Self-adaptive fuzzy PID flow field control method and device

Abstract

The invention discloses a self-adaptive fuzzy PID flow field control method and device, and the method comprises the steps: generating an initial PID gain parameter in real time through a fuzzy reasoning module, introducing an RBF neural network module to carry out the dynamic optimization compensation of the initial parameter, and generating a correction gain. And a final PID parameter is dynamically synthesized by adopting a weighted fusion formula, and a weight coefficient is adjustable so as to balance the contribution of fuzzy rules and neural network learning. The nonlinear approximation capability of the RBF network is utilized, network parameters are updated in real time through a gradient descent method, a Jacobian matrix of a controlled system is output, and the sensitivity of the control quantity to input changes is accurately recognized. The parameter updating rate is dynamically adjusted through a momentum item and an exponential decay function in combination with environmental sensor data, flow feedback and historical deviation, and a closed-loop control loop of monitoring, fuzzy reasoning, RBF optimization, PID output and feedback is formed. The accuracy of flow field control is improved.

Description

Technical Field

[0001] The embodiments of the present application relate to flow field control technology, and in particular to an adaptive fuzzy PID flow field control method and device. Background Art

[0002] Traditional fuzzy PID (proportional, integral, derivative) controllers face significant limitations in complex industrial scenarios such as port water quality control: they rely on a preset fuzzy rule base (such as a seven-level classification system including NB, NM, NS, ZO, PS, PM, and PB) to output PID gain parameters. When sudden environmental disturbances cause the control deviation to exceed the fuzzy domain, the rule base cannot provide valid parameters. At the same time, the introduction of nonlinear characteristics of actuators such as control valves further leads to steady-state errors, while fixed-parameter PID lacks dynamic adaptability. Improvements to traditional PID control generally improve the dynamic and static performance of the system by adjusting the three parameters of the controller. However, in actual application scenarios, control accuracy decreases due to external environmental interference factors and changes in the internal performance parameters of the control valve. PID controllers also have weak anti-interference capabilities and are not very applicable in nonlinear systems.

[0003] In the field of regulating valve control, traditional PID control algorithms, while widely used, present three problems when regulating large-diameter valves: First, insufficient stability, resulting in overshoot and system oscillation when flow rates change suddenly; second, dynamic response hysteresis, making it difficult for fixed parameters to adapt to real-time load changes; and third, degraded steady-state accuracy, resulting in accumulated flow deviations and reduced accuracy over long periods of operation. While existing improvement schemes can partially optimize performance, they still suffer from a strong reliance on manual experience, delayed parameter adjustment responses, and a lack of real-time self-learning mechanisms. Summary of the Invention

[0004] The present application provides an adaptive fuzzy PID flow field control method and device to at least solve the above technical problems existing in the prior art.

[0005] According to a first aspect of the present application, there is provided an adaptive fuzzy PID flow field control method, comprising: Collecting water quality index data in a target water area and determining the deviation and rate of change of the deviation between the water quality index data and a target value; the water quality index data including at least inorganic nitrogen concentration and active phosphate concentration; inputting the deviation and the rate of change of the deviation into a fuzzy inference module, performing fuzzification processing through a seven-level fuzzy classification system, and generating an initial PID gain parameter based on a preset fuzzy rule base; The initial PID gain parameters and real-time feedback values ​​are input into a radial basis function network RBF, the Jacobian matrix is ​​calculated by the radial basis function, and the network parameters are updated online using the gradient descent method; Through the adaptive parameter fusion mechanism, the initial PID gain parameters output by the fuzzy inference module and the correction parameters output by the RBF are weighted and fused to obtain the final PID parameters; A control signal is generated according to the final PID parameters to drive the actuator to adjust the port flow field to form a closed-loop control circuit.

[0006] In some optional embodiments, the method further includes: The rules in the fuzzy rule base are generated in the form of conditional statements, and an intelligent environment perception system is established to form a fuzzy rule base; Calculate the correction of proportional, integral and differential coefficients in real time according to the rules in the fuzzy rule base 、 、 , the exponential decay function is used to dynamically adjust the correction coefficient, and its mathematical expression is: ,in, 、 、 Represent the correction coefficients of the proportional term, integral term and differential term respectively, is the exponential decay function value, is the initial value of each correction coefficient, represents the decay rate constant, represents the base of natural logarithms, is the time elapsed from the initial moment; Proportional coefficient The value of is determined by the coefficient of the previous moment With the current adjustment Determine together, the integral coefficient The value is determined by the integration coefficient of the previous moment With the current adjustment Determine together, the differential coefficient value By the differential coefficient of the previous moment With the current adjustment jointly determined; among which, 、 、 Respectively expressed in The initial setting value of each parameter at the moment, 、 、 It is the parameter adjustment amount determined based on the fuzzy rule base.

[0007] In some optional embodiments, the calculating of the Jacobian matrix by using the radial basis function and the online updating of the network parameters by using the gradient descent method include: The Gaussian function is selected as the specific implementation form of the radial basis function, and its mathematical expression is as follows:

[0008] in, For the The basis width vector of nodes, Indicates the The center vector of the node, =1,2,…m; Calculate the neural network output according to the weight coefficient of the RBF neural network and network identification performance indicator function ,as follows:

[0009]

[0010] in, Indicates the measured value of water quality index in target water area, is the network weight coefficient from the hidden layer to the output layer, is the hidden layer The output of a neuron; Update the node center vector by iteratively , basis width vector , network weight coefficient and the Jacobian matrix Identify sensitivity information, update network identification indicators, and obtain the optimal value of network identification indicators.

