Power plant electrical automation control system and method based on hierarchical management

Abstract

The invention discloses a power plant electrical automation control system and method based on hierarchical management, and relates to the technical field of power plant electrical control, and the system comprises a dynamic resource distribution coordination mechanism, a fault self-healing mechanism, and a local optimization control and energy efficiency optimization strategy based on edge calculation. The resource output of the dynamic resource allocation coordination mechanism is used as the optimized resource input of the local optimization control based on the edge computing, and the energy efficiency optimization mechanism is used as a dynamic interaction layer of the dynamic resource allocation coordination mechanism and the local optimization control based on the edge computing. The fault self-healing mechanism is used as a result execution mechanism of the energy efficiency optimization strategy. By designing a dynamic resource allocation coordination mechanism, the resource load dynamic allocation function of the power plant equipment is realized, the problems of unbalanced resource utilization, equipment overload or low-efficiency operation and incapability of quickly responding to resource requirements in a traditional power plant are solved, resource allocation is optimized, and the operation efficiency of the whole system is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of power plant electrical control, and in particular to a power plant electrical automation control system and method based on hierarchical management. Background Art

[0002] With the rapid development of the power industry, the automation level of power plants is constantly improving, but traditional power plant electrical control systems still face many challenges in dealing with complex working conditions, equipment failures, and energy efficiency optimization. The collaborative work and real-time dispatching capabilities between power plant equipment are relatively limited, and they cannot effectively deal with sudden load changes and failure risks, resulting in insufficient resource utilization, high energy consumption, and poor system operation reliability. Therefore, there is an urgent need for a more intelligent and flexible electrical automation control system that can achieve dynamic resource allocation, rapid fault response, and energy efficiency optimization to improve the overall operating efficiency and economic benefits of power plants.

[0003] Patent document CN112381146B discloses a method for self-organizing aggregation and collaborative control of distributed resources under a virtual power plant. The above patent realizes the optimal combination and collaborative control of resources through self-organizing aggregation of subjects, reduces the overall regulation cost, and significantly improves the operating efficiency of the virtual power plant; at the same time, a multi-level self-organizing aggregation method for a virtual power plant is proposed, which provides an underlying mechanism for revealing the emergence mechanism of the system; and this method proposes an implementation method for self-organizing aggregation of adaptive subjects, which can quickly and accurately solve the optimal joint behavior and benefits of the adaptive subject combination, accelerate the convergence process of self-organizing aggregation, and improve the overall decision-making efficiency, but the above patent cannot realize the dynamic allocation function of resource loads for power plant equipment.

[0004] Patent document CN112448396B discloses a self-optimizing load tracking control method for a virtual power plant. The above patent is based on the self-optimizing operation theory, adopts a bottom-up approach, and drives the virtual power plant as a whole to track the expected load target curve through adaptive adjustment of massive heterogeneous and time-varying distributed resources, thereby realizing dynamic mining and efficient utilization of the flexibility of massive distributed resources. This method can quickly explore the regulation capability of the virtual power plant under complex conditions, accurately track the target load curve within its capability, and effectively drive the whole to approach the theoretical optimal operating point while the individual is operating in an optimal manner. However, the above patent cannot realize the function of quickly identifying the equipment status and isolating faults.

[0005] Patent document CN110000158B discloses a program-controlled gas-phase electrical switch cabinet dust removal system and control method. The above patent realizes the cleaning of various electrical switch cabinets in thermal power plants; it adopts an automated integrated control system, on-site sampling, analysis and calculation, and automatic output control to realize the equipment cleaning operation of the electrical switch cabinet under long-term continuous and interlocking working modes. The system has a simple structure, mature technology, low failure rate, high reliability, simple cost and low daily operation and maintenance costs, but the above patent cannot realize the monitoring and calculation adjustment functions of the equipment layer.

