A Multi-Valve Collaborative Control Method and System Based on Group Optimization and Anti-Saturation

Abstract

This invention discloses a multi-way valve collaborative control method and system based on swarm optimization and anti-saturation, comprising the following steps: using PID gain parameters and anti-saturation gain parameters as individual control parameters, and using the time-weighted absolute error integral as the fitness function, the overall swarm optimization algorithm is used to iteratively optimize and output the optimal control parameters; when in a non-saturated linear operating state, the theoretical and actual control quantities of the current are output according to the standard PID mode based on the PID gain parameters; when in a saturated state, the theoretical and actual control quantities of the current are output according to the anti-saturation compensated PID mode based on the PID gain parameters and anti-saturation gain parameters; the actual control quantity of the current is converted into the actual current driving the two electromagnet coils of the proportional valve and controlling the actual displacement of the valve core, which has the control effect of high control accuracy, fast dynamic response, strong anti-saturation capability, and good robustness.

Description

Technical Field

[0001] This invention belongs to the field of electro-hydraulic proportional control technology, specifically relating to a multi-way valve collaborative control method and system based on group optimization and anti-saturation. Background Technology

[0002] Electro-hydraulic proportional valves are widely used in robotics and industrial hydraulic systems due to their high power density and fast dynamic response. However, electro-hydraulic proportional multi-way valve systems exhibit complex high-order dynamic characteristics, strong nonlinearity, and model uncertainty, making it difficult for traditional PID tuning methods such as the Ziegler-Nichol (ZN) method to achieve the expected performance, resulting in problems such as poor dynamic performance and tracking errors.

[0003] The spool displacement control of a proportional multi-way valve driven by dual proportional electromagnets is affected by several nonlinear factors: 1. Proportional electromagnet dead zone: the output force is almost zero when the current is below a certain threshold; 2. Spool spring pre-compression force: the initial resistance generated by the spool centering spring; 3. Static friction: Coulomb friction between the spool and the valve body; 4. Hydraulic force: the force exerted on the spool due to changes in flow velocity and direction during liquid flow. Under traditional control methods, electro-hydraulic proportional valves generally have a dead zone, which typically accounts for 10% to 30% of the rated input signal value. To improve the control accuracy of proportional valves, the dead zone must be overcome, and its adverse effects must be minimized to improve control performance.

[0004] Challenges in Electro-hydraulic Proportional Valve Control: Electro-hydraulic proportional multi-way valves are widely used in high-performance hydraulic actuators due to their high power density and fast response. However, their control faces multiple challenges: a) strong nonlinearity, including flow-pressure characteristics, dead zone, and hysteresis; b) actuator saturation, with physical limitations on valve spool displacement or drive current; c) parameter sensitivity and coupling effects. Traditional proportional-integral-derivative (PID) controllers have fixed parameters, and when faced with the above complex dynamics, they are prone to integral saturation (windup), leading to large system overshoot, long settling time, and even instability, severely restricting further improvements in control accuracy. Current Technological Improvements and Shortcomings: Parameter optimization methods: To improve PID performance, researchers use intelligent methods such as particle swarm optimization (PSO) and genetic algorithm (GA) to tune PID gain parameters. However, the objective function of PID optimization (such as ITAE) is usually nonlinear and multimodal. Traditional optimization algorithms have limited search capabilities and often suffer from premature convergence and local optimum traps, resulting in higher engineering application costs.

[0005] Anti-integral saturation strategy: Back-Calculation Anti-Windup (BCAW) is a classic dynamic compensation method that can effectively suppress integrator accumulation when the output is saturated. However, the tuning of its compensation gain Ks is itself a difficult problem, and traditional empirical tuning is difficult to obtain the global optimal performance.

[0006] Current control logic: In order to compensate for the dead zone of hydraulic valves and improve the valve core drive efficiency, some methods use current control logic based on the deviation direction. However, it is usually applied in open-loop control and is separated from parameter optimization and anti-saturation mechanism, lacking systematic integration.