[0011] In some optional embodiments, the method further includes: During the network parameter update process, the following operations are performed in sequence at each time step t: , actual system output With model output The difference between the The output value of a neuron , calculate the weight parameter The amount of change ; Update weight parameters , through the adder Weight coefficient of time and Add together and incorporate the momentum term , multiplied by the weight coefficient through the multiplier exist Moment and The difference between the time and the weight coefficient Make updates; Based on the network learning rate , error difference , No. The output value of a neuron , updated weight coefficient , the input vector calculated by the dot product With the center vector The square of the Euclidean distance , and the modulus of the basis width vector , calculate the basis width vector The amount of change ; Update the base width vector When Basis width vector at time and Added by adder and introduced momentum term Multiply the base width vector by the multiplier exist Moment and The difference between the two moments is used as the updated weight coefficient.

[0012] In some optional embodiments, the method further includes: Update the calculated center vector Each component , according to the network learning efficiency , error difference , updated weight coefficient value , input vector and the center vector component The difference between the values ​​of , calculate the change of the center vector component through the divider and multiplier ; Update the center vector components When The central vector component at time and Add together and add momentum , multiplied by the center vector component exist Moment and The difference in time; The adder accumulates the and , center vector component The control change of the PID output to the regulating valve The difference is the same The product of the ratios of .

[0013] In some optional embodiments, generating a control signal according to the final PID parameter includes: PID parameter increment output by RBF neural network 、 、 Determined by the following formula:

[0014] in, 、 、 Respectively represent the PID gain learning rate; Generating a control signal according to the final PID parameter includes: The control signal is determined by the following formula:

[0015] Among them, the control signal The calculation involves the following parameters: time point The controller output value , time point The deviation change , and time points The error second-order backward difference , the second derivative of the approximation error, time point The deviation change ; Calculating the PID controller output value During the process, the time point in the historical deviation data The deviation change and time point The deviation change Used to predict changing trends of water quality indicators.

[0016] According to a second aspect of the present application, there is provided an adaptive fuzzy PID flow field control device, comprising: a calculation unit for collecting water quality index data in a target water area and determining a deviation and a rate of change of the deviation between the water quality index data and a target value; wherein the water quality index data includes at least an inorganic nitrogen concentration and an active phosphate concentration; and a fuzzy processing unit for inputting the deviation and the rate of change of the deviation into a fuzzy inference module, performing fuzzification processing through a seven-level fuzzy classification system, and generating an initial PID gain parameter based on a preset fuzzy rule base. An updating unit is used to input the initial PID gain parameters and real-time feedback values ​​into a radial basis function network RBF, calculate the Jacobian matrix through the radial basis function, and update the network parameters online using a gradient descent method; The fusion unit is used to perform weighted fusion of the initial PID gain parameters output by the fuzzy inference module and the correction parameters output by the RBF through an adaptive parameter fusion mechanism to obtain the final PID parameters; The driving unit is used to generate a control signal according to the final PID parameters and drive the actuator to adjust the port flow field to form a closed-loop control circuit.

[0017] In some optional embodiments, the device further comprises: The generation unit generates the rules in the fuzzy rule base in the form of conditional statements, and establishes an intelligent environment perception system to form a fuzzy rule base; The adjustment unit is used to calculate the correction amount of proportional, integral and differential coefficients in real time according to the rules in the fuzzy rule base 、 、 , the exponential decay function is used to dynamically adjust the correction coefficient, and its mathematical expression is: ,in, 、 、 Represent the correction coefficients of the proportional term, integral term and differential term respectively, is the exponential decay function value, is the initial value of each correction coefficient, represents the decay rate constant, represents the base of natural logarithms, is the time elapsed from the initial moment; the proportional coefficient The value of is determined by the coefficient of the previous moment With the current adjustment Determine together, the integral coefficient The value is determined by the integration coefficient of the previous moment With the current adjustment Determine together, the differential coefficient value By the differential coefficient of the previous moment With the current adjustment jointly determined; among which, 、 、 Respectively expressed in The initial setting value of each parameter at the moment, 、 、 It is the parameter adjustment amount determined based on the fuzzy rule base.

[0018] In some optional embodiments, the updating unit is further configured to: The Gaussian function is selected as the specific implementation form of the radial basis function, and its mathematical expression is as follows:

[0019] in, For the The basis width vector of nodes, Indicates the The center vector of the node, =1,2,…m; Calculate the neural network output according to the weight coefficient of the RBF neural network and network identification performance indicator function ,as follows:

[0020]

[0021] in, Indicates the measured value of water quality index in target water area, is the network weight coefficient from the hidden layer to the output layer, is the hidden layer The output of a neuron; Update the node center vector by iteratively , basis width vector , network weight coefficient and the Jacobian matrix Identify sensitivity information, update network identification indicators, and obtain the optimal value of network identification indicators.

