Centralized control method and system

Abstract

The invention discloses a centralized control method and system, applied to first control devices in the centralized control system, one first control device is correspondingly deployed in each device matrix, and a second control device is connected with all the first control devices; generating a description file, and sending the description file to a second control device when the second control device is connected for the first time, so that the second control device performs centralized control template configuration based on the description file; collecting monitoring data corresponding to the equipment matrix according to the target collection frequency, and sending the monitoring data to second control equipment according to a subscription mechanism to train a neural network model; receiving an operation instruction returned by the second control equipment and a trained neural network model; and inputting the operation instruction into the neural network model to obtain a target operation, and executing the target operation. First control equipment is arranged for a square matrix, a square matrix level neural network model operation platform is provided, modular construction and operation and maintenance of a plant station are achieved, the complexity of construction and operation and maintenance of a control system is reduced, and the operation performance and reliability of the control system are improved.

Description

Technical Field

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

[0002] In related technologies, devices are usually deployed in arrays, and multiple arrays are directly connected to a centralized control station to centrally monitor, operate, and manage multiple geographically dispersed devices. Figure 1 As shown, the centralized control station is connected to Matrix 1, Matrix 2, and Matrix 3. However, when all Matrixes are directly connected to the centralized control station, the centralized control station needs to process data from each Matrix, perform monitoring tasks, and make real-time adjustments. As the number of Matrixes increases, this leads to a sharp increase in the workload of the centralized control station, increasing the complexity of maintaining the centralized control system. Summary of the Invention

[0003] In view of this, the present application provides a centralized control method and system for solving the problem of high complexity in maintaining the centralized control system in the prior art.

[0004] The purpose of this application can be achieved through the following technical solutions: A first aspect of the present application is to provide a centralized control method, wherein a first control device is deployed in each device array, and a second control device is connected to all first control devices. The centralized control method includes: Generate a description file and send the description file to the second control device when connecting to the second control device for the first time, so that the second control device configures a centralized control template based on the description file, and enters the operation mode after the centralized control template configuration is completed, where the centralized control template includes at least a device matrix and a connection relationship between the first control device and the second control device; collecting first monitoring data corresponding to the device array according to a target collection frequency; sending the first monitoring data to the second control device according to a subscription mechanism, so that the second control device trains the first neural network model based on the first monitoring data, the first neural network model being configured in the second control device; receiving an operation instruction and a trained first neural network model returned by a second control device based on the first monitoring data; Inputting the operation instruction into the trained first neural network model received by the first control device to obtain a target operation; Execute the target action.

[0005] In an optional embodiment, the description file includes at least the following information: a list of measurement points, the IP address of the first control device, the device number, the device name, the device location, and the device type, device location, number of devices, and rated power of the devices included in the device array connected to the first control device.

[0006] In an optional embodiment, the first control device includes a pre-trained second neural network model and further includes: Inputting the first monitoring data into the second neural network model to obtain second monitoring data corresponding to the device matrix; Accordingly, sending the first monitoring data to the second control device according to the subscription mechanism includes: The first monitoring data and the second monitoring data are sent to the second control device according to the subscription mechanism.

[0007] In an optional embodiment, the first neural network model is configured in the second control device in the form of a graphical interface, and the first monitoring data is sent to the second control device according to a subscription mechanism, so that the second control device trains the first neural network model based on the first monitoring data, including: The first monitoring data is sent to the second control device according to the subscription mechanism, so that the second control device switches the first neural network model to a training mode after completing the configuration of the first neural network model, and trains the first neural network model using the first monitoring data as a sample.

[0008] In an optional embodiment, the method further includes: In a case where a first neural network model is already running in the first control device, the running first neural network model is replaced with the received trained first neural network model.

[0009] In an optional embodiment, obtaining the target acquisition frequency further includes: When collecting the first monitoring data for the first time, determining the first preset initial frequency as the target collection frequency; In the case where the first monitoring data is not collected for the first time, a historical PID control output and a historical collection frequency are obtained, where the historical PID control output is the PID control output corresponding to the historically collected first monitoring data, and the historical collection frequency is the collection frequency corresponding to the historically collected first monitoring data; According to the historical PID control output and the historical acquisition frequency, the first acquisition frequency is calculated; The first acquisition frequency is determined as the target acquisition frequency.

[0010] In an optional embodiment, obtaining the target acquisition frequency further includes: Get historical CPU usage parameters; Calculate the CPU limit parameter based on the CPU performance parameter and the CPU maximum occupancy limit parameter; if the historical CPU usage parameter exceeds the CPU limit parameter, determine the second preset acquisition frequency as the target acquisition frequency; When the historical CPU usage parameter does not exceed the CPU limit parameter, a second collection frequency is calculated based on the first collection frequency, the second preset collection frequency, the historical CPU usage parameter, and the CPU limit parameter; The second acquisition frequency is determined as the target acquisition frequency.

[0011] In an optional embodiment, sending the first monitoring data to the second control device according to the subscription mechanism includes: receiving subscription information sent by the second control device; Determine the subscription mechanism based on the subscription information; The first monitoring data is sent to the second control device in real time according to the subscription mechanism.