[0007] In summary, most existing technologies address a single problem in parameter tuning, anti-saturation, or driving logic in isolation, failing to construct a composite control architecture that is collaboratively optimized and considers the overall situation. As a result, the performance improvement of closed-loop control is limited when dealing with complex operating conditions that are highly dynamic, strongly nonlinear, and frequently saturated. Summary of the Invention

[0008] In view of the above, the purpose of this invention is to provide a multi-way valve collaborative control method and system based on group optimization anti-saturation. By deeply integrating overall group optimization, integral anti-saturation compensation and actuator command current superposition logic, it achieves collaborative optimization of the entire process from parameter optimization, dynamic compensation to execution drive, and has the control effect of high control accuracy, fast dynamic response, strong anti-saturation capability and good robustness.

[0009] To achieve the above-mentioned objectives, an embodiment provides a multi-way valve cooperative control system based on group optimization and anti-saturation, comprising: The global optimization layer uses control parameters such as PID gain parameters and anti-saturation gain parameters as individuals, and employs a global swarm optimization algorithm to iteratively optimize and output the optimal control parameters using the time-weighted absolute error integral as the fitness function. The dynamic compensation control layer is used to output the theoretical and actual control quantities of the current according to the standard PID mode based on the PID gain parameters when the linear operation is in a non-saturated state; and to output the theoretical and actual control quantities of the current according to the anti-saturation compensation PID mode based on the PID gain parameters and anti-saturation gain parameters when the operation is in a saturated state. The logic driving layer is used to convert the actual control quantity output by the dynamic compensation control layer into the actual current driving the two electromagnet coils of the proportional valve and control the actual displacement of the valve core.

[0010] Preferably, the absolute error is the absolute value of the error between the valve core displacement command reference value and the actual normalized displacement feedback value.

[0011] Preferably, the fitness function is expressed as: in, Represents the fitness function. and These represent the initial and final times of the integration process, respectively. Indicates the current time. Indicates time The absolute error at that time. Preferably, the initial time. and termination time Set them to 0.05 seconds and 0.25 seconds respectively.

[0012] Preferably, the theoretical and actual control quantities of the output current, based on the PID gain parameters and anti-saturation gain parameters in the PID mode with anti-saturation compensation, include: in, , , ,as well as These represent proportional gain, integral gain, derivative gain, and anti-saturation gain, respectively. Indicates time The error between the valve core displacement command reference value and the actual normalized displacement feedback value. This indicates the actuator's maximum rated saturation current.

[0013] Preferably, converting the actual control quantity into the actual current driving the two electromagnet coils of the proportional valve includes: in, It is the current in the coil of proportional electromagnet 1. This refers to the current in coil 2 of the proportional electromagnet; the maximum rated current is... , , The initial current, which serves as the threshold current, represents the current value required for a single proportional electromagnet to output force just enough to overcome the pre-compression force of the slide valve spring. This represents the actual control quantity of the current.

[0014] Preferably, the threshold current is determined in the following manner: An open-loop control strategy is adopted, and the threshold current for skipping the dead zone is calibrated according to the one-sided control method. The current value required to make the output force of a single proportional electromagnet just overcome the nonlinear factors and generate a displacement value is found.

[0015] To achieve the above-mentioned objectives, embodiments of the present invention also provide a multi-way valve cooperative control method based on group optimization and anti-saturation, the method employing the above-mentioned system and including the following steps: Using a global optimization layer, the PID gain parameter and anti-saturation gain parameter are treated as individuals. The time-weighted absolute error integral is used as the fitness function. The global swarm optimization algorithm is used to iteratively optimize and output the optimal control parameters. Feedback anti-saturation control is performed using a dynamic compensation control layer. When the system is in a non-saturated linear operating state, the theoretical and actual control quantities of the output current are based on the PID gain parameters and follow the standard PID mode. When the system is in a saturated state, the theoretical and actual control quantities of the output current are based on the PID gain parameters and anti-saturation gain parameters and follow the anti-saturation compensation PID mode. The logic driving layer is used to convert the actual control quantity output by the dynamic compensation control layer into the actual current driving the two electromagnet coils of the proportional valve and control the actual displacement of the valve core.