[0022] In some optional embodiments, the updating unit is further configured to: During the network parameter update process, the following operations are performed in sequence at each time step t: , actual system output With model output The difference between the The output value of a neuron , calculate the weight parameter The amount of change ; Update weight parameters , through the adder Weight coefficient of time and Add together and incorporate the momentum term , multiplied by the weight coefficient through the multiplier exist Moment and The difference between the time and the weight coefficient Make updates; Based on the network learning rate , error difference , No. The output value of a neuron , updated weight coefficient , the input vector calculated by the dot product With the center vector The square of the Euclidean distance , and the modulus of the basis width vector , calculate the basis width vector The amount of change ; Update the base width vector When Basis width vector at time and Added by adder and introduced momentum term Multiply the base width vector by the multiplier exist Moment and The difference between the two moments is used as the updated weight coefficient.

[0023] According to a third aspect of the present application, an electronic device is provided, including: at least one processor; and a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the steps of the adaptive fuzzy PID flow field control method described in the present application.

[0024] According to a fourth aspect of the present application, a non-transitory computer-readable storage medium storing computer instructions is provided, wherein the computer instructions are used to enable the computer to execute the steps of the adaptive fuzzy PID flow field control method described in the present application.

[0025] The technical solution of the present application generates initial PID gain parameters in real time and introduces an RBF neural network to dynamically optimize and compensate the initial parameters, thereby improving the accuracy of the PID gain parameters and the control accuracy of the control signal. The weight coefficients corresponding to the final PID parameters are adjustable, thereby balancing the contributions of fuzzy rules and neural network learning. Utilizing the nonlinear approximation capability of the RBF network, the network parameters are updated in real time through the gradient descent method, achieving accurate identification of the sensitivity of the control quantity to input changes. The parameter update rate is dynamically adjusted through the momentum term and exponential decay function to form a closed-loop control loop, ensuring that the flow field control always complies with the dynamic PID parameters.

[0026] It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present application, nor is it intended to limit the scope of the present application. Other features of the present application will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] The above and other objects, features and advantages of the present application will become more apparent from the following detailed description read in conjunction with the accompanying drawings, in which: In the drawings, identical or corresponding components are denoted by identical or corresponding reference numerals.

[0028] Figure 1 A flowchart of an adaptive fuzzy PID flow field control method of an embodiment of the present application is shown; Figure 2 A schematic diagram of dynamic adjustment of a correction coefficient using an exponential decay function according to an embodiment of the present application is shown; Figure 3 A schematic diagram of online updating of network parameters using a gradient descent method according to an embodiment of the present application is shown; Figure 4 A schematic diagram of real-time determination of a deviation between water quality index data and a target value and a deviation change rate according to an embodiment of the present application is shown; Figure 5 A schematic diagram of overall closed-loop control according to PID parameters according to an embodiment of the present application is shown; Figure 6 A schematic diagram of a composition structure of an adaptive fuzzy PID flow field control device according to an embodiment of the present application is shown; Figure 7 A schematic diagram of a composition structure of an electronic device according to an embodiment of the present application is shown. DETAILED DESCRIPTION

[0029] In order to make the objectives, features and advantages of the present application more apparent and easy to understand, the technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only some of the embodiments of the present application, but not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work fall within the scope of protection of the present application.

[0030] Figure 1 A flowchart of an adaptive fuzzy PID flow field control method of an embodiment of the present application is shown, as shown in Figure 1 The adaptive fuzzy PID flow field control method according to an embodiment of the present application includes the following processing steps: Step 101, water quality index data in a target water area is collected, and a deviation between the water quality index data and a target value and a deviation change rate are determined.

[0031] In this embodiment, real-time data collection of key water quality indicators, such as inorganic nitrogen and active phosphate, is required for the target port waters. The deviation between the measured value and the target value, as well as the rate of change, is calculated and then fed into a fuzzy inference module based on the intelligent environmental perception system. In step 102, the deviation and rate of change are fed into the fuzzy inference module, which then performs fuzzification processing using a seven-level fuzzy classification system. Initial PID gain parameters are generated based on a pre-set fuzzy rule base.

[0032] The seven-level fuzzy classification system of the embodiment of the present application includes NB, NM, NS, ZO, PS, PM, PB, etc., which quantitatively evaluates the deviation between the measured value and the target value and dynamically outputs the initial gain parameters of the PID controller according to the preset rule base. 、 、 .

[0033] Step 103: input the initial PID gain parameters and the real-time feedback value into the radial basis function network RBF, calculate the Jacobian matrix through the radial basis function, and use the gradient descent method to update the network parameters online.

[0034] In this embodiment of the present application, the initial control amount The real-time feedback value is input into the RBF neural network, and with the help of the nonlinear approximation ability and autonomous learning characteristics of the RBF neural network, the Jacobian matrix of the sensitivity information of the controlled system control quantity to the change of the controller input is output.

[0035] Step 104 : Through the adaptive parameter fusion mechanism, the initial PID gain parameters output by the fuzzy inference module and the correction parameters output by the RBF are weightedly fused to obtain the final PID parameters.

[0036] The embodiment of the present application optimizes and compensates the PID parameters through a self-learning mechanism to generate a correction gain 、 、 , thereby achieving dynamic adjustment of control parameters and continuous optimization of system performance.

[0037] Step 105 : generating a control signal according to the final PID parameters to drive the actuator to adjust the port flow field to form a closed-loop control circuit.

[0038] The following is a detailed explanation of the technical solution of the embodiment of the present application.

