Parameter setting system and method of intelligent coal preparation density controller based on edge-cloud cooperation

Abstract

The application discloses an end-edge cloud cooperative intelligent coal density controller parameter setting system and method, realizes intelligent implementation on the original manual parameter setting mode depending on the experience of engineers, introduces an edge device, an industrial cloud server and an edge system, realizes automatic correction of parameters in the running process of the control system, avoids the problem of control performance decline caused by dynamic characteristic change in the traditional parameter setting method, and allows an operator to manually modify parameters or the system to automatically update control parameters according to the on-site situation, thereby reducing the work intensity of personnel system maintenance and improving the running stability of the system. Since the system collects full process data, the powerful storage capacity and computing power of the cloud are utilized to update the numerical twin model in real time, and the model accuracy is improved. The system is independent of the control system of the factory, is convenient to deploy, and does not introduce safety problems.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of industrial production process control, and particularly to an end-edge-cloud collaborative intelligent coal preparation density controller parameter setting system and method. BACKGROUND

[0002] Density control in the coal preparation process is one of the key control links in production. Since the coal preparation production process is always in frequent dynamic changes, the medium density often fluctuates, which makes it difficult for the coal preparation density controller to achieve the expected control effect. In the coal preparation density control process, the existing control method mainly relies on the PID controller, which does not require an accurate mathematical model and can achieve good control effect by relying on the on-site experiments of engineers. However, this parameter setting method seriously depends on the experience of engineers, which is time-consuming and laborious. In addition, since the dynamic characteristics of the industrial process are prone to change, the control effect of the PID controller cannot be kept optimal for a long time. SUMMARY

[0003] Therefore, the embodiments of the present application provide an end-edge-cloud collaborative intelligent coal preparation density controller parameter setting system and method.

[0004] According to one aspect of the present application, an end-edge-cloud collaborative intelligent coal preparation density controller parameter setting system is provided, comprising: an end-side system, an edge-side system and a cloud-side system, wherein the end-side system is deployed in an industrial site;

[0005] The end-side system comprises a PLC field control system, which is configured to perform process control on the industrial site by using a first intelligent coal preparation density controller according to the medium density control loop process data collected in the industrial site, and send the medium density control loop process data of the industrial site to the cloud-side system;

[0006] The cloud-side system comprises a first coal preparation density control process digital twin model, and the cloud-side system is configured to perform iterative updating on the first coal preparation density control process digital twin model according to the medium density control loop process data, and send the model parameters of the first coal preparation density control process digital twin model to the edge-side system when the second coal preparation density control process digital twin model in the edge-side system meets an updating condition;

[0007] The edge side system includes a PLC edge control system and an edge server, the edge side system including the PLC edge control system is a second intelligent coal density controller, the edge server includes a second coal density control process digital twin model, the edge side system is used for receiving the combined medium density control loop process data sent by the cloud side system, based on the combined medium density control loop process data and the second coal density control process digital twin model, the controller parameter of the second intelligent coal density control is optimized, and the controller parameter after optimization is sent to the end side system;

[0008] The end side system is also used for updating the first intelligent coal density controller based on the controller parameter after optimization sent by the edge side system;

[0009] The edge side system is also used for updating the second coal density control process digital twin model based on the received model parameter.

[0010] Optionally, the cloud side system includes an industrial big data storage server and an artificial intelligence computing platform, the artificial intelligence computing platform is deployed with the first coal density control process digital twin model, and the industrial big data storage server is used for storing the combined medium density control loop process data.

[0011] Optionally, the second coal density control process digital twin model and the first coal density control process digital twin model are represented as: Wherein y=[y(k-1), y(k-2),..., y(k-n)] is a density feedback value, u=[u(k-1), u(k-2),..., u(k-n)] is a shunt valve opening value, the density feedback value y(k) at k time is z -d b0u(k-1)+c, b0 and c are model parameters, d represents a delay time, z represents a z transform, u(k-1) represents a shunt valve opening value at k-1 time,

[0012] is a nonlinear dynamic compensation system output value, represents a nonlinear dynamic compensation system output at k time,

[0013] Kp, Ki are controller parameters, and e(k-1) represents a difference between a density set value and a density feedback value at k-1 time, q1(k) represents the combined medium level feedback value at time k, q2(k) represents the dilute medium density feedback value at time k, q3(k) represents the dilute medium level feedback value at time k, q4(k) represents the thick medium density feedback value at time k, q5(k) represents the water supply valve opening degree feedback value at time k, and n is the number of neurons of the second coal preparation density control process digital twin model and the first coal preparation density control process digital twin model.

[0014] Optionally, the edge-side system further comprises an edge server, and the edge server is deployed with the second coal preparation density control process digital twin model and a PID parameter correction algorithm. The edge-side system is specifically configured to use the second coal preparation density control process digital twin model and the second intelligent coal preparation density controller under the self-correcting mechanism as an environment based on the combined medium density control loop process data, run the PID parameter correction algorithm to obtain tuning parameters, and use the second coal preparation density control process digital twin model to determine the evaluation result of the tuning parameters. When the evaluation result meets the expectation, the second intelligent coal preparation density controller is controlled based on the tuning parameters.