[0012] In an optional embodiment, the subscription information includes at least information on the maximum data volume of the first monitoring data uploaded at a single time, whether it is compressed, and whether it is encrypted.

[0013] A second aspect of the present application is to provide a centralized control system, comprising at least one device matrix, a first control device and a second control device, wherein each device matrix has a corresponding first control device, and the second control device is connected to all first control devices; the device matrix includes multiple devices; The first control device includes: a generation module, configured to generate a description file and, when connected to a second control device for the first time, send the description file to the second control device, so that the second control device configures a centralized control template based on the description file, and enters an operation mode after the centralized control template configuration is completed, wherein the centralized control template at least includes a device matrix, a connection relationship between the first control device and the second control device; An acquisition module, configured to acquire first monitoring data corresponding to the device array according to a target acquisition frequency; a sending module, configured to send the first monitoring data to the second control device according to a subscription mechanism, so that the second control device trains the first neural network model based on the first monitoring data, the first neural network model being configured in the second control device; A receiving module, configured to receive an operation instruction and a trained first neural network model returned by the second control device based on the first monitoring data; An input module, configured to input an operation instruction into the trained first neural network model received by the first control device to obtain a target operation; Execution module, used to execute target operations.

[0014] The third aspect of the present application is to provide an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the method of the first aspect when executing the computer program.

[0015] The fourth aspect of the present application is to provide a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, it implements the method of the first aspect.

[0016] Compared with the existing technology, the centralized control method provided by this application is applied to the first control device in the centralized control system. A first control device is deployed in each device matrix, and the second control device is connected to all the first control devices. A description file is generated and sent to the second control device when it is first connected to it, so that it can configure the centralized control template based on the description file. The monitoring data corresponding to the device matrix is ​​collected according to the target collection frequency and sent to the second control device according to the subscription mechanism to train the neural network model. The operation instructions and the trained neural network model returned by the second control device are received. The operation instructions are input into the neural network model to obtain the target operation and execute the target operation. The matrix is ​​equipped with a first control device and an matrix-level neural network model operation platform is provided to realize modular construction and operation and maintenance of the plant and station, reduce the complexity of control system construction and operation and maintenance, and improve its computing performance and reliability. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0018] Figure 1 A structural block diagram of a centralized control system provided for related technologies; Figure 2 A schematic structural block diagram of a centralized control system provided in an embodiment of the present application; Figure 3 Another structural block diagram of the centralized control system provided in an embodiment of the present application; Figure 4 A schematic diagram of a flow chart of the centralized control method provided in an embodiment of the present application; Figure 5 A structural block diagram of an electronic device for implementing a centralized control method provided in an embodiment of the present application. DETAILED DESCRIPTION

[0019] In order to enable those skilled in the art to better understand the present invention, the following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts should fall within the scope of protection of this application.

[0020] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequential order. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in a sequence other than those illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0021] It should be understood that in the embodiments of the present application, "at least one" refers to one or more, and "more than one" refers to two or more. "And / or" is merely a description of the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. The character " / " generally indicates that the previous and subsequent associated objects are in an "or" relationship. "Including A, B and / or C" means including any one, any two, or any three of A, B, and C.

[0022] It should be understood that in the embodiments of the present application, "B corresponding to A," "B corresponding to A," "A corresponds to B," or "B corresponds to A" means that B is associated with A and B can be determined based on A. Determining B based on A does not mean determining B based solely on A; B can also be determined based on A and / or other information.

[0023] In order to solve the technical problems existing in the related art, the embodiment of the present application provides a centralized control method, which is applied to Figure 2In the centralized control system shown, the centralized control system includes a second control device, a first control device 1, a first control device 2, a first control device 3, a device matrix 1, a device matrix 2, and a device matrix 3. Device matrix 1 is connected to the second control device via the first control device 1, device matrix 2 is connected to the second control device via the first control device 2, and device matrix 3 is connected to the second control device via the first control device 3. Each device matrix includes multiple devices.

[0024] In an optional embodiment, a high-performance processor is provided in the first control device so that large amounts of data can be processed subsequently. For example, the main frequency of the processor can be 2.5 GHz, and the power consumption is less than 40 W. A large-capacity hard disk can be provided in the first control device to store historical data, for example, a 2 TB hard disk is provided. A larger running memory is configured for the first control device, for example, a 16 GB running memory is configured. Multiple Ethernet ports and serial ports are provided in the first control device so that it can be connected and communicated with multiple other devices in the future, thereby adapting to different industrial control and data acquisition requirements. The Linux system is run on the first control device to meet the actual needs of the user. The user is allowed to interact with the first control device through operating methods such as a graphical user interface and a command line, so as to facilitate non-technical personnel while being able to adapt to complex tasks.

[0025] It should be noted that the first control device can run different communication protocols to communicate with the device. For example, the communication protocol includes at least one of the following protocols: TCP (Transmission Control Protocol), OPC UA (Open Production Control and Unified Architecture, Opto-Autonomous Intelligence and Cognitive Control), and BACnet (Building Automation and Control Networks). In addition, the communication protocol may also be other protocols, which are not limited in this application.