[0016] Compared with the prior art, the beneficial effects of the present invention include at least the following: 1. Global optimal performance: The global optimization layer uses a holistic swarm optimization algorithm that leverages overall population information (such as root mean square fitness) to guide the search. It dynamically allocates the movement coefficient by calculating the difference between the individual fitness and the mean fitness of the population. Combined with simulated annealing selection and population adaptive mutation mechanism, it collaboratively optimizes the PID gain parameter and anti-saturation gain parameter from a global search perspective. This effectively improves premature convergence and local optima, increases training efficiency, and saves engineering application costs.

[0017] 2. High efficiency in resisting saturation: The feedback anti-saturation mechanism adopted by the dynamic compensation control layer after parameter optimization can quickly and smoothly suppress integral saturation, significantly reduce overshoot and settling time, and improve dynamic response characteristics.

[0018] 3. High-precision tracking: The single-sided threshold current superposition logic used in the logic driving layer, combined with the optimized controller, can achieve a steady-state error close to zero (for example, 0% steady-state error in step response), and the mean absolute error (MAE) of tracking a 1Hz sine signal is reduced by more than 50% compared with the traditional ZN-PID.

[0019] 4. Strong robustness and adaptability: The three-layer architecture makes the system more robust to valve nonlinearity and input changes. It maintains excellent tracking performance in sinusoidal signal tests with different amplitudes (25%-100%).

[0020] 5. Strong engineering applicability: The method is modular and clear, and easy to implement in embedded controllers (such as DSP and PLC), providing a directly deployable solution for high-precision electro-hydraulic proportional control systems. Attached Figure Description

[0021] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0022] Figure 1 This is a schematic diagram of the structure of the multi-way valve cooperative control system based on group optimization and anti-saturation provided in the embodiment; Figure 2 This is a flowchart of the multi-valve collaborative control method based on group optimization and anti-saturation provided in the embodiment; Figure 3 This is a diagram showing the relationship between the control current command and the main valve core displacement under open-loop control provided in the embodiment (static characteristics). Figure 4 This is the convergence curve of the fitness function value of HSO / PID in 100 optimization iterations provided in the example; Figure 5 This is the convergence curve of the fitness function value of HSO / BC-AW in 100 optimization iterations provided in the example; Figure 6 This is a dynamic performance comparison chart provided in the embodiment; Figure 7 , Figure 8 , Figure 9 ,as well as Figure 10 These are comparison charts showing the tracking performance under the sinusoidal input signal conditions provided in the embodiments, with amplitude values ​​of (100%), (75%), (50%), and (25%). Figure 11 This is a comparison chart of the dead zone compensation effect provided in the embodiment. Detailed Implementation

[0023] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and do not limit the scope of protection of this invention.

[0024] The inventive concept of this invention is to address the dead zone problem caused by various nonlinear factors affecting the valve core displacement control of a proportional multi-way valve driven by dual proportional electromagnets; to solve the problems of poor electro-hydraulic control accuracy and decreased dynamic system response performance caused by difficulties in control parameter tuning and continuous saturation of the PID integral term under complex engineering problems; and to solve the problem of zero-position dead zone lag caused by nonlinear factors such as electromagnet delay when using sinusoidal input. Therefore, this embodiment provides a multi-way valve collaborative control scheme based on group optimization and anti-saturation. Through global optimization of control parameters, dynamic compensation control, and logic driving, the above three problems are solved simultaneously, achieving precise control of the proportional multi-way valve driven by dual proportional electromagnets.

[0025] like Figure 1 As shown in the embodiment, the multi-valve collaborative control system based on group optimization and anti-saturation includes a global optimization layer, a dynamic compensation control layer, and a logic driving layer.

