Control method and device for pneumatic conveying system of bulk grain, and electronic device

Abstract

The present invention relates to the technical field of pneumatic conveying of bulk grain, and discloses a control method, a device and an electronic device for a pneumatic conveying system of bulk grain, aiming at solving the problem that the effect of the existing control scheme in reducing the breakage rate of bulk grain is poor. The solution mainly includes: training a sparse non-linear dynamics identification model according to the control parameters of the pneumatic conveying system of bulk grain and their corresponding response parameters, where the response parameters at least include the breakage rate of bulk grain; when using the pneumatic conveying system of bulk grain to convey bulk grain, using model predictive control to predict the response parameters at future moments based on the sparse non-linear dynamics identification model at each moment; using an optimization algorithm to solve to obtain the optimal control parameter sequence at the current moment, with the optimization objective being the minimum breakage rate of bulk grain; and adjusting the control parameters of the pneumatic conveying system of bulk grain to the first set of optimal control parameters in the optimal control parameter sequence. The present invention improves the control accuracy and significantly reduces the breakage rate of bulk grain during pneumatic conveying. It is applicable to the conveying of bulk grain.

Description

Technical Field

[0001] The present invention relates to the technical field of pneumatic conveying of bulk grain, and particularly relates to a control method and device for a pneumatic conveying system of bulk grain, and an electronic device. Background Art

[0002] A pneumatic conveying system for bulk grain is a system that uses the energy of air flow to convey granular bulk grain in a closed pipeline along the direction of the air flow. Its basic principle is to use compressed air as the conveying medium, and utilize the pressure and speed of the air flow to move the bulk grain from one place to another through the pipeline. During the conveying process, the air flow suspends and pushes the material forward to achieve continuous and uniform conveying, and is applicable to horizontal, vertical or inclined conveying. The system mainly consists of a feeding device, a conveying pipeline, a gas source device, a discharging device, etc.

[0003] In order to reduce the breakage of bulk grain particles during pneumatic conveying, the existing pneumatic conveying systems for bulk grain mainly adjust the fan frequency and valve opening through PID control, so as to adjust variables such as the air flow rate and pressure in the pneumatic conveying system, and achieve the control of the conveying process of bulk grain particles. Specifically, the PID controller performs feedback adjustment on the error of the system (i.e., the difference between the actual bulk grain speed in the air flow and the set value), and through the three control actions of proportional, integral and differential, attempts to stabilize the speed and pressure of the air flow by changing the fan frequency to adjust the air intake volume and adjusting the valve opening to adjust the feeding volume, so as to avoid excessive impact of bulk grain particles or being subjected to excessive air flow force and causing breakage.

[0004] However, the existing PID control method has the following limitations in solving the problem of bulk grain particle breakage: First, in the pneumatic conveying system of bulk grain, the interaction between the behavior of bulk grain particles and the air flow is highly non-linear and changes constantly with the change of the system state. The PID controller cannot accurately capture this non-linearity and time-varying effect, and can only perform linear adjustment according to the error of the system, which makes its effect limited in dealing with the complex interaction between the air flow and bulk grain particles. Second, in the pneumatic conveying system of bulk grain, the air flow speed, pressure and the flow state of bulk grain particles are a multi-input multi-output (MIMO) system with mutual coupling. The PID control does not deal with this coupling effect sufficiently, because it usually adjusts for a single variable and lacks overall optimization and comprehensive consideration of the coupling effect. Summary of the Invention

[0005] The present invention aims to solve the problem that the control scheme of the existing pneumatic conveying system for bulk grain has a poor effect in reducing the breakage rate of bulk grain, and proposes a control method and device for a pneumatic conveying system of bulk grain, and an electronic device.