[0026] In a more specific embodiment, Figure 3As shown, the first control device in the centralized control system is equipped with a collection service function, a second neural network, a processing service function, a communication function, a processor, and a database. The collection service function is used to collect first monitoring data corresponding to the device array at a target collection frequency; the second neural network is used to obtain second monitoring data corresponding to the device array; the processing service function is used to generate target operations based on the operation instructions returned by the second control device; the communication function is used to communicate with the second control device; the processor is used to perform tasks such as data processing, control logic execution, and communication management; and the database is used to store data, save parameters, and log records.

[0027] In a more specific embodiment, if the centralized control system is used in a photovoltaic power station, the device array in the centralized control system includes, but is not limited to, the following devices: inverters and combiner boxes. The number of inverters and combiner boxes can be n, where n is a positive integer.

[0028] Next, the data configuration method applied to the first control device provided in the embodiment of the present application will be explained in detail: like Figure 4 As shown, Figure 4 This is a flowchart of an example of a centralized control method provided in an embodiment of the present application. It should be noted that the steps shown can be performed in a different logical order than that shown in the flowchart of the method. The method may include the following steps S401 to S406.

[0029] Step S401: Generate a description file and send the description file to the second control device when connecting to the second control device for the first time, so that the second control device configures the centralized control template based on the description file and enters the operation mode after the centralized control template configuration is completed.

[0030] It should be noted that the centralized control template at least includes a device matrix and a connection relationship between a first control device and a second control device.

[0031] In an optional embodiment, a device array refers to devices arranged in a square array. In an optional embodiment, if the centralized control method is applied to a photovoltaic power station, the devices refer to solar photovoltaic modules, and the device array refers to a photovoltaic array. If the centralized control method is applied to a wind power station, the devices refer to wind turbines, and the device array refers to wind turbines arranged in a square array.

[0032] In an optional embodiment, the description file includes at least the following information: a list of measurement points, the IP address of the first control device, the device number, the device name, the device location, and the device type, device location, number of devices, and rated power of the devices included in the device array connected to the first control device.

[0033] Interact with the secondary control device through both a graphical user interface and command line, and quickly deploy the primary control device by configuring a centralized control template. Simply modify the array name to batch modify the device names and measurement point names of all devices in the array.

[0034] In another specific embodiment, for a newly added device matrix, it is only necessary to configure a first control device for it and connect the first control device to a second control device, so that the newly added device matrix can be added to the centralized control system, thereby achieving rapid introduction and expansion.

[0035] Step S402: collecting first monitoring data corresponding to the device array according to the target collection frequency.

[0036] In an optional embodiment, taking the application of the centralized control method in a photovoltaic power station as an example, the first monitoring data may include at least one of the following data: the output power of the generator, the utilization rate of the generator, the operating temperature of components such as the generator, the intensity of solar radiation, etc.

[0037] In an optional embodiment, an intelligent data acquisition system may be configured in the first control device, and the first monitoring data may be acquired by utilizing the intelligent data acquisition system.

[0038] In an optional embodiment, a pre-trained second neural network model is configured in the first control device, and the centralized control method provided in the embodiment of the present application further includes: The first monitoring data is input into the pre-trained second neural network model to obtain second monitoring data corresponding to the device matrix.

[0039] In an optional embodiment, the first monitoring data may include at least one of the following data: operating temperature of components such as the generator, and solar radiation intensity. The second monitoring data may include at least predicted output power.

[0040] It should be noted that the monitoring data can be adaptively changed according to the application scenario of the centralized control method, and this application does not limit this. In this embodiment, the first monitoring data is comprehensively analyzed using advanced machine learning algorithms to more accurately predict the output power of the generator. Based on accurate power prediction, the second control device can better plan power generation resources, optimize scheduling strategies, and ensure a balance between power supply and power demand. It can identify factors that may affect power generation (such as severe weather conditions) in advance, allowing dispatchers to have sufficient time to take measures to maintain grid stability.

[0041] Next, the training process of the pre-trained second neural network model is described in detail: obtaining multiple first monitoring data samples and second monitoring data labels corresponding to the first monitoring data samples; inputting the multiple first monitoring data samples into the second neural network model to obtain predicted second monitoring data corresponding to the multiple first monitoring data samples; and adjusting the model parameters of the second neural network model based on the difference between the predicted second monitoring data and the second monitoring data labels. The model training process in this application is similar to the model training process in the prior art and will not be described in detail in this application.

[0042] In an optional embodiment, for monitoring data that changes steadily, it is often only necessary to use a lower acquisition frequency for collection, while for monitoring data that changes dramatically, it is necessary to use a higher acquisition frequency for collection. High-frequency collection of problem monitoring data helps to discover potential problems or abnormal trends in advance so that the problems can be dealt with in a timely manner later. Low-frequency collection of problem-free data can reduce the amount of data that needs to be stored and save storage space. In this way, while ensuring the quality of monitoring, the operation and maintenance costs of the centralized control system can be effectively controlled. Dynamically adjusting the acquisition strategy according to the needs of different application scenarios enhances the flexibility and adaptability of the centralized control system.