[0026] In this embodiment, the global optimization layer runs offline and is responsible for finding the optimal control parameters for the dynamic compensation control layer. This will be explained in detail from three aspects: the optimization algorithm, the optimization objective, and the optimization process.

[0027] Optimization Algorithm: The Human Population Optimization (HSO) algorithm is adopted. Compared with Particle Swarm Optimization (PSO) and Genetic Algorithm (GA), HSO uses overall population information (such as root mean square fitness) to guide the search. It dynamically allocates movement coefficients by calculating the difference between individual fitness and population mean, and combines simulated annealing selection and adaptive mutation mechanisms to achieve a better balance between exploration and exploitation, making it easier to find the global optimum.

[0028] Optimization objective: Use the time-weighted integral of absolute error (ITAE) as the fitness function. in, Represents the fitness function. and These represent the initial and final times of the integration process, respectively. Indicates the current time. Indicates the reference value for valve core displacement command. Compared with the actual normalized displacement feedback value The error between them Indicates error The absolute value of.

[0029] Optimization process: The HSO population is initialized with random values ​​within a given range of control parameters to ensure a comprehensive exploration of the feasible parameter space. For each individual in the HSO population (i.e., a set of control parameters)... The fitness function is evaluated and calculated. The value drives the HSO iteration until convergence, and outputs the optimal control parameters. ,in, , , These represent the proportional gain, integral gain, and derivative gain, respectively, and are the PID gain parameters. This represents the anti-saturation gain parameter. Minimizing ITAE allows for a balance between response speed, overshoot, and steady-state accuracy.

[0030] Note: To ensure consistency of initial conditions, the displacement of the main spool valve of the proportional multi-way valve is stabilized at zero under the action of the centering spring during the initial 0-0.05 seconds. Subsequently, a step command is applied at 0.05 seconds, and the system response is considered complete at 0.25 seconds. Therefore, the start and end times of the ITAE calculation are set to 0.05 seconds and 0.25 seconds, respectively.

[0031] In this embodiment, the dynamic compensation control layer is an online core controller, and the input to the controller is the valve core displacement command reference value. Compared with the actual normalized displacement feedback value error The controller outputs the theoretical control quantity (the calculated output value of the controller before passing through the limiting module). Actual control quantity (The controller's calculated output value when passing through the limiting module) or control current command.

[0032] In the dynamic compensation control layer, the BCAW (Back-Calculation Anti-Windup Method) compensation mechanism: when the theoretical control quantity... Exceeding the actuator saturation limit When the actuator (proportional electromagnet coil) reaches its maximum rated current, a saturation error occurs. This saturation error is controlled by the anti-saturation gain parameter. Feedback is sent to the integrator input to dynamically correct the integral term and prevent its continuous accumulation. The key innovation lies in the inverse integral saturation gain. The parameters, like the PID gain parameters, are obtained through collaborative optimization by the global optimization layer using the HSO algorithm, rather than being independently tuned or entirely based on experience, thus ensuring the optimal match between anti-saturation action and control response.

[0033] When operating in a non-saturated linear state, the theoretical control quantity of the output current is based on the PID gain parameter and follows the standard PID mode. and actual control quantity ,Right now: When the calculated output value remains within the limit range, that is... The controller will operate in standard PID mode.

[0034] When in a saturated state (i.e.) When the anti-saturation mechanism is activated, the theoretical control quantity of the output current is determined according to the PID gain parameter and the anti-saturation gain parameter in the PID mode with anti-saturation compensation. and actual control quantity ,Right now: When saturation occurs, the BCAW method adjusts the controller's dynamic characteristics to quickly exit the saturation state. The key change in the control output occurs in the integral term, where a correction feedback term is introduced. This term is proportional to the difference between the calculated control output and the actual control output, thereby effectively reducing the integral rate during saturation.

