidal responses, indicating that the combined controller has better overall control performance. In addition, the tracking performance of the control system is better when the input signal is sinusoidal than when the input signal is step, which is consistent with the expected experimental results and confirms the validity of the experimental results.

[0078] In summary, under low-speed operating conditions, compared to the traditional dual-loop PI controller, the combined controller reduces the average settling time of the step response by 30.0%, and the average angular velocity error of the sinusoidal response by 46.0% without external load. Under external load disturbance, the combined controller reduces the average angular velocity error of the step effect and sinusoidal response by 22.6% and 28.3%, respectively, indicating that the combined controller has a faster response speed and stronger anti-interference capability. Furthermore, the average dispersion of the angular velocity of the combined controller is also smaller, confirming its superior performance.

[0079] The embodiments of the present invention have been described in detail above with reference to the accompanying drawings. However, the present invention is not limited to the above embodiments. For those skilled in the art, after learning the contents described in the present invention, several equivalent changes and substitutions can be made without departing from the principle of the present invention. These equivalent changes and substitutions should also be considered to fall within the protection scope of the present invention.

Claims

1. A combined control system for a solar array drive mechanism, characterized in that, include: A dual fuzzy PI controller outputs a current value based on the difference between the given angular velocity value of the drive mechanism and the angular velocity feedback value. as well as The difference and the rate of change of the difference are fuzzy processed to obtain an adjustment factor, which is used to adjust the output of the dual fuzzy PI controller online. The deadbeat current prediction controller predicts the current value of the next cycle based on the output current value, and outputs a voltage value based on the current value of the next cycle. The voltage value acts on the drive mechanism to form a closed-loop control. In the dq rotating coordinate system: the predicted current value of the dq axis in the (k+1)th cycle is used as the delay compensation for the deadbeat current prediction controller, where the deadbeat current prediction equation for delay compensation is: ； in, Predicted voltage value for the d-axis. R is the predicted voltage value along the q-axis. s For armature wire resistance, This is the d-axis compensation current value. Here, L is the q-axis compensation current value, and L is the stator inductance. The rotor's electric angular velocity, For rotor permanent magnet flux linkage.

2. The combined control system for a solar array drive mechanism according to claim 1, characterized in that: The dual-fuzzy PI controller includes a fuzzy controller and a PI controller; the difference and the rate of change of the difference are fuzzified by the first quantization factor and then input into the fuzzy controller for fuzzy inference, and the output value of the fuzzy inference is then defuzzified by the scaling factor to obtain the adjustment factor.

3. The combined control system for a solar array drive mechanism according to claim 2, characterized in that: The dual-fuzzy PI controller also includes a universe-of-discourse factor fuzzy controller. The difference and the rate of change of the difference are fuzzified by the second quantization factor and then input to the universe-of-discourse factor fuzzy controller for fuzzy inference and defuzzification. The output is a universe-of-discourse adjustment factor, which is then input to the fuzzy controller as an input quantity.

4. The combined control system for a solar array drive mechanism according to claim 3, characterized in that: Both the fuzzy controller and the universe factor fuzzy controller adopt the Mamdani-type fuzzy system.

5. The combined control system for a solar array drive mechanism according to claim 4, characterized in that: The fuzzy sets of the fuzzy controller and the universe factor fuzzy controller are both {NB, NM, NS, ZE, PS, PM, PB}, where NB and PB adopt Z-shaped and S-shaped functions to ensure coverage of all input quantities, and the remaining functions are triangular functions.

Images