[0039] Figure 2 FIG. 4 shows a schematic diagram of dynamically adjusting the correction coefficient using an exponential decay function according to an embodiment of the present application. Figure 2 As shown, the PID flow field control of the embodiment of the present application uses an exponential decay function to dynamically adjust the correction coefficient, and its mathematical expression is: ,in 、 、 Represent the correction coefficients of the proportional term, integral term and differential term respectively, is the exponential decay function value, is the initial value of each correction coefficient, represents the decay rate constant, represents the base of natural logarithms, The time from the initial moment is collected in real time to collect key water quality indicators such as inorganic nitrogen and active phosphate in the target port waters. , water quality index setting value , generating deviation values and deviation change value The fuzzy rule base used in the embodiment of the present application is established based on the intelligent environment perception system, and its rules are in the form of conditional statements: If is…and is…, then is…, is…, The system calculates the correction amount of proportional, integral and differential coefficients in real time according to these rules in the fuzzy rule base. 、 、 , and dynamically update the PID control parameters.

[0040] During the parameter update process, the proportional coefficient The value is determined by the coefficient of the previous moment With the current adjustment Determine together, the integral coefficient The value is determined by the integration coefficient of the previous moment With the current adjustment Determine together, the differential coefficient value Also from the differential coefficient of the previous moment With the current adjustment Jointly determined. Among them, 、 、 Respectively indicate that the controller is The initial setting value of each parameter at the moment, 、 、 It is the parameter adjustment amount calculated by the fuzzy rule base.

[0041] The embodiment of the present application adopts a three-layer feedforward RBF neural network, the input layer of the RBF neural network It is composed of signal source nodes and directly receives external input parameters. The hidden layer is constructed by the nonlinear radial basis function of the network weight coefficient. The input parameters are spatially transformed, and the output layer is transformed by the network weight coefficient The output of the hidden layer A linear weighted summation is performed, and finally efficient function approximation is achieved through the nonlinear mapping of the hidden layer and the linear combination of the output layer.

[0042] The RBF neural network of the embodiment of the present application first performs a nonlinear feature conversion on the input data. Its core mechanism is to achieve nonlinear mapping of the input data through the radial basis function. In the RBF neural network structure, each hidden layer node calculates the distance between the input vector and the center of the node, and uses the distance value as the output of the node. Three key parameters need to be solved during the RBF network training process: the center position of the radial basis function, the basis function width parameter, and the connection weight from the hidden layer to the output layer. The embodiment of the present application uses the Gaussian function as the specific implementation form of the radial basis function, and its mathematical expression is as follows: (1.1) in, For the The basis width vector of nodes, Indicates the The center vector of the node, =1,2,…m. According to the weight coefficient of RBF neural network, the output of neural network can be calculated and network identification performance indicator function ,as follows: (1.2) (1.3) in, Indicates the measured value of water quality index, is the network weight coefficient from the hidden layer to the output layer, is the hidden layer In order to achieve the best control effect, the embodiment of the present application adopts the gradient descent method to update the parameters. Specifically, the performance index searches for the optimal solution along the negative gradient direction, and updates the node center vector by iterative , basis width vector , network weight coefficient and the Jacobian matrix Identify sensitivity information, update network identification indicators, and obtain the optimal value of network identification indicators.

[0043] Figure 3 FIG. 1 shows a schematic diagram of an embodiment of the present application using a gradient descent method to update network parameters online, as shown in FIG. Figure 3As shown, during the parameter update process, the following operations are performed in sequence at each time step t: First, according to the network learning rate , actual system output With model output The difference between the The output value of a neuron , calculate the weight parameter The amount of change . Then update the weight parameters , through the adder Weight coefficient of time and Add and introduce momentum term , multiplied by the weight coefficient through the multiplier exist Moment and The difference between the moments, so as to achieve the weight coefficient Updates.

[0044] Then based on the network learning rate , error difference , No. The output value of a neuron , updated weight coefficient , the input vector obtained by the dot product With the center vector The square of the Euclidean distance , and the modulus of the basis width vector , calculate the basis width vector The amount of change . Update the base width vector When Basis width vector at time and Added by adder and introduced momentum term , multiplied by the basis width vector through the multiplier exist Moment and The difference between the moments, so as to achieve the weight coefficient Updates.

[0045] The embodiment of the present application also updates the calculation center vector Each component , according to the network learning efficiency , error difference , updated weight coefficient value , input vector and the center vector component The difference between the values ​​of , calculate the change of the center vector component through the divider and multiplier . In updating the center vector component When The central vector component at time and Add together and add momentum , multiplied by the center vector component through the multiplier exist Moment and Finally, the adder accumulates the values ​​of all neurons. and , center vector component The control change of the PID output to the regulating valve The difference is the same The product of the ratios of The entire calculation process is done through the momentum term The historical parameter change trends are continuously introduced to optimize the convergence characteristics of the parameters.

[0046] Figure 4 A schematic diagram of determining the deviation between water quality index data and target value and the deviation change rate in real time according to an embodiment of the present application is shown. Figure 4 As shown, the embodiment of the present application calculates the deviation at the current moment based on the difference between the water quality index monitored in real time by the environmental data monitoring module and the set target water quality index. The deviation is input into the fuzzy control system, and after fuzzification processing and fuzzy reasoning module, the fuzzy controller gain of the initial stage is output. 、 、 Subsequently, the control quantity generated by the initial PID controller gain is And the actual feedback value The RBF neural network is inputted together. Utilizing the RBF's approximation ability for nonlinear systems and its self-learning characteristics, the neural network is able to learn the dynamic mapping relationship between the system's control variable and the controller input, that is, the Jacobian matrix of the sensitivity of the controlled system's output control variable to changes in the controller input.