[0035] In this embodiment, the logic driving layer is used for unilateral threshold current superposition, mainly to superimpose the actual control amount of the current output by the dynamic compensation control layer. This is converted into the actual current driving the two electromagnet coils of the proportional valve. The logic is shown below, aiming to solve the nonlinear dead-zone problem of the proportional valve: in, It is the current in the coil of proportional electromagnet 1. This refers to the current in coil 2 of the proportional electromagnet; the maximum rated current is... , , The initial current, which serves as the threshold current, represents the current value required for a single proportional electromagnet to output force just enough to overcome the pre-compression force of the slide valve spring. These values ​​are typically calibrated experimentally and reflect the size of the dead zone.

[0036] It should also be noted that the displacement of the main slide valve of the proportional multi-way valve is stabilized at zero position under the action of the centering spring, that is, zero position is maintained.

[0037] This logic not only achieves directional superposition of current, but more importantly, it is deeply coupled with the upper-level global optimization layer and dynamic compensation control layer. In the global optimization layer, the HSO algorithm optimizes parameters within a complete closed loop that includes this current-driven logic. Therefore, the optimized PID gain parameters and anti-saturation gain parameters are aware of and adapt to the downstream logic drive characteristics, thus achieving the optimal mapping from control commands to physical drives.

[0038] like Figure 2 As shown in the embodiment, a multi-valve cooperative control method based on group optimization to resist saturation is also provided. This method uses the above-mentioned system and includes the following steps: Step S1: Determine the threshold current when the single-sided threshold currents are superimposed.

[0039] An open-loop control strategy is adopted. First, the threshold current for skipping the dead zone is calibrated using the traditional single-sided control method. Then, the current value required for the output force of a single proportional electromagnet to just overcome the nonlinear factors such as the pre-compression force and friction of the slide valve spring before displacement is generated is found. This value is then substituted into... Figure 1 The dead zone in the static characteristics has been completely eliminated by testing and verifying the superposition formula of the single-sided threshold current. The results are as follows: Figure 3 As shown.

[0040] Step S2: Using the global optimization layer, the control parameters such as PID gain parameters and anti-saturation gain parameters are treated as individuals, and the time-weighted absolute error integral is used as the fitness function. The global swarm optimization algorithm is used to iteratively optimize and output the optimal control parameters.

[0041] First, the HSO algorithm is coded, specifically the HSO algorithm script, and its fitness function is written. For ITAE. Algorithm parameters are set as follows: population size is 50, maximum number of iterations is 100, initial temperature is 10000, cooling rate is 0.995, and mutation rate is linearly reduced from 0.5 to 0.1.

[0042] Then, parameter optimization is performed by running the HSO optimization program. The search range for optimization parameters is shown in Table 1.

[0043] Table 1 The HSO algorithm optimization process is set with a population size of 50 and 100 iterations until termination, at which point the optimal set of control parameters is returned. Key settings for the optimization algorithm also include: an initial simulated annealing temperature of 10,000 °C and a cooling rate of 0.995; in the population adaptive mutation mechanism, the adaptive mutation rate and step size are linearly reduced from 0.5 to 0.1, and from 0.3 to 0.1, respectively. Under the same settings, the convergence curves of the fitness function values ​​of the PID control (HSO / PID) and the feedback anti-saturation control (HSO / BCAW) optimized by the HSO algorithm in 100 optimization iterations are shown below. Figure 4 and Figure 5 As shown in Table 2, the optimal parameter sets for each controller are summarized in Table 2.

[0044] Table 2 Step S3: The optimal control parameters obtained through iterative optimization are input to the dynamic supplementary control layer for dynamic control output of the actual control quantity. The actual control quantity is then input to the logic drive layer to convert and output the actual current of the two electromagnet coils of the proportional valve and control the actual displacement of the valve core.

[0045] Deploy the iteratively optimized optimal control parameters to Figure 1 In the online controller shown, a series of performance test signals are applied to the system, such as a step signal with an amplitude of 50% and a sine signal with a frequency of 1Hz and a variable amplitude.