[0043] In an optional embodiment, obtaining the target acquisition frequency includes: When the first monitoring data is collected for the first time, the first preset initial frequency is determined as the target collection frequency.

[0044] For example, the first preset initial frequency may be 0.03 Hz, 10 Hz, 50 Hz, etc.

[0045] In another more specific embodiment, obtaining the target acquisition frequency includes: When the first monitoring data is not collected for the first time, a historical PID (Proportional-Integral-Derivative Controller) control output and a historical collection frequency are obtained, where the historical PID control output is the PID control output corresponding to the historically collected first monitoring data, and the historical collection frequency is the collection frequency corresponding to the historically collected first monitoring data; the first collection frequency is calculated based on the historical PID control output and the historical collection frequency; and the first collection frequency is determined as the target collection frequency.

[0046] For example, the historical PID control output may be the PID control output corresponding to the last collection of the first monitoring data, and the historical collection frequency may be the collection frequency corresponding to the last collection of the first monitoring data.

[0047] In a more specific embodiment, taking the acquisition process i as an example, obtaining the historical PID control output may include the following steps: Step 1: Define variables: Define the range of the acquisition frequency fi. For example, the acquisition frequency fi can range from 0.03 Hz to 10 Hz.

[0048] Define the initial error term, initial integral term, values ​​of initially collected monitoring data, the CPU performance of the centralized control system, and the maximum CPU usage limit α. For example, the values ​​of the initial error term, initial integral term, and initially collected monitoring data can be defined as 0. It should be noted that the definitions of the above variables are user-defined based on actual circumstances and are not restricted by this application.

[0049] Step 2: Based on historical monitoring data and historical collection frequency, use the following formula to calculate the historical change rate: ΔDi= (Dprevious2_i – Dprevious1_i) / Δti(1); Where ΔDi is the historical rate of change, Dprevious2_i is the last collected first monitoring data, Dprevious1_i is the last two collected first monitoring data, Δt_i is the historical periodic collection interval between the two first monitoring data collections, Δt_i = 1 / fi, and fi is the collection frequency of the last first monitoring data collection. It should be noted that Dprevious2_i and Dprevious1_i are continuously updated.

[0050] Step 3: Based on the historical rate of change and the historical expected rate of change, calculate the historical error term using the following formula: Ei = ΔDi - ΔD_ref (2); Among them, Ei is the historical error term, ΔDi is the historical rate of change, and ΔD_ref is the historical expected rate of change.

[0051] Based on the historical period collection interval and the historical error term, the historical integral term is calculated using the following formula: integral_e_i = Ei * Δt_i (3); Among them, integral_e_i is the historical integral term, Ei is the historical error term, and Δt_i is the historical period collection interval between two collections of the first monitoring data.

[0052] Step 4: Based on the historical error term and the historical period collection interval, use the following formula to calculate the historical differential term: derivative_e_i = (Eprev_2i - Eprev_1i) / Δt_i (4); Among them, derivative_e_i is the historical differential term, Eprev_2i is the historical error term of the last collection of the first monitoring data, Eprev_1i is the historical error term of the last two collections of the first monitoring data, and Δt_i is the historical period collection interval between the two collections of the first monitoring data.

[0053] Step 5: Based on the proportional coefficient, integral coefficient, differential coefficient, historical error term, historical integral term, and historical differential term, use the following formula to calculate the historical PID control output: u_i = Kp_i * Ei + Ki_i * integral_e_i + Kd_i * derivative_e_i (5); Among them, u_i is the historical PID control output, Kp_i is the proportional coefficient, Ki_i is the integral coefficient, Kd_i is the derivative coefficient, Ei is the historical error term, integral_e_i is the historical integral term, and derivative_e_i is the historical derivative term.

[0054] In a more specific embodiment, the first acquisition frequency is calculated based on the historical PID control output and the historical acquisition frequency using the following formula: fnew_1i = fi + u_i(6); Among them, fnew_1i is the first acquisition frequency, fi is the historical acquisition frequency, and u_i is the historical PID control output.

[0055] In another optional embodiment, obtaining the target acquisition frequency further includes: Obtain historical CPU (Central Processing Unit) usage parameters; calculate a CPU limit parameter based on the CPU performance parameters and the CPU maximum occupancy limit parameter; if the historical CPU usage parameters exceed the CPU limit parameter, determine a second preset collection frequency as a target collection frequency; if the historical CPU usage parameters do not exceed the CPU limit parameter, calculate a second collection frequency based on the first collection frequency, the second preset collection frequency, the historical CPU usage parameters, and the CPU limit parameter; and determine the second collection frequency as the target collection frequency.

[0056] In this embodiment, the first control device adopts a PID algorithm and a CPU load dual closed-loop control mechanism to dynamically adjust the collection frequency of field monitoring data according to the real-time monitored CPU load and data accuracy requirements to reduce the communication network load.