[0006] The technical solution adopted by the present invention to solve the above technical problems is:

[0007] In a first aspect, the present invention provides a control method for a pneumatic conveying system of bulk grain, and the method includes:

[0008] S1. Obtain multiple sets of control parameters and their corresponding response parameters of the pneumatic conveying system of bulk grain, and train a sparse non-linear dynamics identification model based on the control parameters and their corresponding response parameters, where the response parameters at least include the bulk grain breakage rate;

[0009] S2. When using the pneumatic conveying system of bulk grain to convey bulk grain, perform the following operations at each moment by using model predictive control:

[0010] S21. Predict the response parameters at future moments according to the control parameters at the current moment and based on the sparse non-linear dynamics identification model;

[0011] S22. According to the response parameters at future moments and the set optimization objective and constraint conditions, use an optimization algorithm to solve and obtain the optimal control parameter sequence at the current moment, where the optimization objective is to minimize the bulk grain breakage rate;

[0012] S23. Adjust the control parameters of the pneumatic conveying system of bulk grain to the first set of optimal control parameters in the optimal control parameter sequence.

[0013] Further, the control parameters include the fan frequency and valve opening degree of the pneumatic conveying system of bulk grain;

[0014] The constraint conditions include the variable range of the fan frequency and the variable range of the valve opening degree of the pneumatic conveying system of bulk grain.

[0015] Further, the control parameters further include the ambient temperature and ambient humidity;

[0016] The constraint conditions further include the variable range of the ambient temperature and the variable range of the ambient humidity of the pneumatic conveying system of bulk grain.

[0017] Further, the response parameters further include the momentum parameters of bulk grain particles, the pressure drop of the conveying pipeline, and the density of bulk grain particles in the pipeline, and the momentum parameters include velocity and acceleration.

[0018] Further, training the sparse non-linear dynamics identification model based on the control parameters and their corresponding response parameters includes:

[0019] Taking the control parameters as input data and the corresponding response parameters as output data, constructing a training data set, and training the sparse non-linear dynamics identification model according to the training data set until the sparse non-linear dynamics identification model converges.

[0020] Further, the method further includes:

[0021] Regularly use the validation dataset to determine the error of the sparse non-linear dynamics identification model, and judge whether the error is greater than the preset error. If so, retrain the sparse non-linear dynamics identification model.

[0022] In a second aspect, the present invention provides a control device for a bulk grain pneumatic conveying system, the device comprising:

[0023] A training unit, configured to obtain multiple groups of control parameters of the bulk grain pneumatic conveying system and their corresponding response parameters, and train a sparse non-linear dynamics identification model according to the control parameters and their corresponding response parameters, where the response parameters at least include the bulk grain breakage rate;

[0024] A model predictive control unit, configured to, when conveying bulk grain using the bulk grain pneumatic conveying system, at each moment, predict the response parameters at a future moment according to the control parameters at the current moment and based on the sparse non-linear dynamics identification model; and according to the response parameters at the future moment and the set optimization objective and constraint conditions, use an optimization algorithm to solve and obtain the optimal control parameter sequence at the current moment, where the optimization objective is to minimize the bulk grain breakage rate;

[0025] A parameter adjustment unit, configured to adjust the control parameters of the bulk grain pneumatic conveying system to the first group of optimal control parameters in the optimal control parameter sequence.

[0026] In a third aspect, the present invention provides an electronic device, characterized in that the electronic device includes a processor, a memory, and a communication bus;

[0027] The communication bus is used to implement connection communication between the processor and the memory;

[0028] The processor is configured to execute one or more programs in the memory to implement the steps of the control method for the bulk grain pneumatic conveying system as described in the first aspect.

[0029] The beneficial effects of the present invention are as follows: The control method, device, and electronic equipment for the bulk grain pneumatic conveying system provided by the present invention perform rolling prediction and rolling optimization on the bulk grain pneumatic conveying system based on SINDY-MPC (Sparse Identification of Nonlinear Dynamics for Model Predictive Control), that is, at each moment, the model predictive control (MPC) predicts the response parameters of the bulk grain pneumatic conveying system at future moments based on the sparse nonlinear dynamics identification model (SINDYc), solves the control parameters according to the prediction results with the minimum bulk grain breakage rate as the optimization goal, and then uses the solved control parameters to optimize the control of the bulk grain pneumatic conveying system. SINDY-MPC can capture complex nonlinear dynamics, maintain high control performance under multi-input and multi-output conditions, and can adapt to different working conditions in real time, thereby improving the response speed of control and the system robustness. Through precise model predictive control, the precise optimization of the control parameters of the bulk grain pneumatic conveying system is achieved, and the bulk grain breakage rate during the pneumatic conveying process is significantly reduced. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] Figure 1 It is a schematic flowchart of a control method for a bulk grain pneumatic conveying system provided in an embodiment;