[0047] Finally, the PID parameters of the input controller are optimized and fitted through the self-learning module. The gradient descent algorithm is used to adjust the neural network parameters to obtain the PID parameter increment output by the RBF neural network. 、 、 , as shown in Equation 1.4 below, where 、 、 Represents the PID gain learning rate respectively. Therefore, by combining the weights of the fuzzy controller and the RBF neural network, the adjustment increment of the PID control parameters can be calculated as 、 、 .in, , and The weight coefficients reflect the contributions of the fuzzy control and neural network to the parameter increment. This approach enables the system to adaptively adjust PID parameter gains in real time, effectively suppressing interference from environmental disturbances and forming a closed-loop control loop. This strategy significantly improves the system's control accuracy and dynamic response performance. The system's control signal is given by Equation 1.5.

[0048] (1.4) (1.5) As shown in the above formula 1.5, in the control system, the output value of the PID controller is ,in Represents a discrete time point. Control signal The calculation process involves the following parameters: time point The controller output value , time point The deviation change , and time points The error of the second-order backward difference , the second derivative of the approximation error, time point The deviation change The calculation of the PID controller output value During the process, the time point in the historical deviation data The deviation change and time point The deviation change It is used to predict the changing trend of water quality indicators and dynamically optimize the rule weights through the sliding window algorithm.

[0049] Figure 5 The overall schematic diagram of the closed-loop control according to the PID parameters of the embodiment of the present application is shown as follows: Figure 5As shown, the embodiment of the present application first collects the environmental parameters of the port waters through environmental data sensors, and then inputs these parameters together with the real-time flow and flow rate data obtained by the flow and velocity sensors into the core control component - the adaptive fuzzy control PID controller based on neural networks. This intelligent controller dynamically generates accurate control signals and outputs them to the regulating valve by integrating the self-learning ability of the neural network and the adaptability of fuzzy control, and realizes closed-loop control of the flow by changing the opening of the regulating valve. The flow and velocity sensors continuously feed back the latest data under the influence of the controlled flow, forming a two-way data interaction with the controller, thereby forming a complete closed-loop control system with environmental parameter perception, intelligent algorithm decision-making, and real-time feedback regulation.

[0050] Figure 6 FIG. 1 shows a schematic diagram of the structure of the adaptive fuzzy PID flow field control device according to an embodiment of the present application. Figure 6 As shown, the adaptive fuzzy PID flow field control device of the embodiment of the present application includes: A calculation unit 60 is configured to collect water quality index data in a target water area and determine the deviation and rate of change of the deviation between the water quality index data and a target value; wherein the water quality index data includes at least inorganic nitrogen concentration and active phosphate concentration; a fuzzy processing unit 61 is configured to input the deviation and rate of change of the deviation into a fuzzy inference module, perform fuzzification processing using a seven-level fuzzy classification system, and generate initial PID gain parameters based on a preset fuzzy rule base; An updating unit 62 is configured to input the initial PID gain parameters and the real-time feedback value into a radial basis function network RBF, calculate the Jacobian matrix by the radial basis function, and update the network parameters online using a gradient descent method; The fusion unit 63 is used to perform weighted fusion of the initial PID gain parameters output by the fuzzy inference module and the correction parameters output by the RBF through an adaptive parameter fusion mechanism to obtain the final PID parameters; The driving unit 64 is used to generate a control signal according to the final PID parameters to drive the actuator to adjust the port flow field to form a closed-loop control circuit.

[0051] middle Figure 6 Based on the adaptive fuzzy PID flow field control device shown in the figure, the adaptive fuzzy PID flow field control device of the embodiment of the present application further includes: Generate unit ( Figure 6 (not shown), the rules in the fuzzy rule base are generated in the form of conditional statements, and an intelligent environment perception system is established to form a fuzzy rule base; Adjustment unit ( Figure 6 (not shown) for calculating the correction amount of proportional, integral and differential coefficients in real time according to the rules in the fuzzy rule base 、 、 , the exponential decay function is used to dynamically adjust the correction coefficient, and its mathematical expression is: ,in, 、 、 Represent the correction coefficients of the proportional term, integral term and differential term respectively, is the exponential decay function value, is the initial value of each correction coefficient, represents the decay rate constant, represents the base of natural logarithms, is the time elapsed from the initial moment; the proportional coefficient The value of is determined by the coefficient of the previous moment With the current adjustment Determine together, the integral coefficient The value is determined by the integration coefficient of the previous moment With the current adjustment Determine together, the differential coefficient value By the differential coefficient of the previous moment With the current adjustment jointly determined; among which, 、 、 Respectively expressed in The initial setting value of each parameter at the moment, 、 、 It is the parameter adjustment amount determined based on the fuzzy rule base.

