Liquid level control system and method

Abstract

The present application relates to the field of liquid level control technology, and in particular to a liquid level control system and method. The present application analyzes the received liquid level data based on a predictive iterative learning control model to obtain a liquid level control result, generates a liquid level control instruction based on the liquid level control result, and sends the liquid level control instruction to the actuator so that the actuator controls the liquid level of the liquid in the water tank according to the received liquid level control instruction, wherein the predictive iterative learning control model is a predictive model established along the iteration axis based on a linear data model, and the linear data model is a nonlinear mechanism model established by analyzing the dynamic material balance relationship of the water tank. In this way, the present application is aimed at a networked liquid level control system with a repeated operation feature, and by combining predictive control and iterative learning control methods, it reduces the adverse effects of the data attenuation problem existing in the transmission of the networked system on the controller performance, and improves the accuracy of the liquid level control.

Description

Technical Field

[0001] The present application relates to the field of liquid level control technology, and in particular to a liquid level control system and method. Background Art

[0002] Liquid level control plays an irreplaceable role in industrial systems, improving production efficiency and safety. It is widely used in industries such as steel metallurgy, petrochemicals, food processing, and solution filtration. Conventional PID control strategies struggle to achieve effective control of complex batch process liquid level control systems, a factor that has become a significant factor impacting production quality and economic profitability. Therefore, research into advanced liquid level control methods and strategies is crucial for ensuring the rationality, efficiency, and safety of industrial processes.

[0003] It should be noted that the information disclosed in the above background technology section is only used to enhance the understanding of the background of the present disclosure, and therefore may include information that does not constitute prior art known to ordinary technicians in the field. Summary of the Invention

[0004] In view of this, the embodiments of the present application at least provide a liquid level control system and method, which reduces the adverse impact of data attenuation problems existing in network system transmission on controller performance and improves the accuracy of liquid level control.

[0005] This application mainly includes the following aspects:

[0006] In a first aspect, an embodiment of the present application provides a liquid level control system, comprising a controller, an actuator, and a sensor, wherein the sensor is mounted on the surface of the liquid or water tank, and the controller is communicatively connected to the sensor and the actuator via a communication network; wherein:

[0007] The sensor is used to obtain liquid level data of the liquid in the water tank in real time and send the liquid level data to the controller;

[0008] The controller is configured to analyze received liquid level data based on a predictive iterative learning control model to obtain a liquid level control result, generate a liquid level control instruction based on the liquid level control result, and send the liquid level control instruction to the actuator; the predictive iterative learning control model is a predictive model established along an iteration axis based on a linear data model, and the linear data model is a nonlinear mechanism model established by analyzing the dynamic material balance relationship of the water tank;

[0009] The actuator is used to control the liquid level of the liquid in the water tank according to the received liquid level control instruction.

[0010] In a second aspect, an embodiment of the present application further provides a liquid level control method, the liquid level control method comprising:

[0011] Receive real-time liquid level data of the liquid in the water tank sent by the sensor;

[0012] Analyzing the received liquid level data based on the predictive iterative learning control model to obtain a liquid level control result, generating a liquid level control instruction based on the liquid level control result, and sending the liquid level control instruction to the actuator so that the actuator controls the liquid level of the liquid in the water tank according to the received liquid level control instruction;

[0013] The predictive iterative learning control model is a predictive model established along the iteration axis based on a linear data model, and the linear data model is a nonlinear mechanism model established by analyzing the dynamic material balance relationship of the water tank.

[0014] In a third aspect, an embodiment of the present application further provides an electronic device comprising: a processor, a memory and a bus, wherein the memory stores machine-readable instructions executable by the processor. When the electronic device is running, the processor and the memory communicate through the bus, and the machine-readable instructions are executed by the processor to execute the steps of the liquid level control method described in the second aspect above.

[0015] In a fourth aspect, an embodiment of the present application further provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the liquid level control method described in the second aspect are executed.

[0016] The liquid level control system and method provided in the embodiments of the present application analyzes received liquid level data based on a predictive iterative learning control model to obtain a liquid level control result, generates a liquid level control instruction based on the liquid level control result, and sends the liquid level control instruction to an actuator so that the actuator controls the liquid level of the liquid in the water tank according to the received liquid level control instruction, wherein the predictive iterative learning control model is a predictive model established along the iteration axis based on a linear data model, and the linear data model is a nonlinear mechanism model established by analyzing the dynamic material balance relationship of the water tank. In this way, the present application is aimed at a networked liquid level control system with a repetitive operation characteristic, and by combining predictive control and iterative learning control methods, it reduces the adverse effects of data attenuation problems existing in networked system transmission on controller performance and improves the accuracy of liquid level control.