[0031] Figure 2 It is a schematic flowchart of another control method for a bulk grain pneumatic conveying system provided in an embodiment;

[0032] Figure 3 It is a schematic structural diagram of a control device for a bulk grain pneumatic conveying system provided in an embodiment;

[0033] Figure 4 It is a schematic structural diagram of an electronic device provided in an embodiment. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0034] In order to enable those skilled in the art to better understand the solution of the present invention, the technical solutions in this embodiment will be clearly and completely described below in conjunction with the accompanying drawings in this embodiment.

[0035] In some processes described in the specification of the present invention and the above-mentioned drawings, there are multiple operations that appear in a specific order. However, it should be clearly understood that these operations may not be executed in the order in which they appear in this article or may be executed in parallel. The serial numbers of the operations are only used to distinguish different operations, and the serial numbers themselves do not represent any execution order. In addition, these processes may include more or fewer operations, and these operations may be executed in sequence or in parallel.

[0036] The technical solution of the present invention is applicable to application scenarios of transporting bulk grain particles using a pneumatic conveying system for bulk grain, such as rice, wheat, corn, etc. Currently, the control scheme of the pneumatic conveying system for bulk grain usually adopts PID control, that is, the PID controller adjusts the system error by feedback, based on the three control actions of proportional, integral, and differential, and tries to stabilize the air flow velocity and pressure by changing the fan frequency to adjust the air intake and adjusting the valve opening to adjust the feed rate, so as to avoid excessive impact of bulk grain particles or being subjected to excessive air flow force and causing breakage. The inventor has found through research that this scheme has limited effect in dealing with the complex interaction between the air flow and bulk grain particles, and lacks comprehensive consideration of global optimization and coupling effect, resulting in poor effect of reducing the breakage rate of bulk grain particles.

[0037] Based on this, the technical solution of the present invention is proposed. In the present invention, multiple groups of control parameters and their corresponding response parameters of the pneumatic conveying system for bulk grain are obtained, and a sparse nonlinear dynamics identification model is trained according to the control parameters and their corresponding response parameters. The response parameters at least include the breakage rate of bulk grain; when transporting bulk grain using the pneumatic conveying system for bulk grain, model predictive control is used to perform the following operations at each moment: predicting the response parameters at a future moment according to the control parameters at the current moment and based on the sparse nonlinear dynamics identification model; according to the response parameters at the future moment and the set optimization objective and constraint conditions, using an optimization algorithm to solve and obtain the optimal control parameter sequence at the current moment, and the optimization objective is to minimize the breakage rate of bulk grain; adjusting the control parameters of the pneumatic conveying system for bulk grain to the first group of optimal control parameters in the optimal control parameter sequence.

[0038] Specifically, the present invention controls the pneumatic conveying system for bulk grain based on SINDY-MPC. First, a sparse nonlinear dynamics identification model is constructed. Then, in the closed-loop control mode, model predictive control uses the sparse nonlinear dynamics identification model to predict the response parameters of the pneumatic conveying system for bulk grain at a future moment, and based on the set optimization objective of minimizing the breakage rate of bulk grain and related constraint conditions, through the way of rolling optimization, a corresponding set of optimal control parameter sequences is solved at each moment according to the prediction result. This optimization process will find an equilibrium point to ensure that the breakage rate of bulk grain particles is minimized while not violating the constraint conditions. Finally, the first group of control parameters in the optimal control parameter sequence is input into the pneumatic conveying system. Then, re-prediction and optimization are performed to achieve closed-loop control. By combining sparse nonlinear dynamics identification and model predictive control technologies, the present invention can effectively learn and control a complex pneumatic conveying system for bulk grain in a limited data and noise environment, and has advantages such as high performance, low data dependence, high computational efficiency, and strong interpretability. Through precise model predictive control, the precise optimization of the control parameters of the pneumatic conveying system for bulk grain is realized, and the breakage rate of bulk grain during pneumatic conveying is significantly reduced.

[0039] The following will clearly and completely describe the technical solutions in this embodiment in conjunction with the accompanying drawings in this embodiment. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments.