[0015] Optionally, the edge-side system is further configured to verify the calculation error θ1(k) of the second coal preparation density control process digital twin model based on the combined medium density control loop process data, and determine that the second coal preparation density control process digital twin model reaches an updating condition when the calculation error θ1(k) is greater than a preset threshold value δ, wherein the calculation error θ1(k) is calculated according to the following formula: y0(k) represents the field density feedback value at time k.

[0016] Optionally, the end-side system comprises an edge device and a switch. The edge device is configured to collect the combined medium density control loop process data of an industrial field, and perform data transmission between the edge device and the PLC field control system based on a TCP protocol or an IP protocol through the switch. The edge device is configured to send the combined medium density control loop process data to an industrial big data storage server in the cloud-side system through a wireless communication module.

[0017] Optionally, the combined medium density control loop process data comprises a combined medium density, a shunt valve opening degree, a water supply valve start-stop signal, and a combined medium bucket liquid level.

[0018] According to another aspect of the present application, there is provided an end-edge-cloud collaborative intelligent coal preparation density controller parameter tuning method, which is applied to the end-edge-cloud collaborative intelligent coal preparation density controller parameter tuning system described above, and comprises the following steps:

[0019] The end-side system collects the combined medium density control loop process data of the industrial site, controls the process of the industrial site according to the combined medium density control loop process data by using the first intelligent coal density controller, and sends the combined medium density control loop process data of the industrial site to the cloud-side system;

[0020] The cloud-side system iteratively updates the first coal density control process digital twin model according to the combined medium density control loop process data, and sends the model parameters of the first coal density control process digital twin model to the edge-side system when the second coal density control process digital twin model in the edge-side system meets the update condition;

[0021] The edge-side system receives the combined medium density control loop process data sent by the cloud-side system, updates the second coal density control process digital twin model based on the received model parameters, optimizes the controller parameters of the second intelligent coal density controller based on the combined medium density control loop process data and the second coal density control process digital twin model, and sends the optimized controller parameters to the end-side system;

[0022] The end-side system updates the first intelligent coal density controller based on the optimized controller parameters sent by the edge-side system.

[0023] Optionally, the second coal density control process digital twin model and the first coal density control process digital twin model are represented as: Wherein y=[y(k-1), y(k-2), …, y(k-n)] is the density feedback value, u=[u(k-1), u(k-2), …, u(k-n)] is the shunt valve opening value, the density feedback value y(k) at k time is z -d b0u(k-1)+c, b0 and c are model parameters, d represents the delay time, z represents the z transform, u(k-1) represents the shunt valve opening value at k-1 time,

[0024] is the output value of the nonlinear dynamic compensation system, represents the nonlinear dynamic compensation system output at k time,

[0025] Kp, Ki are controller parameters, e(k-1) represents the difference between the density set value and the density feedback value at k-1 time, For the coal preparation density control process variables, q1(k) represents the combined medium liquid level feedback value at time k, q2(k) represents the dilute medium density feedback value at time k, q3(k) represents the dilute medium liquid level feedback value at time k, q4(k) represents the concentrated medium density feedback value at time k, q5(k) represents the water supply valve opening feedback value at time k, and n is the number of neurons in the second coal preparation density control process digital twin model and the first coal preparation density control process digital twin model.

[0026] Optionally, the controller parameters of the second intelligent coal preparation density controller are optimized based on the process data of the combined medium density control loop and the digital twin model of the second coal preparation density control process, including:

[0027] The side system uses the digital twin model of the second coal preparation density control process under the self-calibration mechanism and the second intelligent coal preparation density controller as the environment based on the process data of the combined medium density control loop. It runs a PID parameter correction algorithm to obtain the tuning parameters, and uses the digital twin model of the second coal preparation density control process to determine the evaluation result of the tuning parameters. When the evaluation result meets the expectations, the controller parameters of the second intelligent coal preparation density controller are optimized based on the tuning parameters.

[0028] Optionally, the cloud-side system includes an industrial big data storage server and an artificial intelligence computing platform. The artificial intelligence computing platform is deployed with a digital twin model of the first coal preparation density control process, and the industrial big data storage server is used to store the process data of the combined medium density control loop.

[0029] Optionally, the method further includes: the side-side system verifies the calculation error θ1(k) of the digital twin model of the second coal preparation density control process based on the process data of the combined medium density control loop, and determines that the digital twin model of the second coal preparation density control process has reached the update condition when the calculation error θ1(k) is greater than a preset threshold δ, wherein the calculation error y0(k) represents the field density feedback value at time k.