[0017] In order to make the above-mentioned objects, features and advantages of the present application more obvious and easy to understand, preferred embodiments are given below and described in detail with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following is a brief introduction to the drawings required for use in the embodiments. It should be understood that the following drawings only show certain embodiments of the present application and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other relevant drawings can be obtained based on these drawings without creative work.

[0019] Figure 1 A schematic structural diagram of a liquid level control system provided in an embodiment of the present application is shown;

[0020] Figure 2 Schematic diagrams of the liquid level curves of the water tank at the 3rd, 15th, and 40th iterations are shown;

[0021] Figure 3 A schematic diagram showing a curve of the maximum tracking error of PILC along the iteration axis;

[0022] Figure 4 A schematic diagram of the maximum tracking error of PILC along the iteration axis under different fading channel means is shown;

[0023] Figure 5 A schematic diagram of the maximum tracking error of PILC along the iteration axis under different fading channel variances is shown;

[0024] Figure 6 A schematic diagram showing the effect of PILC and PID in tracking the water tank level;

[0025] Figure 7 The schematic diagram of the effect of comparing the control quantities of PILC and PID is shown;

[0026] Figure 8 A flow chart of a liquid level control method provided in an embodiment of the present application is shown;

[0027] Figure 9 A schematic structural diagram of an electronic device provided in an embodiment of the present application is shown. DETAILED DESCRIPTION

[0028] In order to make the purpose, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. It should be understood that the drawings in the present application only serve the purpose of illustration and description and are not used to limit the scope of protection of the present application. In addition, it should be understood that the schematic drawings are not drawn to scale. The flowcharts used in this application illustrate the operations implemented according to some embodiments of the present application. It should be understood that the operations of the flowcharts can be implemented out of sequence, and steps without logical context can be reversed or implemented simultaneously. In addition, those skilled in the art, under the guidance of the contents of this application, can add one or more other operations to the flowchart, or remove one or more operations from the flowchart.

[0029] In addition, the described embodiments are only a part of the embodiments of the present application, rather than all of the embodiments. The components of the embodiments of the present application generally described and shown in the drawings here can be arranged and designed in various configurations. Therefore, the following detailed description of the embodiments of the present application provided in the drawings is not intended to limit the scope of the claimed application, but merely represents selected embodiments of the present application. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without making creative work are within the scope of protection of the present application.

[0030] In order to enable those skilled in the art to use the contents of this application, the following implementation methods are given in combination with the specific application scenario "liquid level control". For those skilled in the art, the general principles defined here can be applied to other embodiments and application scenarios without departing from the spirit and scope of this application.

[0031] The following methods, devices, electronic devices or computer-readable storage media of the embodiments of the present application can be applied to any scenario requiring liquid level control. The embodiments of the present application are not limited to specific application scenarios. Any solution using the liquid level control system and method provided by the embodiments of the present application is within the scope of protection of this application.

[0032] It is worth noting that before the present application, liquid level control played an irreplaceable role in industrial systems. It can improve production efficiency and safety and is widely used in industries such as steel metallurgy, petrochemicals, food processing, and solution filtration. For a class of complex batch process liquid level control systems, it is difficult to achieve good control effects using conventional PID control strategies, which has become an important factor affecting the production quality and economic benefits of enterprises. Therefore, the study of advanced liquid level control methods and strategies is of great significance to ensure the rationality, efficiency, and safety of industrial processes.

[0033] In recent years, a growing number of researchers have applied advanced control strategies to liquid level control in traditional process systems, achieving satisfactory results. Without requiring a precise model for tank level control, an online fuzzy adaptive PID controller was designed by combining a host computer and a programmable controller, demonstrating good stability and robustness. To further improve controller performance, fuzzy control and fuzzy integral control were respectively introduced into a dual-tank level control scheme, and the advantages and disadvantages of the two intelligent algorithms were analyzed. Similarly, for dual-tank level control, Wang Junwei et al. designed a model predictive controller based on state equations and demonstrated significant advantages over traditional PID controllers through simulation. These results achieved good control results in the traditional point-to-point data transmission environment of process systems. However, the expansion of physical systems and the long-distance transmission of data have increasingly exposed the limitations of point-to-point transmission. Consequently, more and more modern industrial process control systems are being designed with networked architectures. Effective control methods are still lacking to address the limited communication resources and bandwidth of the network itself, as well as the attenuation channel caused by reflection, refraction, and diffraction during transmission. Therefore, studying compensation control methods for fading channels has important theoretical significance and practical value.