[0040] Figure 1 The flowchart of a control method for a pneumatic conveying system of bulk grain is shown. Please refer to Figure 1 , and the method includes the following steps:

[0041] S1. Obtain multiple groups of control parameters and their corresponding response parameters of the pneumatic conveying system of bulk grain, and train a sparse nonlinear dynamics identification model according to the control parameters and their corresponding response parameters. The response parameters at least include the bulk grain breakage rate.

[0042] Please refer to Figure 2 , in practical applications, in the open-loop control mode, the control parameters and their corresponding response parameters can be collected from the pneumatic conveying system of bulk grain through corresponding sensors, and preprocessing such as noise removal and standardization is performed on the control parameters and response parameters to provide accurate basic data for subsequent modeling.

[0043] In the embodiment of the present application, the control parameters include the fan frequency and valve opening of the pneumatic conveying system of bulk grain. The control parameters are used to affect the momentum parameters of bulk grain particle conveying, and thus affect the breakage rate of bulk grain particles. In practical applications, the control parameters can also include external disturbance factors, such as environmental temperature and environmental humidity, and the external disturbance factors will also affect the response parameters during the bulk grain conveying process.

[0044] In the embodiment of the present application, the response parameters further include the momentum parameters of bulk grain particles, the pressure drop of the conveying pipeline, and the density of bulk grain particles in the pipeline. The response parameters reflect the dynamic behavior of the pneumatic conveying system of bulk grain, especially the force and motion state of bulk grain particles during the conveying process.

[0045] In the embodiment of the present application, training a sparse nonlinear dynamics identification model according to the control parameters and their corresponding response parameters includes: using the control parameters as input data, using the corresponding response parameters as output data, constructing a training data set, and training the sparse nonlinear dynamics identification model according to the training data set until the sparse nonlinear dynamics identification model converges.

[0046] In practical applications, nonlinear dynamics feature extraction is respectively performed on the control parameters and response parameters to construct a training data set, and then training data is obtained from the training data set and a suitable training data window is selected to train the sparse nonlinear dynamics identification model. During the training process, the model parameters are adjusted so that the model can accurately predict the response parameters of the pneumatic conveying system of bulk grain. After training, it is stored in the model library to provide model support for closed-loop control.

[0047] S2. When using the bulk grain pneumatic conveying system to convey bulk grain, at each moment, model predictive control is used to execute steps S21 to S23 at each moment.

[0048] S21. Based on the control parameters at the current moment and the sparse non-linear dynamics identification model, predict the response parameters at future moments.

[0049] S22. According to the response parameters at future moments, the set optimization objective and constraint conditions, use an optimization algorithm to solve and obtain the optimal control parameter sequence at the current moment, and the optimization objective is to minimize the bulk grain breakage rate.

[0050] S23. Adjust the control parameters of the bulk grain pneumatic conveying system to the first set of optimal control parameters in the optimal control parameter sequence.

[0051] It can be understood that the embodiments of the present application are based on model predictive control (MPC) to roll-optimize the control parameters of the bulk grain pneumatic conveying system. The core idea of MPC is to use a model to predict the system state in the future for a period of time, and then calculate the control output at the current moment according to these prediction results. This method not only considers the dynamic characteristics of the system, but also can introduce various constraint conditions in the control process, such as input and output limitations, etc.

[0052] In the embodiments of the present application, first, MPC predicts the response parameters of the bulk grain pneumatic conveying system at multiple future moments based on the sparse non-linear dynamics identification model (SINDYc). In the bulk grain pneumatic conveying system, the control objective is to reduce the breakage rate of bulk grain particles by adjusting parameters such as the air flow rate and system pressure of the bulk grain pneumatic conveying system. Therefore, MPC will set an optimization objective and related constraint conditions, and the optimization objective is to minimize the bulk grain breakage rate.