[0030] Optionally, the method further includes: the edge device collecting process data of the density control loop in the industrial field, and transmitting the data between the edge device and the PLC field control system via a switch based on the TCP or IP protocol; the edge device sending the process data of the density control loop to the industrial big data storage server in the cloud system via a wireless communication module.

[0031] Optionally, the process data of the combined medium density control loop includes the combined medium density, the opening degree of the diversion valve, the start / stop signal of the water supply valve, and the liquid level in the combined medium tank.

[0032] According to another aspect of the present application, a storage medium is provided, which stores a computer program, and the program is executed by a processor to implement the parameter setting method of the intelligent coal density controller in the end-edge-cloud collaborative manner.

[0033] According to another aspect of the present application, a computer device is provided, which comprises a storage medium, a processor, and a computer program stored in the storage medium and executable on the processor, and the processor implements the parameter setting method of the intelligent coal density controller in the end-edge-cloud collaborative manner when executing the program.

[0034] Through the close cooperation of the end side, the edge side and the cloud side, the data acquisition, processing, model training and parameter setting in the coal density control process are realized, the intelligent level and response speed of the control system are improved, the powerful capabilities of edge computing and cloud computing are utilized to realize the automatic setting and optimization of the coal density controller parameters, the disadvantages of manual intervention and experience-dependent parameter adjustment are reduced, the high precision and stability are achieved by real-time updating and correcting the cloud digital twin model, the accuracy and stability of the model are improved, and thus the precision and effect of the coal density control are ensured. The original manual parameter adjustment method depending on the experience of engineers is realized intelligently, the edge device, industrial cloud server and edge system are introduced, the automatic correction of parameters during the operation of the control system is realized, the problem of performance degradation of the traditional parameter setting method due to the change of dynamic characteristics is avoided, manual modification of parameters by the operator or automatic updating of control parameters by the system is allowed, the work intensity of personnel system maintenance is reduced, and the system operation stability is improved. Since the system collects full process data, the cloud has powerful storage capacity and computing power, the digital twin model is updated in real time, and the model accuracy is improved. The system is independent of the control system of the factory, is easy to deploy, and does not introduce security problems.

[0035] The above description is only a summary of the technical solutions of the present application, in order to more clearly understand the technical means of the present application, the specific embodiments of the present application can be implemented according to the content of the description, and in order to make the above and other purposes, characteristics and advantages of the present application more obvious and easy to understand, the following specific embodiments of the present application are described. BRIEF DESCRIPTION OF DRAWINGS

[0036] The drawings described herein are used to provide further understanding of the present application, and form a part of the present application. The schematic embodiments of the present application and their descriptions are used to explain the present application, and do not constitute an improper limitation on the present application. In the drawings:

[0037] Figure 1 A structure schematic diagram of a parameter setting system of an intelligent coal density controller in an end-edge-cloud collaborative manner provided by an embodiment of the present application is shown.

[0038] Figure 2 A terminal-edge cloud collaborative intelligent coal preparation density controller setting algorithm structure schematic diagram provided by the embodiment of the application is shown;

[0039] Figure 3 An intelligent coal preparation density controller parameter setting algorithm structure schematic diagram provided by the embodiment of the application is shown;

[0040] Figure 4 A terminal-edge cloud collaborative intelligent coal preparation density controller parameter setting method flow schematic diagram provided by the embodiment of the application is shown;

[0041] Figure 5 A terminal-edge cloud collaborative coal preparation density control process digital twin model algorithm logic schematic diagram provided by the embodiment of the application is shown;

[0042] Figure 6 An intelligent coal preparation density controller parameter setting algorithm logic schematic diagram driven by a digital twin model provided by the embodiment of the application is shown. DETAILED DESCRIPTION

[0043] The application will be described in detail below with reference to the accompanying drawings and in conjunction with the embodiments. It should be noted that the embodiments in the application and the features in the embodiments can be combined with each other without conflict.

[0044] In the embodiment, a terminal-edge cloud collaborative intelligent coal preparation density controller parameter setting system is provided, like Figure 1As shown, it comprises: an end-side system, a side-side system and a cloud-side system, the end-side system is deployed in an industrial site; the end-side system comprises a PLC field control system, the PLC field control system is used for process control of the industrial site according to the mixed medium density control loop process data collected in the industrial site by using a first intelligent coal density controller, and the mixed medium density control loop process data of the industrial site is sent to the cloud-side system; the cloud-side system comprises a first coal density control process digital twin model, the cloud-side system is used for iterative updating of the first coal density control process digital twin model according to the mixed medium density control loop process data, and when the second coal density control process digital twin model in the side-side system reaches an updating condition, the model parameters of the first coal density control process digital twin model are sent to the side-side system; the side-side system comprises a PLC edge control system and an edge server, the PLC edge control system is a second intelligent coal density controller, and the edge server comprises a second coal density control process digital twin model; the side-side system is used for receiving the mixed medium density control loop process data sent by the cloud-side system, optimizing the controller parameters of the second intelligent coal density controller based on the mixed medium density control loop process data and the second coal density control process digital twin model, and sending the optimized controller parameters to the end-side system; the end-side system is also used for updating the first intelligent coal density controller based on the optimized controller parameters sent by the side-side system; and the side-side system is also used for updating the second coal density control process digital twin model based on the received model parameters.