[0034] In industrial process systems, a pump periodically delivers water from a pool to a storage tank in batches, and the water level in the tank is controlled by the valve opening. This process is characterized by typical repetitiveness. For systems that operate repeatedly within a finite time interval, iterative learning control (ILC) uses only the error and control information from the previous one or several previous actions to correct the current control input. Through repeated operation, the system's actual output accurately tracks the desired reference trajectory. ILC has been widely used in fields such as robotics, trains, chemical processes, and multi-agent systems, achieving excellent control results. In recent years, research on applying ILC to networked systems has achieved some success. For linear networked stochastic systems where both input and output data may be subject to data decay, Shen Dong modified the transmitted signal at specific fading locations and designed a P-type ILC controller using the modified signal. For unknown nonlinear networked systems subject to data decay, Bu Xuhui constructed event-triggered conditions along the iteration and time domains to conserve communication resources during the iterative learning process. He proposed an event-triggered model-free adaptive iterative learning control (MFAILC) approach. The iterative controllers in the aforementioned studies are all driven by the previous error information. While they can improve system performance during repeated system operations, they do not take advantage of the network's ability to transmit large amounts of data at once using data packets. Predictive control stems directly from the practical application of industrial process control and has been continuously improved and matured through close integration with industrial applications. This control method can be divided into two parts: output prediction of the prediction model and deviation-based prediction correction. It can fully utilize the characteristics of data packet transmission, using the latest data packet on the control side to calculate multi-step predictive control variables and transmit them in a package to the actuator. Therefore, how to combine predictive control with iterative learning control, taking into account the characteristics of networked liquid level control systems such as nonlinearity, large lags, and large time constants, and designing a predictive iterative learning controller that can address the data decay problem in the system will be a technical challenge of this study.

[0035] In response to the above-mentioned problems, the embodiment of the present application analyzes the received liquid level data based on a predictive iterative learning control model to obtain a liquid level control result, generates a liquid level control instruction based on the liquid level control result, and sends the liquid level control instruction to the actuator so that the actuator controls the liquid level of the liquid in the water tank according to the received liquid level control instruction, wherein the predictive iterative learning control model is a predictive model established along the iteration axis based on a linear data model, and the linear data model is a nonlinear mechanism model established by analyzing the dynamic material balance relationship of the water tank. In this way, the present application is aimed at a networked liquid level control system with a repetitive operation characteristic, and by combining predictive control and iterative learning control methods, it reduces the adverse effects of the data attenuation problem existing in the transmission of the networked system on the controller performance and improves the accuracy of the liquid level control.

[0036] To facilitate understanding of the present application, the technical solutions provided in the present application are described in detail below in conjunction with specific embodiments.

[0037] Figure 1 This is a structural diagram of a liquid level control system 100 provided in an embodiment of the present application. Figure 1 As shown, the liquid level control system 100 provided in an embodiment of the present application includes a controller 110, an actuator 120 and a sensor 130. The sensor 130 is installed on the surface of the liquid or water tank, and the controller 110 is communicatively connected with the sensor 130 and the actuator 120 respectively through a communication network.

[0038] The sensor 130 is used to obtain liquid level data of the liquid in the water tank in real time and send the liquid level data to the controller 110 .

[0039] The controller 110 is used to analyze the received liquid level data based on the predictive iterative learning control model to obtain a liquid level control result, generate a liquid level control instruction based on the liquid level control result, and send the liquid level control instruction to the actuator 120; the predictive iterative learning control model is a predictive model established along the iteration axis based on a linear data model, and the linear data model is a nonlinear mechanism model established by analyzing the dynamic material balance relationship of the water tank.

[0040] The actuator 120 is configured to control the liquid level of the liquid in the water tank according to the received liquid level control instruction.

[0041] In practice, the networked liquid level control system 100 is primarily composed of three nodes, which work together to achieve precise liquid level control. First, there are the nodes consisting of sensors 130, which are installed on the surface of a liquid or container (such as a water tank) to monitor liquid level changes in real time. Next, there is the control node, or controller 110, which receives and analyzes data from the sensors 130. Finally, there are the execution nodes, or actuators 120, which are specifically designed to execute the corresponding control operations. Furthermore, a communication network ensures that information can be smoothly transmitted between nodes, forming a network that covers the entire system.

[0042] It should be noted that this application uses a predictive iterative learning control method to conduct theoretical and applied research on water tank liquid level control, achieving high-precision liquid level tracking for a networked liquid level control system 100 in the presence of attenuation channels on both the input and output sides. Specifically, first, the dynamic material balance relationship of the water tank is analyzed to establish its nonlinear mechanism model, and a dynamic linearization method is used to obtain its linear data model in the iteration domain. Then, based on the linear data model, a predictive model along the iteration axis is established, and the predictive information is used to design a predictive iterative learning control model to achieve precise control of the water tank liquid level.