[0053] Then, MPC calculates an optimal control parameter sequence at each moment in a rolling optimization manner according to the output of the sparse non-linear dynamics identification model. Specifically, MPC will perform the following operations at each moment: using the sparse non-linear dynamics identification model, based on the current control parameters and response parameters, simulate the response parameters at several future moments, and predict the momentum change and breakage rate change of bulk grain particles. According to the prediction results, under the preset optimization objective and constraint conditions, use an optimization algorithm (such as quadratic programming, gradient descent, etc.) to calculate a control parameter sequence that can achieve the optimal objective. This optimization process will find a balance point to ensure that the bulk grain particle breakage rate is minimized without violating the constraint conditions.

[0054] Finally, through the above optimization process, the MPC generates an optimal control parameter sequence, and inputs the first element of the optimal control parameter sequence, i.e., the first set of optimal control parameters, into the bulk grain pneumatic conveying system, so that the bulk grain pneumatic conveying system is adjusted to the optimal control parameters, realizing the optimization of the control parameters.

[0055] At each moment, the response parameters of the bulk grain pneumatic conveying system are re-collected, the model input is updated, and prediction and optimization are repeated to achieve closed-loop control.

[0056] In the embodiment of the present application, the constraint conditions include the variable range of the fan frequency and the variable range of the valve opening of the bulk grain pneumatic conveying system. It also includes the variable range of the ambient temperature and the variable range of the ambient humidity of the bulk grain pneumatic conveying system.

[0057] In the embodiment of the present application, the method further includes: regularly using the verification data set to determine the error of the sparse non-linear dynamics identification model, and judging whether the error is greater than a preset error. If so, re-train the sparse non-linear dynamics identification model.

[0058] Specifically, in order to ensure the adaptability of the system under different operating conditions, the system regularly checks the error of the sparse non-linear dynamics identification model. If the error is greater than the preset error, a model that better conforms to the current conditions is selected from the model library or a new model is trained to ensure the control accuracy of the system under changing conditions.

[0059] In summary, the control method of the bulk grain pneumatic conveying system provided in this embodiment performs rolling prediction and rolling optimization on the bulk grain pneumatic conveying system based on SINDY-MPC, that is, at each moment, model predictive control predicts the response parameters of the bulk grain pneumatic conveying system at future moments based on the sparse non-linear dynamics identification model, solves the control parameters according to the prediction results with the minimum bulk grain breakage rate as the optimization goal, and then uses the solved control parameters to optimize the control of the bulk grain pneumatic conveying system. SINDY-MPC can capture complex non-linear dynamics, maintain high control performance under multi-input multi-output conditions, and can adapt to different working conditions in real time, thereby improving the response speed of control and the system robustness. Through precise model predictive control, the precise optimization of the control parameters of the bulk grain pneumatic conveying system is realized, and the bulk grain breakage rate during pneumatic conveying is significantly reduced.

[0060] Figure 3 The structural schematic diagram of a control device for a bulk grain pneumatic conveying system is shown. Please refer to Figure 3 , the device includes:

[0061] A training unit, configured to obtain multiple sets of control parameters and their corresponding response parameters of a bulk grain pneumatic conveying system, and train a sparse non-linear dynamics identification model according to the control parameters and their corresponding response parameters, where the response parameters at least include the bulk grain breakage rate;

[0062] A model predictive control unit, configured to, when conveying bulk grain by using the bulk grain pneumatic conveying system, at each moment, predict the response parameters at a future moment according to the control parameters at the current moment and based on the sparse non-linear dynamics identification model; and solve for the optimal control parameter sequence at the current moment by using an optimization algorithm according to the response parameters at the future moment and the set optimization objective and constraint conditions, where the optimization objective is to minimize the bulk grain breakage rate;

[0063] A parameter adjustment unit, configured to adjust the control parameters of the bulk grain pneumatic conveying system to the first set of optimal control parameters in the optimal control parameter sequence.

[0064] Figure 4 The structural schematic diagram of an electronic device is shown. Please refer to Figure 4 , where the electronic device includes a processor, a memory, and a communication bus;

[0065] The communication bus is used to implement the connection and communication between the processor and the memory;

[0066] The processor is configured to execute one or more programs in the memory to implement the steps of the control method of the bulk grain pneumatic conveying system as described in this embodiment.

[0067] It can be understood that since the control device and equipment of the bulk grain pneumatic conveying system described in this embodiment are used to implement the control method of the bulk grain pneumatic conveying system described in the embodiment, for the device and equipment disclosed in the embodiment, since they correspond to the method disclosed in the embodiment, the description is relatively simple. For the relevant parts, please refer to the partial description of the method, and details are not described herein again.