[0045] The cloud-side system comprises an industrial big data storage server and an artificial intelligence computing platform, the first coal density control process digital twin model is deployed on the artificial intelligence computing platform, and the industrial big data storage server is used for storing the mixed medium density control loop process data. The mixed medium density control loop process data comprises mixed medium density, shunt valve opening degree, water supply valve start-stop signal and mixed medium barrel liquid level. The end-side system comprises an edge device and a switch, the edge device is used for collecting the mixed medium density control loop process data of the industrial site, and data transmission is performed between the edge device and the PLC field control system through the switch based on a TCP protocol or an IP protocol; and the edge device sends the mixed medium density control loop process data to the industrial big data storage server in the cloud-side system through a wireless communication module.

[0046] In the above embodiment, the system architecture of the end-side cloud collaborative intelligent coal preparation density controller parameter tuning system specifically includes: an end-side system, a side-side system, and a cloud-side system. The end-side system is deployed in the industrial field and directly participates in the control of the actual production process. The end-side system includes a PLC field control system, which is composed of a PLC, an IO module, and a bottom layer device, and is responsible for the acquisition, processing, and control of field process data. The PLC acquires real-time data of the medium density control loop (such as medium density, shunt valve opening, water supply valve start-stop signal, and medium bucket liquid level) through the IO module, processes these data through the built-in first intelligent coal preparation density controller (usually a PID controller), calculates the control amount (such as shunt valve opening), and sends the control signal to the actuator (such as the shunt valve) through the IO module to complete the field process control. The PLC field control system sends the medium density control loop process data in the industrial field to the cloud-side system through a network (such as the TCP / IP protocol) for model updating and parameter tuning. The cloud-side system includes: an industrial big data storage server: used to store all process control data received from the edge device, which is stored in a database (such as a MySQL database) for subsequent data processing and model training. An artificial intelligence computing platform: this platform has strong data processing and computing capabilities, is responsible for preprocessing the stored data (such as filtering, denoising, and outlier removal), and runs the first coal preparation density control process digital twin model (i.e., the coal preparation density control process cloud twin model in the figure) of the coal preparation density control process, and continuously iterates and updates the model to improve its accuracy and adaptability. When the cloud-side system determines that the second coal preparation density control process digital twin model in the side-side system needs to be updated, it sends the latest model parameters (including model weights and biases) of the first coal preparation density control process digital twin model to the side-side system. The side-side system includes an edge server and a PLC edge control system (with a second intelligent coal preparation density controller). The PLC edge control system includes a control module consistent with the field PLC control system, as well as a second coal preparation density control process digital twin model and a parameter tuning algorithm running on the edge server. The PLC edge control system simulates the field environment to provide an experimental platform for the parameter tuning algorithm. The parameter tuning algorithm optimizes the controller parameters of the second intelligent coal preparation density controller based on the second coal preparation density control process digital twin model in the edge system. During the optimization process, the algorithm evaluates the effects of different parameter combinations through simulation, and finally determines the optimal controller parameters. The side-side system receives the medium density control loop process data sent by the cloud-side system through the network, and optimizes the second intelligent coal preparation density controller using these data and its own parameter tuning algorithm. After optimization, the optimized controller parameters are sent to the PLC field control system in the end-side system to improve the field control effect.

[0047] By applying the technical solution of the embodiment, through the close cooperation of the end side, the side side and the cloud side, the closed-loop optimization of data acquisition, processing, model training and parameter setting in the coal density control process is realized, the intelligent level and response speed of the control system are improved, the powerful capabilities of edge computing and cloud computing are utilized to realize the automatic setting and optimization of the coal density controller parameters, the disadvantages of manual intervention and reliance on experience parameter adjustment are reduced, and the high precision and stability: through real-time updating and correction of the cloud digital twin model, the accuracy and stability of the model are improved, thereby ensuring the precision and effect of the coal density control. The original manual parameter adjustment method relying on the experience of engineers is intelligently implemented, the edge device, industrial cloud server and edge system are introduced, the parameter automatic correction during the operation of the control system is realized, the problem of control performance degradation caused by dynamic characteristic changes in the traditional parameter setting method is avoided; the operating personnel can manually modify the parameters or the system automatically updates the control parameters according to the on-site situation, and the like, the personnel system maintenance work intensity is reduced, and the system operation stability is improved. Since the system collects full process data, the cloud has powerful storage capacity and computing power, the digital twin model is updated in real time, and the model accuracy is improved. The system is independent of the control system of the factory, is easy to deploy, and does not introduce security problems.