[0057] In a more specific embodiment, the CPU limit parameter is calculated based on the CPU performance parameter and the CPU maximum occupancy limit parameter using the following formula: CPUlimit = α * CPUtotal(7); CPUlimit is the CPU limit parameter, α is the CPU maximum usage limit parameter, and CPUtotal is the CPU performance parameter.

[0058] In a more specific embodiment, the second preset acquisition frequency may be 0.03 Hz, etc., which is not limited in this application.

[0059] In a more specific embodiment, the second acquisition frequency is calculated using the following formula according to the first acquisition frequency, the second preset acquisition frequency, the historical CPU usage parameter, and the CPU limit parameter: fnew_2i = fx_i + (fnew_1i – fx_i) * (CPUlimit / CPUusage) (8); Among them, fnew_2i is the second collection frequency, fx_i is the second preset collection frequency, fnew_1i is the first collection frequency, CPUlimit is the CPU limit parameter, and CPUusage is the historical CPU usage parameter.

[0060] Step S403: Send the first monitoring data to the second control device according to the subscription mechanism, so that the second control device trains the first neural network model based on the first monitoring data.

[0061] It should be noted that the first neural network model is configured in the second control device.

[0062] When sending the first monitoring data to the second control device, considering that not all data needs to be uploaded to the second control device at all times, through the subscription mechanism, the first monitoring data will be triggered to be sent to the second control device only when specific conditions are met, thereby reducing unnecessary polling requests and reducing the network burden.

[0063] In an optional embodiment, when the second monitoring data corresponding to the device matrix is ​​obtained, sending the monitoring data to the second control device includes: The first monitoring data and the second monitoring data are sent to the second control device according to the subscription mechanism.

[0064] In another optional embodiment, sending the first monitoring data to the second control device according to the subscription mechanism includes: Receive subscription information sent by the second control device; determine a subscription mechanism according to the subscription information; and send the first monitoring data to the second control device in real time according to the subscription mechanism.

[0065] In a specific embodiment, the subscription information at least includes information on the maximum data volume of the first monitoring data to be uploaded at a single time, whether to compress the data, and whether to encrypt the data.

[0066] In a more specific embodiment, the monitoring data is packaged and sent to the second control device, wherein the header includes at least information such as the time the monitoring data was collected and the source of the monitoring data. Of course, other information may also be included, which is not limited by this application. The message content can be displayed in the form of an array, including at least the array name and the monitoring data.

[0067] In an optional embodiment, before sending the monitoring data to the second control device, the first control device must first establish a communication connection with the second control device. Establishing the communication connection specifically includes the following steps: Ensure that the first control device and the second control device are connected via a cable, wireless signal, or other medium; the first control device obtains the IP address and port number of the second control device, and packages and sends its own IP address and port number to the second control device, thereby establishing a communication connection with the second control device. It should be noted that the process of establishing a communication connection between devices may vary depending on different communication protocols and application scenarios, and this application does not limit this.

[0068] In another optional embodiment, the first neural network model is configured in the second control device in the form of a graphical interface, and the first monitoring data is sent to the second control device according to a subscription mechanism, so that the second control device trains the first neural network model based on the first monitoring data. Specifically, the following steps are included: The first monitoring data is sent to the second control device according to the subscription mechanism, so that the second control device switches the first neural network model to a training mode after completing the configuration of the first neural network model, and trains the first neural network model using the first monitoring data as a sample.

[0069] Step S404: Receive the operation instruction and the trained first neural network model returned by the second control device based on the first monitoring data.

[0070] After receiving the first monitoring data, the second control device analyzes the first monitoring data through a staff member or an automated control system and makes a corresponding decision. The second control device then generates corresponding operational instructions based on the decision. These operational instructions may be generated based on staff input or automatically based on a pre-set automated program, though this application does not limit this. Finally, the second control device transmits the generated operational instructions to the first control device.

[0071] In an optional embodiment, the operation instructions include but are not limited to: start instructions, stop instructions, adjustment instructions, protection instructions, status query instructions, and maintenance instructions.

[0072] After receiving the first monitoring data, the second control device may further train the first neural network model based on the first monitoring data and return the trained neural network model to the first control device. Furthermore, if the first neural network model is already running in the first control device, the trained first neural network model is used to replace the running first neural network model.

[0073] The training process of the first neural network model specifically includes the following steps: obtaining multiple first monitoring data samples and operation labels corresponding to the first monitoring data samples; inputting the multiple first monitoring data samples into the first neural network model to obtain predicted operations corresponding to the multiple first monitoring data samples; and adjusting the model parameters of the first neural network model based on the difference between the predicted operations and the operation labels. The model training process in this application is similar to the model training process in the prior art and will not be described in detail in this application.