Claims

1. A control method for a pneumatic conveying system of bulk grain, characterized in that, The method includes: S1. Obtain multiple sets of control parameters and their corresponding response parameters of the bulk grain pneumatic conveying system, and train a sparse nonlinear dynamics identification model based on the control parameters and their corresponding response parameters. The response parameters at least include the bulk grain breakage rate; S2. When using the bulk grain pneumatic conveying system to convey bulk grain, perform the following operations at each moment by using model predictive control: S21. Predict the response parameters at future moments based on the control parameters at the current moment and the sparse nonlinear dynamics identification model; S22. Solve to obtain the optimal control parameter sequence at the current moment by using an optimization algorithm according to the response parameters at future moments and the set optimization objective and constraint conditions. The optimization objective is to minimize the bulk grain breakage rate; S23. Adjust the control parameters of the bulk grain pneumatic conveying system to the first set of optimal control parameters in the optimal control parameter sequence; The control parameters include the fan frequency and valve opening of the bulk grain pneumatic conveying system, and also include the ambient temperature and ambient humidity; The constraint conditions include the variable range of the fan frequency of the bulk grain pneumatic conveying system and the variable range of the valve opening, and also include the variable range of the ambient temperature and the variable range of the ambient humidity of the bulk grain pneumatic conveying system; The optimization algorithm adopts the quadratic programming algorithm or the gradient descent algorithm.

2. The control method of the pneumatic conveying system for bulk grain according to claim 1, characterized in that The response parameters also include the momentum parameters of the bulk grain particles, the pressure drop of the conveying pipeline, and the density of the bulk grain particles in the pipeline. The momentum parameters include velocity and acceleration.

3. The control method of the pneumatic conveying system for bulk grain according to claim 1, characterized in that, Training the sparse nonlinear dynamics identification model based on the control parameters and their corresponding response parameters includes: Taking the control parameters as input data and the corresponding response parameters as output data to construct a training data set, and training the sparse nonlinear dynamics identification model according to the training data set until the sparse nonlinear dynamics identification model converges.

4. The control method of the pneumatic conveying system for bulk grain according to claim 1, characterized in that, The method also includes: Regularly use the validation data set to determine the error of the sparse nonlinear dynamics identification model, and judge whether the error is greater than the preset error. If so, retrain the sparse nonlinear dynamics identification model.

5. A control device for a pneumatic conveying system of bulk grain, characterized in that, The device includes: A training unit, configured to obtain multiple sets of control parameters and their corresponding response parameters of the bulk grain pneumatic conveying system, and train a sparse nonlinear dynamics identification model based on the control parameters and their corresponding response parameters. The response parameters at least include the bulk grain breakage rate; A model predictive control unit, configured to, when using the bulk grain pneumatic conveying system to convey bulk grain, at each moment, predict the response parameters at future moments based on the control parameters at the current moment and the sparse nonlinear dynamics identification model; and solve to obtain the optimal control parameter sequence at the current moment by using an optimization algorithm according to the response parameters at future moments and the set optimization objective and constraint conditions. The optimization objective is to minimize the bulk grain breakage rate; A parameter adjustment unit, configured to adjust the control parameters of the bulk grain pneumatic conveying system to the first set of optimal control parameters in the optimal control parameter sequence; The control parameters include the fan frequency and valve opening of the bulk grain pneumatic conveying system, and also include the ambient temperature and ambient humidity; The constraint conditions include the variable range of the fan frequency and the variable range of the valve opening degree of the bulk grain pneumatic conveying system, and also include the variable range of the ambient temperature and the variable range of the ambient humidity of the bulk grain pneumatic conveying system; The optimization algorithm adopts a quadratic programming algorithm or a gradient descent algorithm.

6. An electronic device, characterized in that, The electronic device includes a processor, a memory, and a communication bus; The communication bus is used to realize the connection and communication between the processor and the memory; The processor is used to execute one or more programs in the memory to realize the steps of the control method of the bulk grain pneumatic conveying system as described in any one of claims 1 to 4.

Images