[0043] Here, the liquid level control system 100 has strong nonlinearity and hysteresis. When the system operating conditions change, the model-based control method will face huge challenges, while the model-free iterative learning control method can dynamically update the pseudo-partial derivatives of the system according to the input and output data of the system, thereby adapting to the changes in the control conditions. This application applies iterative learning control to the liquid level control system 100 with a large time constant and large hysteresis for the first time. For the networked liquid level control system 100 with repeated operation characteristics, a predictive iterative learning control scheme is designed using only the system input and output data. The scheme includes a gain compensation part of the post-fading data, a predictive model of the iterative axis, and a learning control model driven by the prediction information to update. Specifically, by combining predictive control and iterative learning control methods, the adverse effects of the data attenuation problem existing in the transmission of the networked system on the performance of the controller 110 can be reduced, and the accuracy of the liquid level control can be improved.

[0044] It should be noted that, for the networked single-capacity water tank liquid level control system provided in the embodiment of the present application, the networked single-capacity water tank liquid level control system 100 is mainly composed of a water tank, a water tank, a weight sensor 130 (upper weight sensor, lower weight sensor), a controller 110 (such as a host computer) and an actuator 120 (such as a peristaltic pump), wherein the sensor 130, the controller 110 and the actuator 120 are connected via a network. Specifically, the networked single-capacity water tank liquid level control system may include a water inlet 1, an upper water tank, an upper weight sensor, a water inlet 2, a lower water tank, a lower weight sensor, a water reservoir, a water pumping port, a data transmission interface, a peristaltic pump and a host computer. Among them, the amount of water flowing in through the lower water inlet is , is the system input, and the outflow at the outlet is , the controlled object is the liquid level ,Specifically, the mathematical model of a single capacity water tank is derived by analyzing ,the dynamic characteristics of the liquid level.

[0045] In a possible implementation, the controller 110 is specifically configured to establish the linear data model according to the following steps:

[0046] By analyzing the dynamic characteristics of the water tank level, the ideal dynamic characteristic differential equation of a single-capacity water tank is derived;

[0047] Transforming the ideal dynamic characteristic differential equation to obtain a water tank mathematical model of a continuous-time system;

[0048] The water tank mathematical model is expressed as a discrete time system model, and a nonlinear system model is obtained by introducing an iteration index;

[0049] The nonlinear system model is used to obtain the linear data model.

[0050] In a possible implementation, the ideal dynamic characteristic differential equation is:

[0051] Formula (1);

[0052] Formula (2);

[0053] in, is the cross-sectional area of the water tank, is the water inflow from the lower water inlet of the water tank, is the outflow rate of the water tank outlet, is the liquid level of the water tank, For time, is the liquid outflow rate in the water tank, is the cross-sectional area of the water tank outlet pipe, is the gravity coefficient.

[0054] Here, the ideal dynamic characteristic differential equation of a single-capacity water tank is established based on the dynamic material balance relationship. According to the small-hole Bernoulli equation, the water outflow rate in the water tank is satisfy , rewrite formula (1) as: Formula (3).

[0055] In a possible implementation, the water tank mathematical model is , the discrete-time system model is , the nonlinear system model is , the linear data model is ;

[0056] in, , , is the sampling time, is the iteration index, For about and The nonlinear function of , It is a time-varying parameter used to represent the characteristics of the system. Subsystem operation Parameters of the time system; is a non-negative constant, The characterization system has a finite input and a finite output. Different actual systems Has different values.

[0057] In the specific implementation, , , are the output and input of the system respectively, then formula (3) can be expressed as Formula (4). Then, the “zero-order holder discretization method” is adopted, that is, the instantaneous value of the variable at the sampling moment is used as its sampling value in the sampling period, and the water tank model of the continuous time system - Formula (4) is expressed as a discrete time system model: Formula (5); where is the sampling moment. Considering the repeated operation characteristics of the system, formula (5) can be improved by introducing the iteration index , Expressed as Formula (6), record For about and nonlinear function.

[0058] Before using the dynamic linearization method, it is determined whether formula (6) satisfies the following assumptions:

[0059] Assumption 1 right has continuous partial derivatives.

[0060] Assumption 2 System (5) satisfies the generalized Lipschitz condition along the iteration axis, for any , , if there is ,but

[0061] ,

[0062] in, is a positive constant, , .

[0063] For the actual liquid level system, the control input energy of the system is bounded and easy to judge right The partial derivatives of are continuous, and assumptions 1 and 2 are easily satisfied. Therefore, the nonlinear system model can be converted into the following linear data model using the dynamic linearization method: , formula (7),

[0064] in , is a non-negative constant.