[0052] In some optional embodiments, the updating unit 62 is further configured to: The Gaussian function is selected as the specific implementation form of the radial basis function, and its mathematical expression is as follows:

[0053] in, For the The basis width vector of nodes, Indicates the The center vector of the node, =1,2,…m; Calculate the neural network output according to the weight coefficient of the RBF neural network and network identification performance indicator function ,as follows:

[0054]

[0055] in, Indicates the measured value of water quality index in target water area, is the network weight coefficient from the hidden layer to the output layer, is the hidden layer The output of a neuron; Update the node center vector by iteratively , basis width vector , network weight coefficient and the Jacobian matrix Identify sensitivity information, update network identification indicators, and obtain the optimal value of network identification indicators.

[0056] In some optional embodiments, the updating unit 62 is further configured to: During the network parameter update process, the following operations are performed in sequence at each time step t: , actual system output With model output The difference between the The output value of a neuron , calculate the weight parameter The amount of change ; Update weight parameters , through the adder Weight coefficient of time and Add together and incorporate the momentum term , multiplied by the weight coefficient through the multiplier exist Moment and The difference between the time and the weight coefficient Make updates; Based on the network learning rate , error difference , No. The output value of a neuron , updated weight coefficient , the input vector calculated by the dot product With the center vector The square of the Euclidean distance , and the modulus of the basis width vector , calculate the basis width vector The amount of change ; Update the base width vector When Basis width vector at time and Added by adder and introduced momentum term Multiply the base width vector by the multiplier exist Moment and The difference between the two moments is used as the updated weight coefficient.

[0057] Update the calculated center vector Each component , according to the network learning efficiency , error difference , updated weight coefficient value , input vector and the center vector component The difference between the values ​​of , calculate the change of the center vector component through the divider and multiplier ; Update the center vector components When The central vector component at time and Add together and add momentum , multiplied by the center vector component exist Moment and The difference in time; The adder accumulates the and , center vector component The control change of the PID output to the regulating valve The difference is the same The product of the ratios of .

[0058] PID parameter increment output by RBF neural network 、 、 Determined by the following formula:

[0059] in, 、 、 Respectively represent the PID gain learning rate; Generating a control signal according to the final PID parameter includes: The control signal is determined by the following formula:

[0060] Among them, the control signal The calculation involves the following parameters: time point The controller output value , time point The deviation change , and time points The error of the second-order backward difference , the second derivative of the approximation error, time point The deviation change ; Calculating the PID controller output value During the process, the time point in the historical deviation data The deviation change and time point The deviation change Used to predict changing trends of water quality indicators.

[0061] In an exemplary embodiment, each processing unit in the adaptive fuzzy PID flow field control device of the embodiment of the present application can be implemented by one or more central processing units (CPU), graphics processing units (GPU), application specific integrated circuits (ASIC), DSP, programmable logic devices (PLD), complex programmable logic devices (CPLD), field programmable gate arrays (FPGA), general-purpose processors, controllers, microcontrollers (MCU), microprocessors, or other electronic components.

[0062] Regarding the device in the above embodiment, the specific manner in which each module and unit performs operations has been described in detail in the embodiment of the method, and will not be elaborated here.

[0063] According to an embodiment of the present application, the present application also describes an electronic device and a readable storage medium.

[0064] Figure 7A schematic block diagram of an example electronic device 800 that can be used to implement embodiments of the present application is shown. The electronic device 800 is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device 800 can also represent various forms of mobile devices, such as personal digital assistants, cellular phones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present application described and / or claimed herein.

[0065] like Figure 7 As shown, electronic device 800 includes a computing unit 801, which can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) 802 or a computer program loaded from a storage unit 808 into a random access memory (RAM) 803. RAM 803 may also store various programs and data required for the operation of electronic device 800. Computing unit 801, ROM 802, and RAM 803 are interconnected via a bus 804. An input / output (I / O) interface 805 is also connected to bus 804.

[0066] Multiple components in the electronic device 800 are connected to the I / O interface 805, including an input unit 806, such as a keyboard, a mouse, etc.; an output unit 807, such as various types of displays, speakers, etc.; a storage unit 808, such as a magnetic disk, an optical disk, etc.; and a communication unit 809, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 809 allows the electronic device 800 to exchange information / data with other devices via a computer network such as the Internet and / or various telecommunication networks.

[0067] The computing unit 801 can be any general-purpose and / or specialized processing component with processing and computing capabilities. Some examples of the computing unit 801 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various specialized artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. The computing unit 801 performs the various methods and processes described above, such as the adaptive fuzzy PID flow field control method. For example, in some embodiments, the adaptive fuzzy PID flow field control method can be implemented as a computer software program tangibly embodied in a machine-readable medium, such as the storage unit 808. In some embodiments, part or all of the computer program can be loaded and / or installed onto the electronic device 800 via the ROM 802 and / or the communication unit 809. When the computer program is loaded into the RAM 803 and executed by the computing unit 801, one or more steps of the adaptive fuzzy PID flow field control method described above can be performed. Alternatively, in other embodiments, the computing unit 801 may be configured to execute the adaptive fuzzy PID flow field control method in any other appropriate manner (for example, by means of firmware).

[0068] Various implementations of the systems and techniques described above in this specification can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on a chip (SOCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include being implemented in one or more computer programs that are executable and / or interpreted on a programmable system that includes at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0069] The program code for implementing the methods of the present application can be written in any combination of one or more programming languages. Such program code can be provided to a processor or controller of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that when the program code is executed by the processor or controller, the functions / operations specified in the flow charts and / or block diagrams are implemented. The program code can be executed entirely on the machine, partially on the machine, as a stand-alone software package, partially on the machine and partially on a remote machine, or entirely on a remote machine or server.