[0046] Table 3 Table 3 shows the comparison results of dynamic performance indicators. For example... Figure 6As shown in Table 3, the significant overshoot and persistent oscillations observed in the ZN-tuned PID controller primarily stem from the fundamental mismatch between its linear design philosophy and the inherent nonlinear characteristics of the electro-hydraulic system. The ZN method derives parameters based on the response of a linearized model under critical instability conditions. While pursuing aggressive control performance, it fails to adequately consider key nonlinear factors, particularly the current saturation effect of the actuator electromagnet. When the control output reaches saturation, the integrator continues to accumulate errors, leading to integral saturation. This integral saturation effect causes the integrator's decoupling process to lag behind the system's dynamic characteristics, resulting in significant overshoot. Subsequently, as the integrator slowly recovers from saturation, persistent oscillations are induced. Therefore, the aggressive parameters obtained by the ZN method not only fail to achieve the expected control performance but also amplify and exacerbate the system's nonlinear dynamic characteristics, ultimately leading to unstable response and performance degradation. In contrast, both the HSO / PID controller and the HSO / BC-AW controller exhibit superior and near-optimal dynamic performance. Although the overall performance index (ITAE) of the standard HSO / PID controller (without anti-saturation design) is similar to that of the HSO / BC-AW controller, in-depth analysis reveals that the former still exhibits a certain degree of steady-state error. While both the ZN / PID and HSO / PID controllers include integral terms, the persistent steady-state error stems from different intrinsic mechanisms. For the ZN / PID controller, this error is an inherent instability; severe integral saturation leads to continuous oscillations, preventing the system from reaching a true steady state. Therefore, the recorded error is only an instantaneous value within a continuous oscillation period. In contrast, the minimum steady-state error of the HSO / PID controller originates from a proactive performance trade-off in the optimization algorithm. To comprehensively reduce the ITAE performance index, the HSO algorithm converges to the optimal parameter set containing a minimal integral gain. This strategic compromise prioritizes transient performance and stability by deliberately limiting integral action to suppress overshoot, allowing for a small non-zero steady-state error. This limitation of the standard PID structure highlights the key innovative value of the proposed method. By integrating the BC-AW mechanism, the HSO / BC-AW controller dynamically adjusts the integral term during the saturation phase. This method successfully eliminates the aforementioned performance trade-offs, achieving the dual objectives of excellent transient performance and zero steady-state error. Compared to ZN / PID, HSO / PID reduces ITAE by 73.92%, while HSO / BCAW further reduces it by 10.80% compared to HSO / PID.

[0047] To evaluate the robustness of the two optimally tuned controllers, 1Hz sinusoidal input signals with amplitudes of 100%, 75%, 50%, and 25% were used instead of the step reference signal. This method provides a more comprehensive evaluation of the controller performance under continuously varying demands, contrasting with the transient response characteristics captured by the step input. The controller performance was benchmarked against a ZN-tuned PID controller under sinusoidal excitation; the comparison results are detailed in [link to relevant documentation]. Figure 7 , Figure 8 , Figure 9 , Figure 10 To quantify tracking accuracy, the mean absolute error (MAE) is used as the evaluation criterion, as detailed in Table 4.

[0048] The calculated MAE values, quantifying the performance improvement, were obtained under sinusoidal inputs at 1 Hz and varying amplitudes (25% to 100%), as shown in Table 4. First, the HSO / PID controller significantly outperformed the ZN / PID controller in all cases, reducing MAE by 38.29%–50.93%. This confirms that HSO-based optimization substantially enhances robust tracking performance. Second, the HSO / BCAW controller, by introducing the BCAW mechanism, achieved further improvements, further reducing MAE by 9.58%–38.19% compared to HSO / PID. This indicates higher accuracy and robustness under continuously changing operating conditions.

[0049] Table 4 The embodiment also provides a comparison chart of dead zone compensation effects. Figure 11 As can be seen from the magnified image near the zero position, the dead zone hysteresis phenomenon near the zero position is significantly improved after integrating the single-sided threshold current superposition compensation.