[0065] In a possible implementation, for attenuation channel model construction, Figure 2 Shows that the networked system uses the network to transmit output data and control data The present application comprehensively considers the problem of attenuation channels that may exist on both the input and output sides. The existence of attenuation channels causes a multiplicative attenuation damage to the data during the transmission process. Therefore, the two multiplicative random variables , are used to describe the effect of the attenuation channel on the system output and control data The system output after fading and control data They are defined as:

[0066] Formula (8);

[0067] Formula (9);

[0068] in, are the fading coefficients of independent and identically distributed Gaussian random variables, and their expectations are defined as: , the variance can be expressed as: , assuming that the mean and variance of the fading coefficients are known. , are two random white noise sequences with zero expectation.

[0069] In a possible implementation, the controller 110 is specifically configured to establish the predictive iterative learning control model according to the following steps:

[0070] determining a tracking error model based on the linear data model;

[0071] Determining a predictive control algorithm model based on the tracking error model;

[0072] Determining a pseudo partial derivative prediction algorithm model according to the predictive control algorithm model;

[0073] The pseudo partial derivative prediction algorithm model is determined as the predictive iterative learning control model.

[0074] In one possible implementation, the tracking error model is:

[0075] ;

[0076] in, is the expected trajectory, For the The output of times, , , for estimated value.

[0077] In the specific implementation, for the establishment of the predictive iterative learning control model, the prediction model is designed first. Specifically, the tracking error is defined as ,in is the expected trajectory. According to formula (7), The output of this can be expressed as Formula (10);

[0078] Wherein, the expected trajectory is subtracted from both sides of formula (10) , then Second and The relationship between the tracking errors of the iterations is shown as follows:

[0079] Formula (11);

[0080] Since (11) It is unknown, remember for The estimated value of Representative Second system operation The parameters of the time system, then formula (11) can be expressed as Formula (12).

[0081] Furthermore, based on formula (12), The step-by-step iterative prediction model is shown below:

[0082] Formula (13);

[0083] in, express Forward prediction It is worth noting that in formula (13) and In the The number of iterations is unknown. Representing the , , , Second system operation Parameters of the time system.

[0084] Here, the definition

[0085] ,

[0086] ,

[0087] Then the last line of formula (13) It can be expressed as the following vector form,

[0088] Formula (14).

[0089] In a possible implementation, the controller 110 is specifically configured to determine the predictive control algorithm model according to the following steps:

[0090] For a given tracking target , the control objective is to design a suitable control input The system tracking error exist Converges to Using the prediction model formula (14), the designed cost function is Formula (15), using the optimization condition ,but can be expressed as Formula (16).

[0091] in, is the weighting factor, is the step factor. Here, due to the existence of the fading channel on the output side, Not available, but the controller 110 can obtain the measured value after fading , after gain compensation, the corrected system output is obtained .so quilt replace. Expressed as Using formula (16), based on the rolling optimization principle, only The first predicted value in drives the control input along the iteration axis Update, a predictive iterative learning control scheme is proposed as Formula (17);

[0092] in, is a selection matrix. Similarly, considering the existence of the fading channel on the input side of the system, the control signal calculated by the controller 110 is is unavailable, use the corrected control signal Acting on water tank .

[0093] In a possible implementation, the controller 110 is specifically configured to determine the pseudo partial derivative prediction algorithm model according to the following steps:

[0094] From formula (16), it can be seen that the key to designing the predictive controller 110 is to calculate the value of the pseudo partial derivative. Therefore, the pseudo partial derivative is estimated using the following parameter estimation criterion function. The estimation criterion function of the pseudo partial derivative is: Formula (18), where As described in formula (18), it is a typical quadratic function. The independent variables included in the function are , and ;

[0095] Use optimized conditions , Characterizing quadratic index functions about Find the partial derivative, then can be expressed as:

[0096] Formula (19);

[0097] ,if or Formula (20);

[0098] in, is the weight factor, is the step size factor, is a very small constant value. yes The initial value of . Using the estimation algorithm formula (21), we can get the times before the pseudo partial derivative estimate, so use , we can build an autoregressive model to predict The pseudo-partial derivative estimate of Formula (21);

[0099] in, is the unknown coefficient ( , ), is the intrinsic order of the model. By definition , , formula (21) can be rewritten as Formula (22).

[0100] here, The unknown coefficients A vector composed of has no actual physical meaning. It can be determined by the following projection algorithm:

[0101] Formula (23);

[0102] ,if , in, and is a known non-negative constant, the initial value can be chosen to be a bounded constant vector, is a constant. Using formulas (21)-(23), the value in formula (16) It can be obtained that the proposed predictive iterative learning control model is designed, where is Estimated value of The vector composed of Indicates system parameters exist Run Estimated value of the moment.