[0074] In an optional embodiment, an algorithm service program and multiple algorithm running programs are run in the pre-trained neural network model, wherein the algorithm service program is responsible for controlling the working status of all algorithm running programs in real time, ensuring their normal operation, and adjusting or restarting them when necessary. The algorithm service program distributes the model parameters or the entire model structure of the updated pre-trained first neural network model received from the second control device to the corresponding algorithm running programs, so that all algorithm running programs use the latest and most optimized model to predict the target operation and then execute the target operation. The algorithm running program runs a single neural network model in real time, reads monitoring data from the database of the first control device, performs prediction tasks, and stores the results back in the database for subsequent analysis, wherein the database includes a real-time database and a historical database. This is conducive to achieving efficient distributed computing, especially when processing large-scale data sets or diverse computing tasks, which can ensure the flexibility and scalability of the centralized control system and the efficiency and accuracy of task processing.

[0075] Step S405: Input the operation instruction into the trained first neural network model received by the first control device to obtain the target operation.

[0076] In an optional embodiment, the target operation includes but is not limited to: controlling the device to start up, controlling the device to stop, adjusting parameters of the device, controlling the device to stop urgently, and obtaining operating status parameters of the device.

[0077] Step S406: Execute the target operation.

[0078] In an optional embodiment, the first control device controls the device to start up, controls the device to stop, adjusts parameters of the device, controls the device to shut down urgently, and obtains operating status parameters of the device.

[0079] In the embodiment of the present application, a first control device is used in the centralized control system, one first control device is deployed in each device matrix, and a second control device is connected to all first control devices; a description file is generated, and when the second control device is connected for the first time, the description file is sent to the second control device so that the centralized control template is configured based on the description file; monitoring data corresponding to the device matrix is ​​collected according to the target collection frequency, and sent to the second control device according to the subscription mechanism to train the neural network model; the operation instructions and the trained neural network model returned by the second control device are received; the operation instructions are input into the neural network model to obtain the target operation and execute the target operation. The matrix is ​​equipped with the first control device, and a matrix-level neural network model operation platform is provided to realize modular construction and operation and maintenance of the plant and station, reduce the complexity of control system construction and operation and maintenance, and improve its computing performance and reliability.

[0080] like Figure 2 As shown, the embodiment of the present application further provides a centralized control system, which includes at least one device matrix, a first control device and a second control device, wherein each device matrix has a corresponding first control device, and the second control device is connected to all first control devices; the device matrix includes multiple devices; The first control device includes: a generation module, configured to generate a description file and, when connected to a second control device for the first time, send the description file to the second control device, so that the second control device configures a centralized control template based on the description file, and enters an operation mode after the centralized control template configuration is completed, wherein the centralized control template at least includes a device matrix, a connection relationship between the first control device and the second control device; An acquisition module, configured to acquire first monitoring data corresponding to the device array according to a target acquisition frequency; a sending module, configured to send the first monitoring data to the second control device according to a subscription mechanism, so that the second control device trains the first neural network model based on the first monitoring data, the first neural network model being configured in the second control device; A receiving module, configured to receive an operation instruction and a trained first neural network model returned by the second control device based on the first monitoring data; An input module, configured to input an operation instruction into the trained first neural network model received by the first control device to obtain a target operation; Execution module, used to execute target operations.

[0081] In an optional embodiment, the description file includes at least the following information: a list of measurement points, the IP address of the first control device, the device number, the device name, the device location, and the device type, device location, number of devices, and rated power of the devices included in the device array connected to the first control device.

[0082] In an optional embodiment, the first control device includes a pre-trained second neural network model and a module for performing the following operations: Inputting the first monitoring data into the second neural network model to obtain second monitoring data corresponding to the device matrix; Accordingly, the sending module is used to: The first monitoring data and the second monitoring data are sent to the second control device according to the subscription mechanism.

[0083] In an optional embodiment, the second control device configures the first neural network model in the form of a graphical interface, and the sending module is used to: The first monitoring data is sent to the second control device according to the subscription mechanism, so that the second control device switches the first neural network model to a training mode after completing the configuration of the first neural network model, and trains the first neural network model using the first monitoring data as a sample.

[0084] In an optional embodiment, the method further includes a module for performing the following operations: In a case where the first neural network model is already running in the first control device, the first neural network model being run is replaced with the received trained first neural network model.

[0085] In an optional embodiment, the acquisition module is configured to: Get the target acquisition frequency, including: When collecting the first monitoring data for the first time, determining the first preset initial frequency as the target collection frequency; In the case where the first monitoring data is not collected for the first time, a historical PID control output and a historical collection frequency are obtained, where the historical PID control output is the PID control output corresponding to the historically collected first monitoring data, and the historical collection frequency is the collection frequency corresponding to the historically collected first monitoring data; According to the historical PID control output and the historical acquisition frequency, the first acquisition frequency is calculated; The first acquisition frequency is determined as the target acquisition frequency.

[0086] In an optional embodiment, the acquisition module is further configured to: Get historical CPU usage parameters; Calculate the CPU limit parameter based on the CPU performance parameter and the CPU maximum occupancy limit parameter; if the historical CPU usage parameter exceeds the CPU limit parameter, determine the second preset acquisition frequency as the target acquisition frequency; When the historical CPU usage parameter does not exceed the CPU limit parameter, a second collection frequency is calculated based on the first collection frequency, the second preset collection frequency, the historical CPU usage parameter, and the CPU limit parameter; The second acquisition frequency is determined as the target acquisition frequency.