[0050] The specific embodiments described above illustrate the technical solution and beneficial effects of the present invention in detail. It should be understood that the above description is only the most preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, additions, and equivalent substitutions made within the scope of the principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A multi-way valve cooperative control system based on group optimization and anti-saturation, characterized in that, include: The global optimization layer uses control parameters such as PID gain parameters and anti-saturation gain parameters as individuals, and employs a global swarm optimization algorithm to iteratively optimize and output the optimal control parameters using the time-weighted absolute error integral as the fitness function. The dynamic compensation control layer is used to output the theoretical and actual control quantities of the current according to the standard PID mode based on the PID gain parameters when the linear operation is in a non-saturated state; and to output the theoretical and actual control quantities of the current according to the anti-saturation compensation PID mode based on the PID gain parameters and anti-saturation gain parameters when the operation is in a saturated state. The logic driving layer is used to convert the actual control quantity output by the dynamic compensation control layer into the actual current driving the two electromagnet coils of the proportional valve and control the actual displacement of the valve core.

2. The multi-channel valve cooperative control system based on group optimization and anti-saturation as described in claim 1, characterized in that, The absolute error is the absolute value of the error between the valve core displacement command reference value and the actual normalized displacement feedback value.

3. The multi-channel valve cooperative control system based on group optimization and anti-saturation as described in claim 2, characterized in that, The fitness function is expressed as follows: in, Represents the fitness function. and These represent the initial and final times of the integration process, respectively. Indicates the current time. Indicates time The absolute error at that time.

4. The multi-channel valve collaborative control system based on group optimization and anti-saturation as described in claim 3, characterized in that, initial time and termination time Set them to 0.05 seconds and 0.25 seconds respectively.

5. The multi-channel valve cooperative control system based on group optimization and anti-saturation as described in claim 1, characterized in that, The theoretical and actual control quantities of the output current based on the PID gain parameters and anti-saturation gain parameters in PID mode with anti-saturation compensation include: in, , , ,as well as These represent proportional gain, integral gain, derivative gain, and anti-saturation gain, respectively. Indicates time The error between the valve core displacement command reference value and the actual normalized displacement feedback value. and These represent the theoretical control quantity and the actual control quantity, respectively. This indicates the actuator's maximum rated saturation current.

6. The multi-channel valve cooperative control system based on group optimization and anti-saturation as described in claim 1, characterized in that, The actual control quantity is converted into the actual current driving the two electromagnet coils of the proportional valve, including: in, It is the current in the coil of proportional electromagnet 1. This refers to the current in coil 2 of the proportional electromagnet; the maximum rated current is... , , The initial current, which serves as the threshold current, represents the current value required for a single proportional electromagnet to output force just enough to overcome the pre-compression force of the slide valve spring. This represents the actual control quantity of the current.

7. The multi-channel valve cooperative control system based on group optimization and anti-saturation as described in claim 6, characterized in that, The threshold current is determined in the following way: An open-loop control strategy is adopted, and the threshold current for skipping the dead zone is calibrated according to the one-sided control method. The current value required to make the output force of a single proportional electromagnet just overcome the nonlinear factors and generate a displacement value is found.

8. A multi-way valve cooperative control method based on group optimization and anti-saturation, characterized in that, The method employs the system described in any one of claims 1-7 and includes the following steps: Using a global optimization layer, the PID gain parameter and anti-saturation gain parameter are treated as individuals. The time-weighted absolute error integral is used as the fitness function. The global swarm optimization algorithm is used to iteratively optimize and output the optimal control parameters. Feedback anti-saturation control is performed using a dynamic compensation control layer. When the system is in a non-saturated linear operating state, the theoretical and actual control quantities of the output current are based on the PID gain parameters and follow the standard PID mode. When the system is in a saturated state, the theoretical and actual control quantities of the output current are based on the PID gain parameters and anti-saturation gain parameters and follow the anti-saturation compensation PID mode. The logic driving layer is used to convert the actual control quantity output by the dynamic compensation control layer into the actual current driving the two electromagnet coils of the proportional valve and control the actual displacement of the valve core.

Images