[0103] In one possible embodiment, the controller 110 is also used to compare the control effects of the predictive iterative learning control model and the traditional PID algorithm model on the liquid level using a comprehensive liquid level control experimental system, so as to determine the effectiveness of the predictive iterative learning control model through experimental verification.

[0104] In the specific implementation, the superiority of the proposed algorithm was verified by comparing it with the PID algorithm on the LC-T2 platform. The input side and output side of the system and the attenuation gain are selected as , The expected trajectory is a sine signal with an amplitude of 100 and a period of 300 seconds. , the running time of each iteration is 150 seconds, and the sampling interval is 0.1 seconds. The cross-sectional area of the water tank 154 , the cross-sectional area of the outlet pipe 0.5 , the initial liquid level of the water tank in each iteration The parameters of the predictive iterative learning controller used in the embodiment of the present application are selected as follows: , , , , , The parameters in the pseudo partial derivative prediction algorithm formulas (21)-(23) are set as: , , , , ;in, is the step size factor, which is used to control the change of input quantity. It is introduced to make the algorithm flexible; is a positive number, the purpose of which is to prevent the denominator from being zero; Is a small positive number that resets the estimated parameters Conditions, when the sign of the controlled parameter is inconsistent with the sign of the initial value of the estimated parameter or the estimated value of the estimated parameter itself is small; is a normal value, used to ensure that the denominator in formula (23) is not zero; Is a constant value used to reset , that is, when the controlled parameter exceeds 20, Reset to .

[0105] Figure 2 The figure shows the curve diagram of the liquid level of the water tank at the 3rd, 15th and 40th iterations. Figure 3 A schematic diagram showing the maximum tracking error of PILC along the iteration axis is shown. Figure 4 The figure shows the curve diagram of the maximum tracking error of PILC along the iteration axis under different fading channel means. Figure 5 The following is a schematic diagram showing the maximum tracking error of PILC along the iteration axis under different fading channel variances. Figure 2 When there are attenuation channels on both the input and output sides of the system, the PILC algorithm is used. The liquid level of the system at the 3rd, 15th, and 40th iterations is used. To further analyze the performance of the designed PILC controller under different attenuation channels, three different values of mean and variance are selected for experiments. For the mean, set They are 0.95, 0.85, and 0.75 respectively, and the variance is fixed The corresponding maximum tracking error is Figure 4 shown. Figure 4 It shows that the smaller the mean, the worse the data quality and the worse the tracking effect of the controller, but the maximum error can still converge to a suitable range as the number of iterations increases. For the variance, let They are 0.1, 0.2, and 0.4 respectively, and the mean is fixed ,in, is the variance of the attenuation gain on the output side, is the attenuation gain variance on the input side, is the mean value of the attenuation gain on the output side, is the mean attenuation gain on the input side. Figure 5 The maximum tracking error curve of the system under different variances is shown in Figure 2. It can be found that a smaller variance indicates that the fading effect of communication is relatively deterministic and the designed controller can work effectively.

[0106] In order to illustrate the superiority of the proposed PILC algorithm over the traditional PID algorithm, a comparative experiment was conducted in the presence of a decay channel in the system. The PID controller used is shown in the following equation, and the controller parameters are , .

[0107] here, Figure 6 A schematic diagram showing the effect of PILC and PID in tank level tracking comparison. Figure 7 The following figure shows the effect of PILC and PID control (a, b systems have attenuation channels; c, d systems do not have attenuation channels). The system tracking effect of the desired tank level at the 40th iteration using the PILC algorithm is compared with the tracking effect of the desired trajectory using the PID algorithm. Figure 6 shown. Figure 6 This study demonstrates that traditional PID algorithms are not effective for nonlinear systems like water tanks, which have large time constants and hysteresis. The water tank's level control has a significant hysteresis, while the PILC algorithm can improve control by eliminating this hysteresis through learning. The presence of attenuation channels in networked water tank level control systems can reduce the accuracy of both the measured level received by the controller and the control signal received by the actuator. This data attenuation can lead to sudden changes in motor speed, significantly impacting the motor's service life and even affecting its stable operation. Figure 7 Figures (a), (b), (c), and (d) show the motor speed variations of the proposed PILC algorithm and the traditional PID algorithm in two scenarios: with and without a system attenuation channel. (a) and (b) show that when the system has a system attenuation channel, the traditional PID algorithm experiences a random decrease in input and output data quality, causing large speed fluctuations and significantly damaging the actuator. However, the small fluctuations of the PILC algorithm are acceptable in engineering and enable stable control. (c) and (d) show that when the system does not have a system attenuation channel, the PID motor speed is smooth but exhibits lag, preventing the system from responding immediately. Therefore, overall, the PILC control strategy clearly outperforms the PID control strategy.