[0070] In the context of this application, a machine-readable medium can be a tangible medium that contains or stores a program for use by or in connection with an instruction execution system, apparatus, or electronic device. The machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can include but is not limited to an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or electronic device, or any suitable combination of the foregoing. More specific examples of the machine-readable storage medium will include one or more lines of electrical connections, portable computer disks, hard disk drives, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or Flash memory), optical fibers, portable compact disc read-only memories (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0071] To provide for interaction with a user, the systems and techniques described here can be implemented on a computer having a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user and a keyboard and a pointing device (e.g., a mouse or a trackball) by which the user can provide input to the computer. Other kinds of devices can be used to provide for interaction with a user as well; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form, including acoustic, speech, or tactile input.

[0072] The systems and techniques described here can be implemented in a computing system that includes a back end component (e.g., as a data server), or that includes a middleware component (e.g., an application server), or that includes a front end component (e.g., a user computer having a graphical user interface or a Web browser through which a user can interact with an implementation of the systems and techniques described here), or any combination of such back end, middleware, or front end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include a local area network (LAN), a wide area network (WAN), and the Internet.

[0073] A computer system may include a client and a server. The client and server are generally remote from each other and typically interact through a communication network. The client-server relationship arises through computer programs running on the respective computers and having a client-server relationship with each other. The server may be a cloud server, a server in a distributed system, or a server integrated with a blockchain.

[0074] It should be understood that the various forms of the processes shown above can be used to reorder, add, or delete steps. For example, the steps described in this disclosure can be performed in parallel, sequentially, or in a different order, as long as the desired results of the technical solutions disclosed in this application can be achieved. This is not a limitation herein.

[0075] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features being referred to. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one such feature. Throughout the description of this application, "plurality" means two or more, unless otherwise specifically defined.

[0076] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of this application. Therefore, the scope of protection of this application should be based on the scope of protection of the claims.

Claims

1. An adaptive fuzzy PID flow field control method, characterized in that: The method comprises: Collecting water quality index data in a target water area, and determining a deviation and a rate of change of the deviation between the water quality index data and a target value; the water quality index data including at least an inorganic nitrogen concentration and an active phosphate concentration; The deviation and the rate of change of the deviation are input into the fuzzy reasoning module, fuzzified through a seven-level fuzzy classification system, and the initial PID gain parameters are generated based on the preset fuzzy rule base; The initial PID gain parameters and real-time feedback values ​​are input into a radial basis function network RBF, the Jacobian matrix is ​​calculated by the radial basis function, and the network parameters are updated online using the gradient descent method; Through the adaptive parameter fusion mechanism, the initial PID gain parameters output by the fuzzy inference module and the correction parameters output by the RBF are weighted and fused to obtain the final PID parameters; A control signal is generated according to the final PID parameters to drive the actuator to adjust the port flow field to form a closed-loop control circuit.

2. The control method according to claim 1, characterized in that: The method further comprises: The rules in the fuzzy rule base are generated in the form of conditional statements, and an intelligent environment perception system is established to form a fuzzy rule base; Calculate the correction of proportional, integral and differential coefficients in real time according to the rules in the fuzzy rule base 、 、 , the exponential decay function is used to dynamically adjust the correction coefficient, and its mathematical expression is: ,in, 、 、 Represent the correction coefficients of the proportional term, integral term and differential term respectively, is the exponential decay function value, is the initial value of each correction coefficient, represents the decay rate constant, represents the base of natural logarithms, is the time elapsed from the initial moment; Proportional coefficient The value of is determined by the coefficient of the previous moment With the current adjustment Determine together, the integral coefficient The value is determined by the integration coefficient of the previous moment With the current adjustment Determine together, the differential coefficient value By the differential coefficient of the previous moment With the current adjustment jointly determined; among which, 、 、 Respectively expressed in The initial setting value of each parameter at the moment, 、 、 It is the parameter adjustment amount determined based on the fuzzy rule base.

3. The control method according to claim 1, wherein: The Jacobian matrix is ​​calculated by radial basis function, and the network parameters are updated online using gradient descent method, including: The Gaussian function is selected as the specific implementation form of the radial basis function, and its mathematical expression is as follows: in, For the The basis width vector of nodes, Indicates the The center vector of the node, =1,2,…m; Calculate the neural network output according to the weight coefficient of the RBF neural network and network identification performance indicator function ,as follows: in, Indicates the measured value of water quality index in target water area, is the network weight coefficient from the hidden layer to the output layer, is the hidden layer The output of a neuron; Update the node center vector by iteratively , basis width vector , network weight coefficient and the Jacobian matrix Identify sensitivity information, update network identification indicators, and obtain the optimal value of network identification indicators.