[0048] In the embodiments of the present application, the second coal density control process digital twin model and the first coal density control process digital twin model are represented as: Wherein, y = [y(k-1), y(k-2),..., y(k-n)] is the density feedback value, u = [u(k-1), u(k-2),..., u(k-n)] is the shunt valve opening value, the density feedback value y(k) at time k is z -d b0u(k-1)+c, b0 and c are model parameters, d represents the delay time, z represents z

[0049] Transform, u(k-1) represents the shunt valve opening value at time k-1,

[0050] is the nonlinear dynamic compensation system output value, represents the nonlinear dynamic compensation system output at time k,

[0051] K p , K i are controller parameters, e(k-1) represents the difference between the density set value and the density feedback value at time k-1, q1(k) represents the combined medium liquid level feedback value at time k, q2(k) represents the dilute medium density feedback value at time k, q3(k) represents the dilute medium liquid level feedback value at time k, q4(k) represents the thick medium density feedback value at time k, q5(k) represents the water supply valve opening degree feedback value at time k, and n is the number of neurons of the second coal preparation density control process digital twin model and the first coal preparation density control process digital twin model.

[0052] As shown in Figure 2 The end-edge cloud collaborative intelligent coal preparation density controller setting algorithm structure diagram provided by the embodiment of the application is shown. The field control system includes a PLC and field devices controlled by the PLC, and is responsible for real-time data acquisition and control instruction execution. The edge system includes an edge server and a PLC edge control system. The edge server runs a digital twin model and a parameter setting algorithm, and the PLC edge control system is consistent with the field control system and is used to verify the effect of the setting parameters. The cloud server includes an industrial big data storage server and an artificial intelligence computing platform. The former stores data, and the latter runs a cloud digital twin model and updates and corrects the model. Data transmission and communication are realized among the field control system, the edge system and the cloud server.

[0053] In the embodiment of the application, the edge system further includes an edge server, the edge server is deployed with the second coal preparation density control process digital twin model and a PID parameter correction algorithm, and the edge system is specifically used for obtaining the setting parameters by using the second coal preparation density control process digital twin model and the second intelligent coal preparation density controller as an environment under a self-correction mechanism to run a PID parameter correction algorithm based on the combined medium density control loop process data, determining the evaluation result of the setting parameters by using the second coal preparation density control process digital twin model, and performing controller parameter optimization on the second intelligent coal preparation density controller based on the setting parameters when the evaluation result meets the expectation.

[0054] In this embodiment, the edge system is generally deployed close to the industrial field, but is connected to the cloud system through a network, including an edge server and edge devices connected to the PLC. The second intelligent coal preparation density control process edge digital twin model and the parameter setting algorithm are run in the edge server. The parameter setting algorithm interacts with the PLC edge control system as an environment, sends the obtained parameters to the PLC edge control system, controls the digital twin model to obtain the parameter evaluation result, and feeds back to the parameter setting algorithm. When a set of good parameters are obtained, the edge system sends the parameters to the edge devices to improve the field control effect. As shown in Figure 3A structure diagram of a parameter setting algorithm of an intelligent coal selection density controller provided in an embodiment of the application is shown. Network parameter and weight updating algorithm: this part describes how to update the parameters and weights of the neural network. In the Actor-Critic method, there are two different neural networks: the policy network (Actor network) and the value network (Critic network), and the parameters of these two networks will be updated as the training progresses in order to improve the agent's ability to make decisions in the environment. Policy function (Actor network): the policy network (Actor network) is responsible for generating actions, and its goal is to learn a policy function that maps states to a probability distribution of actions. At each time step, the policy network outputs a probability distribution based on the current state, and then the agent selects an action to execute. By observing the reward and the new state, the parameters of the policy network are adjusted to improve the policy. Value function (Critic network): the value network (Critic network) is responsible for evaluating the value of the state or the quality of the policy, and it learns a value function that estimates the expected cumulative return of a particular policy in a certain state. The output of the value network can guide the optimization of the policy network, as it provides information about which states are more advantageous. In the Actor-Critic method, the policy network and the value network interact. The policy network tries to maximize the value given by the value network, while the value network tries to accurately predict the value of the policy generated by the policy network. By alternately optimizing these two networks, the agent can gradually improve its behavior in the environment, thereby obtaining higher cumulative rewards. After finding a set of optimal setting parameters, they are sent to the coal selection density edge control system (i.e. the PLC edge control system) for synchronization to the PLC field control system.

[0055] In an embodiment of the application, the side system is also used to verify the calculation error θ1(k) of the second coal selection density control process digital twin model based on the combined medium density control loop process data, and determine that the second coal selection density control process digital twin model reaches the updating condition when the calculation error θ1(k) is greater than a preset threshold δ, wherein the calculation error y0(k) represents the field density feedback value at time k.