[0108] It should be noted that the effectiveness of the method was verified by comparative experiments using the liquid level control comprehensive experimental system LC-T2. The experimental results show that the predictive iterative learning control method provided by the present application can eliminate hysteresis through learning and achieve precise tracking compared to traditional PID. And in the case of the existence of an attenuation channel in the system, the random decline in the quality of input and output data causes large fluctuations in the speed of the PID algorithm, which will greatly damage the actuator. The predictive iterative learning algorithm can ensure that small fluctuations in the speed are acceptable in engineering and can achieve stable control. Therefore, in a networked liquid level control system with attenuation channels on both the input and output sides, the predictive iterative learning control method provided by the embodiment of the present application has more excellent control performance than traditional PID control.

[0109] To address the issue of attenuated channels on the input and output sides of a networked liquid level control system, the present embodiment proposes a predictive iterative learning control method. This solution utilizes only system input and output data, leveraging the data packet transmission characteristics of networked control systems. The controller uses the latest data packet to predict multi-step control variables and transmit them to the actuator. This predicted information is used to track the liquid level, improving the application of iterative learning control in networked control systems. Experiments on the LC-T2 platform verified the stability and effectiveness of the proposed method for networked liquid level control systems.

[0110] Based on the same application concept, the embodiments of the present application also provide a liquid level control method corresponding to the liquid level control system provided in the above embodiments. Since the principle of solving the problem by the device in the embodiments of the present application is similar to the liquid level control system in the above embodiments of the present application, the implementation of the method can refer to the implementation of the method, and the repeated parts will not be repeated.

[0111] like Figure 8 As shown, Figure 8 This is a flow chart of a liquid level control method provided in an embodiment of the present application, wherein the liquid level control method comprises the following steps:

[0112] S801: receiving the liquid level data of the liquid in the water tank sent by the sensor in real time;

[0113] S802: Analyzing the received liquid level data based on the predictive iterative learning control model to obtain a liquid level control result, generating a liquid level control instruction based on the liquid level control result, and sending the liquid level control instruction to the actuator so that the actuator controls the liquid level in the water tank according to the received liquid level control instruction;

[0114] The predictive iterative learning control model is a predictive model established along the iteration axis based on a linear data model, and the linear data model is a nonlinear mechanism model established by analyzing the dynamic material balance relationship of the water tank.

[0115] Based on the same application concept, see Figure 9 As shown, it is a structural diagram of an electronic device 900 provided in an embodiment of the present application, including: a processor 910, a memory 920 and a bus 930, wherein the memory 920 stores machine-readable instructions executable by the processor 910. When the electronic device 900 is running, the processor 910 and the memory 920 communicate with each other through the bus 930, and the machine-readable instructions are executed by the processor 910 when running, as in any of the steps of the liquid level control method described in the above embodiments.

[0116] Based on the same application concept, an embodiment of the present application further provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the liquid level control method provided in the above embodiment are executed.

[0117] Specifically, the storage medium can be a general storage medium, such as a mobile disk, a hard disk, etc. When the computer program on the storage medium is run, the above-mentioned liquid level control method can be executed. By combining predictive control and iterative learning control methods for a networked liquid level control system with repeated operation characteristics, the adverse effect of data attenuation problems in networked system transmission on controller performance is reduced, and the accuracy of liquid level control is improved.

[0118] In the embodiment of the present application, the computer program can also execute other machine-readable instructions when run by the processor to execute other methods described in the embodiment. For the specific execution method steps and principles, please refer to the description of the embodiment and will not be repeated here.

[0119] In the embodiments provided in this application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are merely schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some communication interface, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0120] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0121] In addition, each functional unit in the embodiments provided in the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0122] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes a number of instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, and other media that can store program code.

[0123] It should be noted that similar numbers and letters represent similar items in the following figures. Therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. In addition, the terms "first", "second", "third", etc. are only used to distinguish the description and are not to be understood as indicating or implying relative importance.

[0124] Finally, it should be noted that the above-described embodiments are only specific implementation methods of the present application, which are used to illustrate the technical solutions of the present application, rather than to limit them. The scope of protection of the present application is not limited thereto. Although the present application has been described in detail with reference to the above-described embodiments, those skilled in the art should understand that any person skilled in the art can modify or easily conceive of changes to the technical solutions described in the above-described embodiments within the technical scope disclosed in the present application, or make equivalent replacements for some of the technical features thereof. However, these modifications, changes, or replacements do not deviate from the spirit and scope of the technical solutions of the embodiments of the present application. They should all be included in the scope of protection of the present application. Therefore, the scope of protection of the present application should be based on the scope of protection of the claims.