4. The control method according to claim 3, characterized in that: The method further comprises: During the network parameter update process, the following operations are performed in sequence at each time step t: , actual system output With model output The difference between the The output value of a neuron , calculate the weight parameter The amount of change ; Update weight parameters , through the adder Weight coefficient of time and Add together and incorporate the momentum term , multiplied by the weight coefficient through the multiplier exist Moment and The difference between the time and the weight coefficient Make updates; Based on the network learning rate , error difference , No. The output value of a neuron , updated weight coefficient , the input vector calculated by the dot product With the center vector The square of the Euclidean distance , and the modulus of the basis width vector , calculate the basis width vector The amount of change ; Update the base width vector When Basis width vector at time and Added by adder and introduced momentum term Multiply the base width vector by the multiplier exist Moment and The difference between the two moments is used as the updated weight coefficient.

5. The control method according to claim 4, characterized in that: The method further comprises: Update the calculated center vector Each component , according to the network learning efficiency , error difference , updated weight coefficient value , input vector and the center vector component The difference between the values ​​of , calculate the change of the center vector component through the divider and multiplier ; Update the center vector components When The central vector component at time and Add together and add momentum , multiplied by the center vector component exist Moment and The difference in time; The adder accumulates the and , center vector component The control change of the PID output to the regulating valve The difference is the same The product of the ratios of .

6. The control method according to claim 5, characterized in that: Generating a control signal according to the final PID parameter includes: PID parameter increment output by RBF neural network 、 、 Determined by the following formula: in, 、 、 Respectively represent the PID gain learning rate; Generating a control signal according to the final PID parameter includes: The control signal is determined by the following formula: Among them, the control signal The calculation involves the following parameters: time point The controller output value , time point The deviation change , and time points The error of the second-order backward difference , the second derivative of the approximation error, time point The deviation change ; Calculating the PID controller output value During the process, the time point in the historical deviation data The deviation change and time point The deviation change Used to predict changing trends of water quality indicators.

7. An adaptive fuzzy PID flow field control device, characterized in that: The device comprises: a calculation unit, configured to collect water quality index data in a target water area and determine a deviation and a rate of change of the deviation between the water quality index data and a target value; wherein the water quality index data includes at least an inorganic nitrogen concentration and an active phosphate concentration; The fuzzy processing unit is used to input the deviation and the deviation change rate into the fuzzy reasoning module, perform fuzzification processing through a seven-level fuzzy classification system, and generate the initial PID gain parameters based on the preset fuzzy rule base; An updating unit is used to input the initial PID gain parameters and real-time feedback values ​​into a radial basis function network RBF, calculate the Jacobian matrix through the radial basis function, and update the network parameters online using a gradient descent method; The fusion unit is used to perform weighted fusion of the initial PID gain parameters output by the fuzzy inference module and the correction parameters output by the RBF through an adaptive parameter fusion mechanism to obtain the final PID parameters; The driving unit is used to generate a control signal according to the final PID parameters and drive the actuator to adjust the port flow field to form a closed-loop control circuit.

8. The control device according to claim 7, characterized in that: The device further comprises: The generation unit generates the rules in the fuzzy rule base in the form of conditional statements, and establishes an intelligent environment perception system to form a fuzzy rule base; The adjustment unit is used to calculate the correction amount of proportional, integral and differential coefficients in real time according to the rules in the fuzzy rule base 、 、 , the exponential decay function is used to dynamically adjust the correction coefficient, and its mathematical expression is: ,in, 、 、 Represent the correction coefficients of the proportional term, integral term and differential term respectively, is the exponential decay function value, is the initial value of each correction coefficient, represents the decay rate constant, represents the base of natural logarithms, is the time elapsed from the initial moment; the proportional coefficient The value of is determined by the coefficient of the previous moment With the current adjustment Determine together, the integral coefficient The value is determined by the integration coefficient of the previous moment With the current adjustment Determine together, the differential coefficient value By the differential coefficient of the previous moment With the current adjustment jointly determined; among which, 、 、 Respectively expressed in The initial setting value of each parameter at the moment, 、 、 It is the parameter adjustment amount determined based on the fuzzy rule base.

9. The control device according to claim 7, characterized in that: The updating unit is further configured to: The Gaussian function is selected as the specific implementation form of the radial basis function, and its mathematical expression is as follows: in, For the The basis width vector of nodes, Indicates the The center vector of the node, =1,2,…m; Calculate the neural network output according to the weight coefficient of the RBF neural network and network identification performance indicator function ,as follows: in, Indicates the measured value of water quality index in target water area, is the network weight coefficient from the hidden layer to the output layer, is the hidden layer The output of a neuron; Update the node center vector by iteratively , basis width vector , network weight coefficient and the Jacobian matrix Identify sensitivity information, update network identification indicators, and obtain the optimal value of network identification indicators.

10. The control device according to claim 9, characterized in that: The updating unit is further configured to: During the network parameter update process, the following operations are performed in sequence at each time step t: , actual system output With model output The difference between the The output value of a neuron , calculate the weight parameter The amount of change ; Update weight parameters , through the adder Weight coefficient of time and Add together and incorporate the momentum term , multiplied by the weight coefficient through the multiplier exist Moment and The difference between the time and the weight coefficient Make updates; Based on the network learning rate , error difference , No. The output value of a neuron , updated weight coefficient , the input vector calculated by the dot product With the center vector The square of the Euclidean distance , and the modulus of the basis width vector , calculate the basis width vector The amount of change ; Update the base width vector When Basis width vector at time and Added by adder and introduced momentum term Multiply the base width vector by the multiplier exist Moment and The difference between the two moments is used as the updated weight coefficient.

Images