[0056] In this embodiment, based on the real-time acquisition of the combined medium density control loop process data from the field control system (such as PLC), the second coal preparation density control process digital twin model under the self-correcting mechanism and the second intelligent coal preparation density controller as the control environment are used to run the PID parameter correction algorithm. Through the operation of the algorithm, the optimal PID controller parameters after tuning are obtained. During operation, the side system will continuously verify the error between the calculation output of the second coal preparation density control process digital twin model and the actual process data based on the latest combined medium density control loop process data. By comparing the difference between the calculation output of the model and the actual measurement value, the calculation error of the model is evaluated. When the calculation error is greater than the preset threshold, the side system determines that the accuracy of the second coal preparation density control process digital twin model no longer meets the current control requirements, reaching the update condition. At this time, the side system will request the cloud side system for new model weight and bias parameters in order to update the digital twin model and maintain the accuracy and effectiveness of the model. Through the design and function realization of the above-mentioned side system, the intelligent coal preparation density controller parameter tuning system can adjust and optimize the control parameters in real time according to the field data, while ensuring the accuracy and reliability of the digital twin model, thereby improving the control effect and stability of the entire coal preparation process.

[0057] Further, as a refinement and extension of the above embodiment, in order to fully describe the specific implementation process of the embodiment, an end-side cloud collaborative intelligent coal preparation density controller parameter tuning method is provided, as shown in Figure 4 The method comprises:

[0058] Step 401, the end-side system collects the combined medium density control loop process data of the industrial field, and uses the first intelligent coal preparation density controller to perform process control on the industrial field according to the combined medium density control loop process data, and sends the combined medium density control loop process data of the industrial field to the cloud-side system;

[0059] Step 402, the cloud-side system iteratively updates the first coal preparation density control process digital twin model according to the combined medium density control loop process data, and sends the model parameters of the first coal preparation density control process digital twin model to the side system when the second coal preparation density control process digital twin model in the side system reaches the update condition;

[0060] Step 403, the side system receives the combined medium density control loop process data sent by the cloud-side system, updates the second coal preparation density control process digital twin model based on the received model parameters, optimizes the controller parameters of the second intelligent coal preparation density controller based on the combined medium density control loop process data and the second coal preparation density control process digital twin model, and sends the optimized controller parameters to the end-side system;

[0061] Step 404, the end-side system updates the first intelligent coal preparation density controller based on the optimized controller parameters sent by the side-side system.

[0062] In this embodiment, the end-side cloud collaborative intelligent coal preparation density controller parameter setting method is applied to the end-side cloud collaborative intelligent coal preparation density controller parameter setting system as described above. For specific explanations, please refer to the content about the end-side cloud collaborative intelligent coal preparation density controller parameter setting system in the foregoing, which will not be repeated here.

[0063] As shown in Figure 5 The logic diagram of the end-side cloud collaborative coal density control process twin model algorithm. The specific process is as follows: edge device collects coal density control process related process data: the edge device in the production field collects real-time data related to coal density control. Data is stored in an industrial big data storage server: the collected data is transmitted and stored in an industrial big data storage server for subsequent analysis and processing. Artificial intelligence computing platform acquires data and pre-processes data: the artificial intelligence computing platform acquires these data from the storage server and performs preliminary cleaning, sorting and formatting to facilitate model training and analysis. Cloud updates coal density control process cloud twin model: based on the pre-processed data, the cloud-side artificial intelligence model (cloud twin model) is updated, which is a continuous process aimed at reflecting changes in the actual production process. Self-correcting mechanism evaluates coal density control process cloud twin model and edge twin model: the self-correcting mechanism periodically evaluates the accuracy of the cloud twin model and the edge twin model to ensure the accuracy of the model. Model parameters are transmitted to the edge device: if the edge twin model meets the accuracy requirements, the model parameters of the cloud twin model are sent to the edge server, so that the edge server can update the edge twin model according to the latest model parameters.

[0064] As shown in Figure 6As shown, it is a digital twin model driven coal density controller parameter intelligent setting algorithm logic diagram. The specific process is as follows: initializing model parameters: first, randomly initialize model parameters as the starting point of training. Obtain the initial state of the edge number twin model of the coal density control process: according to the initialized model parameters, obtain the initial state of the edge number twin model. Obtain initial parameters: obtain the initial controller parameters for controlling the coal density from the edge device. Run the edge number twin model for k times in the coal density control process: run the edge number twin model using the above initial parameters to simulate one cycle of the coal density control process. Evaluate the parameter control effect: according to the running result, evaluate the effect of the current parameters on the coal density control. The accuracy meets: if the accuracy of the model meets the predetermined standard, the newly obtained model parameters are transmitted to the edge device for actual coal density control process. Otherwise, store the results in the cache, check whether the cache is full, if not, determine new controller parameters; if full, randomly extract a certain number (for example, m) of results from the cache, train the model parameters with these results, and determine new controller parameters. The whole process embodies the characteristics of iterative learning, and through continuous simulation and feedback, the model parameters are continuously optimized to improve the control effect of the coal density control process.