[0087] In an optional embodiment, the sending module is configured to: receiving subscription information sent by the second control device; Determine the subscription mechanism based on the subscription information; The first monitoring data is sent to the second control device in real time according to the subscription mechanism.

[0088] In an optional embodiment, the subscription information includes at least information on the maximum data volume of the first monitoring data uploaded at a single time, whether it is compressed, and whether it is encrypted.

[0089] Corresponding to a centralized control method provided in an embodiment of the present application, an embodiment of the present application further provides an electronic device for executing the centralized control method, such as Figure 5 As shown, the electronic device includes: a processor 501; and a memory 502, which is used to store a program of the centralized control method. After the device is powered on and the program of the centralized control method is run by the processor, the following steps are performed: Generate a description file and send the description file to the second control device when connecting to the second control device for the first time, so that the second control device configures a centralized control template based on the description file, and enters the operation mode after the centralized control template configuration is completed, where the centralized control template includes at least a device matrix and a connection relationship between the first control device and the second control device; collecting first monitoring data corresponding to the device array according to a target collection frequency; sending the first monitoring data to the second control device according to a subscription mechanism, so that the second control device trains the first neural network model based on the first monitoring data, the first neural network model being configured in the second control device; receiving an operation instruction and a trained first neural network model returned by a second control device based on the first monitoring data; Inputting the operation instruction into the trained first neural network model received by the first control device to obtain a target operation; Execute the target action.

[0090] Corresponding to the centralized control method provided in the embodiment of the present application, the embodiment of the present application further provides a computer-readable storage medium storing a program of the centralized control method, which is executed by a processor to perform the following steps: Generate a description file and send the description file to the second control device when connecting to the second control device for the first time, so that the second control device configures a centralized control template based on the description file, and enters the operation mode after the centralized control template configuration is completed, where the centralized control template includes at least a device matrix and a connection relationship between the first control device and the second control device; collecting first monitoring data corresponding to the device array according to a target collection frequency; sending the first monitoring data to the second control device according to a subscription mechanism, so that the second control device trains the first neural network model based on the first monitoring data, the first neural network model being configured in the second control device; receiving an operation instruction and a trained first neural network model returned by a second control device based on the first monitoring data; Inputting the operation instruction into the trained first neural network model received by the first control device to obtain a target operation; Execute the target action.

[0091] Corresponding to the centralized control method provided in the embodiment of the present application, the embodiment of the present application further provides a computer program comprising instructions. When the program is executed by a computer, the instructions cause the computer to perform the following steps: Generate a description file and send the description file to the second control device when connecting to the second control device for the first time, so that the second control device configures a centralized control template based on the description file, and enters the operation mode after the centralized control template configuration is completed, where the centralized control template includes at least a device matrix and a connection relationship between the first control device and the second control device; collecting first monitoring data corresponding to the device array according to a target collection frequency; sending the first monitoring data to the second control device according to a subscription mechanism, so that the second control device trains the first neural network model based on the first monitoring data, the first neural network model being configured in the second control device; receiving an operation instruction and a trained first neural network model returned by a second control device based on the first monitoring data; Inputting the operation instruction into the trained first neural network model received by the first control device to obtain a target operation; Execute the target action.

[0092] It should be noted that for the detailed description of the system, electronic device and computer-readable storage medium provided in the embodiments of the present application, reference can be made to the relevant description of the centralized control method embodiment provided in the embodiments of the present application, and no further details will be given here.

[0093] Although the present application is disclosed as above with the preferred embodiments, it is not intended to limit the present application. Any person skilled in the art may make possible changes and modifications without departing from the spirit and scope of the present application. Therefore, the scope of protection of the present application shall be based on the scope defined by the claims of the present application.

[0094] In a typical configuration, an electronic device includes one or more processors (CPUs), input / output interfaces, network interfaces, and memory.

[0095] Memory may include non-permanent storage in a computer-readable medium, random access memory (RAM), and / or non-volatile memory, such as read-only memory (ROM) or flash memory. Memory is an example of a computer-readable medium.

[0096] 1. Computer-readable media, including permanent and non-permanent, removable and non-removable media, can be implemented using any method or technology to store information. Information can be computer-readable operations, data structures, program modules, or other data. Examples of computer storage media include, but are not limited to, programmable analog modules (PAMs), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RANM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPRM), flash memory or other memory technologies, compact disc read-only memory (CDROM), digital versatile disc (DVCD) or other optical storage, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer-readable media does not include non-transitory computer-readable media, such as modulated data signals and carrier waves.

[0097] 2. Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware. Furthermore, the present application may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, compact disc read-only memory, optical storage, etc.) containing computer-usable program code.

[0098] Although the present application is disclosed as above with the preferred embodiments, it is not intended to limit the present application. Any person skilled in the art may make possible changes and modifications without departing from the spirit and scope of the present application. Therefore, the scope of protection of the present application shall be based on the scope defined by the claims of the present application.