Claims

1. A liquid level control system, characterized in that: The liquid level control system includes a controller, an actuator and a sensor. The sensor is installed on the surface of the water tank. The controller is connected to the sensor and the actuator through a communication network. The sensor is used to obtain liquid level data of the liquid in the water tank in real time and send the liquid level data to the controller; The controller is configured to analyze the received liquid level data based on the predictive iterative learning control model to obtain a liquid level control result, generate a liquid level control instruction according to the liquid level control result, and send the liquid level control instruction to the actuator; By designing a cost function related to the linear data model and control input along the iteration axis, the control input driven along the iteration axis is obtained. ; Based on the rolling optimization principle, only use The first predicted value in drives the control signal along the iteration axis renew, represents the control signal calculated by the controller; Control Input The estimation of the pseudo partial derivatives in is obtained using a criterion function based on the difference of the pseudo partial derivatives on the iteration axis and the tracking error; The actuator is configured to control the liquid level of the liquid in the water tank according to the received liquid level control instruction; The controller is specifically configured to establish the linear data model according to the following steps: An ideal dynamic characteristic differential equation of a single-capacity water tank is derived by analyzing the dynamic characteristics of the water tank liquid level; the ideal dynamic characteristic differential equation is transformed to obtain a water tank mathematical model as a continuous-time system; the water tank mathematical model is expressed as a discrete-time system model, and a nonlinear system model is obtained by introducing an iteration index; the nonlinear system model is dynamically linearized on an iteration axis to obtain the linear data model; The controller is specifically configured to establish the predictive iterative learning control model according to the following steps: determining a tracking error model based on the linear data model; determining a predictive control algorithm model based on the tracking error model; determining a pseudo partial derivative prediction algorithm model based on the predictive control algorithm model; and determining the pseudo partial derivative prediction algorithm model as the predictive iterative learning control model; The controller is also used to use the liquid level control comprehensive experimental system to compare the control effect of the predictive iterative learning control model and the traditional PID algorithm model on the liquid level, so as to determine the effectiveness of the predictive iterative learning control model through experimental verification; make full use of the characteristics of data packet transmission of the networked control system, and use the system output with attenuation characteristics on the controller side. Predict the control amount, and after the control amount is compensated Transmitted to the actuator to achieve liquid level tracking, in order to improve the application of predictive iterative learning control in networked control systems, is the fading coefficient of an independent and identically distributed Gaussian random variable, are two random white noise sequences with zero expectation, Indicates the output data transmitted by the system using the network. represents the system output after fading, Indicates control data after fading.

2. The liquid level control system according to claim 1, characterized in that: The ideal dynamic characteristic differential equation is: ； in, , is the cross-sectional area of the water tank, is the water inflow from the lower water inlet of the water tank, is the outflow rate of the water tank outlet, is the liquid level of the water tank, For time, is the liquid outflow rate in the water tank, is the cross-sectional area of the water tank outlet pipe, is the gravity coefficient.

3. The liquid level control system according to claim 2, characterized in that: The mathematical model of the water tank is , the discrete-time system model is , the nonlinear system model is , the linear data model is ; in, , , is the sampling time, is the iteration index, For about and The nonlinear function of , is a non-negative constant.

4. The liquid level control system according to claim 1, characterized in that: The iterative axis error prediction model is: ； in, express Forward prediction The predicted value of the iteration, For the Run The liquid level tracking error at the moment, is the expected trajectory, For the Run Output at any moment, , for estimated value.

5. A liquid level control method, characterized in that: A controller applied to a liquid level control system according to any one of claims 1 to 4, wherein the liquid level control system further comprises an actuator and a sensor, wherein the controller communication network is communicatively connected to the sensor and the actuator, respectively; and the liquid level control method comprises: Receive real-time liquid level data of the liquid in the water tank sent by the sensor; Analyzing the received liquid level data based on the predictive iterative learning control model to obtain a liquid level control result, generating a liquid level control instruction based on the liquid level control result, and sending the liquid level control instruction to the actuator so that the actuator controls the liquid level of the liquid in the water tank according to the received liquid level control instruction; The predictive iterative learning control model is an error prediction model established along the iteration axis based on a linear data model.

6. An electronic device, characterized in that: include: A processor, a memory and a bus, wherein the memory stores machine-readable instructions executable by the processor. When the electronic device is running, the processor and the memory communicate through the bus, and the machine-readable instructions are executed by the processor to execute the steps of the liquid level control method as described in claim 5.

7. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the liquid level control method according to claim 5 are executed.

Images