[0065] The embodiment of the present application also provides a computer device, which can be a personal computer, a server, a network device, etc. The computer device comprises a bus, a processor, a memory and a communication interface, and can further comprise an input / output interface and a display device. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device comprises a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store location information. The network interface of the computer device is used to communicate with external terminals through network connection. The computer program is executed by the processor to implement the steps in each method embodiment.

[0066] Those skilled in the art can understand that the structure of the computer device described above is only part of the structure related to the scheme of the present application, and does not constitute a limitation on the computer device to which the scheme of the present application is applied. The specific computer device can include more or fewer components, or combine certain components, or have a different component arrangement.

[0067] In one embodiment, a computer readable storage medium is provided, which can be non-volatile or volatile, and has a computer program stored thereon. The computer program is executed by the processor to implement the steps in each method embodiment described above.

[0068] In an embodiment, a computer program product is provided, comprising a computer program which, when executed by a processor, implements the steps of any of the above method embodiments.

[0069] It should be noted that the user information (including but not limited to user equipment information, user personal information, etc.) and data (including but not limited to data for analysis, stored data, displayed data, etc.) involved in the present application are all information and data authorized by the user or authorized by all parties.

[0070] It can be understood by those skilled in the art that all or part of the processes in the above-mentioned embodiments can be completed by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer readable storage medium. When the computer program is executed, it can include the processes of the above-mentioned embodiments. Any reference to memory, database or other medium used in the embodiments provided by the present application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical storage, high-density embedded non-volatile memory, resistive memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration but not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc. The database involved in the embodiments provided by the present application can include at least one of a relational database and a non-relational database. The non-relational database can include a distributed database based on a block chain, etc., without limitation. The processor involved in the embodiments provided by the present application can be a general-purpose processor, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., without limitation.

[0071] Any technical features in the above embodiments can be combined, and for the sake of brevity, not all possible combinations are described above, however, as long as the combinations of technical features do not have contradictions, they shall be considered within the scope of the present disclosure.

[0072] The above embodiments only express several implementation manners of the present application, and the description is relatively specific and detailed, but it shall not be understood as a limitation on the patent scope of the present application. It should be pointed out that, for ordinary skilled persons in the art, several modifications and improvements can be made without departing from the concept of the present application, and these shall be within the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the appended claims.

Claims

1. An end-to-end cloud collaborative intelligent coal preparation density controller parameter setting system, characterized in that, The application relates to an intelligent coal preparation density control system, which comprises an end-side system, a side-side system and a cloud-side system, wherein the end-side system is deployed in an industrial field. The end-side system comprises a PLC field control system, which is used for performing process control on the industrial field according to collected medium density control loop process data of the industrial field by using a first intelligent coal preparation density controller, and sending the medium density control loop process data of the industrial field to the cloud-side system. The cloud-side system comprises a first coal preparation density control process digital twin model, which is used for iteratively updating the first coal preparation density control process digital twin model according to the medium density control loop process data, and sending model parameters of the first coal preparation density control process digital twin model to the side-side system when a second coal preparation density control process digital twin model in the side-side system meets an updating condition. The side-side system comprises a PLC edge control system and an edge server, wherein the PLC edge control system is a second intelligent coal preparation density controller, and the edge server comprises a second coal preparation density control process digital twin model. The side-side system is used for receiving the medium density control loop process data sent by the cloud-side system, performing controller parameter tuning on the second intelligent coal preparation density controller based on the medium density control loop process data and the second coal preparation density control process digital twin model, and sending the tuned controller parameters to the end-side system. The end-side system is further used for updating the first intelligent coal preparation density controller based on the tuned controller parameters sent by the side-side system. The side-side system is further used for updating the second coal preparation density control process digital twin model based on the received model parameters. The second coal preparation density control process digital twin model and the first coal preparation density control process digital twin model are represented as: wherein, is a density feedback value, is a split valve gate opening value, the density feedback value at time k , and are model parameters, d represents a delay time, z represents a z transform, and u(k-1) represents a split valve gate opening value at time k-1, is a nonlinear dynamic compensation system output value, represents a nonlinear dynamic compensation system output at time k, , K p , K i are controller parameters, and e(k-1) represents a difference between a density set value and the density feedback value at time k-1, is a coal preparation density control process variable, q1(k) represents a combined medium level feedback value at time k, q2(k) represents a dilute medium density feedback value at time k, q3(k) represents a dilute medium level feedback value at time k, q4(k) represents a thick medium density feedback value at time k, q5(k) represents a water supply valve opening feedback value at time k, and n is a number of neurons of the second coal preparation density control process digital twin model and the first coal preparation density control process digital twin model.