Claims

1. A centralized control method, characterized in that: A first control device is applied to a centralized control system, wherein each device array has a corresponding first control device deployed therein, and a second control device is connected to all the first control devices; the centralized control method includes: Generate a description file, and send the description file to the second control device when connecting to the second control device for the first time, so that the second control device configures a centralized control template based on the description file, and enters the operation mode after the centralized control template configuration is completed, wherein the centralized control template includes at least a device matrix and a connection relationship between the first control device and the second control device; Collecting first monitoring data corresponding to the device array according to a target collection frequency; sending the first monitoring data to the second control device according to a subscription mechanism, so that the second control device trains a first neural network model based on the first monitoring data, where the first neural network model is configured in the second control device; receiving an operation instruction and the trained first neural network model returned by the second control device based on the first monitoring data; Inputting the operation instruction into the trained first neural network model received by the first control device to obtain a target operation; Execute the target action.

2. The centralized control method according to claim 1, characterized in that: The description file includes at least the following information: a list of measurement points, the IP address, device number, device name, and device location of the first control device, as well as the device type, device location, device quantity, and device rated power of the devices included in the device array connected to the first control device.

3. The centralized control method according to claim 1, characterized in that: The first control device includes a pre-trained second neural network model, and further includes: Inputting the first monitoring data into the second neural network model to obtain second monitoring data corresponding to the device matrix; Accordingly, sending the first monitoring data to the second control device according to the subscription mechanism includes: The first monitoring data and the second monitoring data are sent to the second control device according to a subscription mechanism.

4. The centralized control method according to claim 1, characterized in that: The first neural network model is configured in the second control device in the form of a graphical interface, and the first monitoring data is sent to the second control device according to the subscription mechanism so that the second control device trains the first neural network model based on the first monitoring data, including: The first monitoring data is sent to the second control device according to the subscription mechanism, so that the second control device switches the first neural network model to training mode after completing the configuration of the first neural network model, and trains the first neural network model using the first monitoring data as a sample.

5. The centralized control method according to claim 1, characterized in that: Also includes: In a case where the first neural network model is already running in the first control device, the first neural network model being run is replaced with the received trained first neural network model.

6. The centralized control method according to claim 1, characterized in that: The acquiring target acquisition frequency includes: When collecting the first monitoring data for the first time, determining a first preset initial frequency as the target collection frequency; In a case where this is not the first time that the first monitoring data is collected, obtaining a historical PID control output and a historical collection frequency, wherein the historical PID control output is the PID control output corresponding to the historical collection of the first monitoring data, and the historical collection frequency is the collection frequency corresponding to the historical collection of the first monitoring data; Calculating a first acquisition frequency according to the historical PID control output and the historical acquisition frequency; The first acquisition frequency is determined as a target acquisition frequency.

7. The centralized control method according to claim 6, characterized in that: The acquiring target acquisition frequency further includes: Get historical CPU usage parameters; Calculating a CPU limit parameter based on the CPU performance parameter and the CPU maximum occupancy limit parameter; and determining the second preset acquisition frequency as the target acquisition frequency when the historical CPU usage parameter exceeds the CPU limit parameter; When the historical CPU usage parameter does not exceed the CPU limit parameter, calculating a second acquisition frequency according to the first acquisition frequency, the second preset acquisition frequency, the historical CPU usage parameter, and the CPU limit parameter; The second acquisition frequency is determined as a target acquisition frequency.

8. The centralized control method according to claim 1, characterized in that: The sending the first monitoring data to the second control device according to the subscription mechanism includes: receiving subscription information sent by the second control device; determining a subscription mechanism according to the subscription information; The first monitoring data is sent to the second control device in real time according to the subscription mechanism.

9. The centralized control method according to claim 8, characterized in that: The subscription information at least includes information on the maximum data volume of the first monitoring data uploaded at a single time, whether it is compressed, and whether it is encrypted.

10. A centralized control system, characterized in that: The centralized control system includes at least one device matrix, a first control device and a second control device in the device matrix, one first control device is deployed in each device matrix, and the second control device is connected to all the first control devices; the device matrix includes multiple devices; The first control device includes: a generation module, configured to generate a description file and send the description file to the second control device when connecting to the second control device for the first time, so that the second control device configures a centralized control template based on the description file, and enters an operation mode after the centralized control template configuration is completed, wherein the centralized control template at least includes a connection relationship between a device matrix, a first control device, and a second control device; An acquisition module, configured to acquire first monitoring data corresponding to the device array according to a target acquisition frequency; a sending module, configured to send the first monitoring data to the second control device according to a subscription mechanism, so that the second control device trains a first neural network model based on the first monitoring data, where the first neural network model is configured in the second control device; A receiving module, configured to receive an operation instruction and a trained first neural network model returned by the second control device based on the first monitoring data; an input module, configured to input the operation instruction into the trained first neural network model received by the first control device to obtain a target operation; Execution module, used to execute target operations.

Images