2. The end-to-end cloud synergy intelligent coal preparation density controller parameter setting system according to claim 1, characterized in that, The cloud-side system comprises an industrial big data storage server and an artificial intelligence computing platform, wherein the artificial intelligence computing platform is deployed with the first coal preparation density control process digital twin model, and the industrial big data storage server is used for storing the medium density control loop process data.

3. The end-to-end cloud synergy based intelligent coal preparation density controller parameter tuning system according to claim 1, wherein, The side-side system further comprises an edge-end server, which is deployed with the second coal preparation density control process digital twin model and a PID parameter correction algorithm. The side-side system is specifically used for obtaining tuning parameters by using the second coal preparation density control process digital twin model and the second intelligent coal preparation density controller under a self-correction mechanism as an environment to run the PID parameter correction algorithm based on the medium density control loop process data, determining an evaluation result of the tuning parameters by using the second coal preparation density control process digital twin model, and performing controller parameter tuning on the second intelligent coal preparation density controller based on the tuning parameters when the evaluation result meets an expectation.

4. The end-to-end cloud synergy based intelligent coal preparation density controller parameter tuning system according to claim 3, characterized in that, The side system is also used to verify the calculation error of the second coal separation density control process digital twin model based on the combined medium density control loop process data , and when the calculation error is greater than a preset threshold , it is determined that the second coal separation density control process digital twin model reaches an updating condition, wherein the calculation error , y0(k) represents the on-site density feedback value at time k.

5. The end-to-end cloud coordinated intelligent coal preparation density controller parameter tuning system of claim 1, wherein, The end-side system comprises an edge device and a switch, the edge device is used to collect the process data of the mixed medium density control loop of the industrial field, and the data transmission is performed between the edge device and the PLC field control system through the switch based on the TCP protocol or the IP protocol; the edge device sends the process data of the mixed medium density control loop to the industrial big data storage server in the cloud-side system through a wireless communication module.

6. The end-to-end cloud synergy based intelligent coal preparation density controller parameter tuning system according to claim 1, wherein, The process data of the mixed medium density control loop comprises the mixed medium density, the shunt valve opening degree, the water supply valve start-stop signal, and the mixed medium barrel liquid level.

7. An edge-cloud collaborative intelligent coal preparation density controller parameter tuning method, characterized in that, The application is applied to the intelligent coal preparation density controller parameter setting system of the end-edge-cloud cooperation as claimed in any one of claims 1 to 6, comprising: The end-side system collects the process data of the mixed medium density control loop of the industrial field, uses the first intelligent coal preparation density controller to perform process control on the industrial field according to the process data of the mixed medium density control loop, and sends the process data of the mixed medium density control loop of the industrial field to the cloud-side system; The cloud-side system iteratively updates the first coal preparation density control process digital twin model according to the process data of the mixed medium density control loop, and when the second coal preparation density control process digital twin model in the edge-side system meets the update condition, sends the model parameters of the first coal preparation density control process digital twin model to the edge-side system; The edge-side system receives the process data of the mixed medium density control loop sent by the cloud-side system, updates the second coal preparation density control process digital twin model based on the received model parameters, optimizes the controller parameters of the second intelligent coal preparation density controller based on the process data of the mixed medium density control loop and the second coal preparation density control process digital twin model, and sends the optimized controller parameters to the end-side system; The end-side system updates the first intelligent coal preparation density controller based on the optimized controller parameters sent by the edge-side system.

8. The method of claim 7, wherein, The second coal preparation density control process digital twin model and the first coal preparation density control process digital twin model are represented as: wherein, is a density feedback value, is a split valve gate opening value, the density feedback value at time k , and are model parameters, d represents a delay time, z represents a z transform, u(k-1) represents the split valve gate opening value at time k-1, is a nonlinear dynamic compensation system output value, represents the nonlinear dynamic compensation system output at time k, , Kp, Ki are controller parameters, e(k-1) represents a difference between the density set value and the density feedback value at time k-1, is a coal preparation density control process variable, q1(k) represents a clean medium level feedback value at time k, q2(k) represents a dilute medium density feedback value at time k, q3(k) represents a dilute medium level feedback value at time k, q4(k) represents a thick medium density feedback value at time k, q5(k) represents a water supply valve opening feedback value at time k, and n is a number of neurons of the second coal preparation density control process digital twin model and the first coal preparation density control process digital twin model.

9. The method of claim 8, wherein, The optimization of the controller parameters of the second intelligent coal preparation density controller based on the process data of the mixed medium density control loop and the second coal preparation density control process digital twin model comprises: The edge-side system uses the second coal preparation density control process digital twin model and the second intelligent coal preparation density controller under the self-correcting mechanism as the environment, runs the PID parameter correction algorithm to obtain the setting parameters based on the process data of the mixed medium density control loop, and uses the second coal preparation density control process digital twin model to determine the evaluation result of the setting parameters, and when the evaluation result meets the expectation, optimizes the controller parameters of the second intelligent coal preparation density controller based on the setting parameters.

Images