[0006] Patent document CN113489002B discloses a method for regulating the AVC substation system of a gas and steam combined cycle unit. The above patent optimizes the AVC regulation mechanism and improves the safety of AVC regulation. It plays a role in ensuring the improvement of the AVC regulation performance of the unit and the correctness of the unit's regulation in two detailed assessments, thereby improving the safety of the unit's operation. However, the above patent cannot achieve the optimal configuration function of energy utilization.

[0007] In summary, the above patents cannot realize the functions of dynamic resource load allocation for power plant equipment, rapid identification of equipment status and isolation of faults, monitoring and calculation adjustment of equipment layer and optimal configuration of energy utilization, resulting in technical problems such as unreasonable resource allocation, slow fault response, excessive energy consumption and low equipment operation efficiency; To this end, the present application proposes a power plant electrical automation control system and method based on hierarchical management that can realize the dynamic allocation of resource loads for power plant equipment, the function of quickly identifying equipment status and isolating faults, the monitoring and calculation adjustment function of the equipment layer, and the optimization configuration function of energy utilization. Summary of the invention

[0008] The purpose of the present invention is to provide an electrical automation control system and method for a power plant based on hierarchical management, so as to solve the technical problems proposed in the above background technology that the resource load dynamic allocation function of the power plant equipment, the function of quickly identifying the equipment status and isolating the fault, the monitoring and calculation adjustment function of the equipment layer and the optimal configuration function of energy utilization cannot be realized, resulting in unreasonable resource allocation, slow fault response, excessive energy consumption and low equipment operation efficiency.

[0009] To achieve the above-mentioned purpose, the present invention provides the following technical solutions: an electrical automation control system for a power plant based on hierarchical management, comprising a dynamic resource allocation coordination mechanism, a fault self-healing mechanism, a local optimization control based on edge computing, and an energy efficiency optimization strategy, wherein the resource output of the dynamic resource allocation coordination mechanism is used as the optimization resource input of the local optimization control based on edge computing, the energy efficiency optimization mechanism is used as a dynamic interaction layer between the dynamic resource allocation coordination mechanism and the local optimization control based on edge computing, and the fault self-healing mechanism is used as a result execution mechanism of the energy efficiency optimization strategy; The dynamic resource allocation coordination mechanism includes: a dynamic resource allocation control unit, a load prediction module and a priority-based task scheduling algorithm. The dynamic resource allocation coordination mechanism is used to adjust the computing resources and communication bandwidth allocation of each layer of the control system to ensure that key equipment responds first under high load conditions; The dynamic resource allocation control unit uses an embedded controller in conjunction with an artificial intelligence algorithm to predict system load trends and dynamically allocate system resources based on historical operating data and real-time data; The load forecasting module obtains the current load data of the power plant through real-time data acquisition equipment, and inputs the load data into the prediction models of the cloud and edge computing nodes to generate load forecasting results, which are fed back to the dynamic resource allocation control unit for dynamic adjustment; The priority-based task scheduling algorithm introduces a priority queue into the system scheduler, and dynamically sorts tasks according to their urgency, ensuring that high-priority tasks can obtain resources first under high-load conditions.

[0010] Preferably, the fault self-healing mechanism includes: a multi-layer communication network, a fault detection and isolation module, a redundant control unit and a fast recovery logic; A multi-layer communication network means that in a layered system, the device layer, control layer and management layer are transmitted through independent communication networks. The power plant communication network based on industrial Ethernet and IEC 61850 protocol ensures the reliability and real-time performance of communication between different layers through the isolation of the physical layer and the logical layer. The fault detection and isolation module monitors the status of system equipment in real time, determines equipment abnormalities through multi-sensor data fusion technology, and processes the data collected by equipment sensors using multi-sensor data fusion algorithms. Once an abnormality is detected, an isolation command is issued through the SCADA system, and other equipment is reconfigured to take over the function. The redundant control unit immediately starts taking over the control task of the device after receiving the isolation instruction from the fault detection and isolation module. The introduction of redundant design ensures that the redundant control unit can seamlessly take over the control task when a single device fails. The fast recovery logic coordinates the status of the multi-layer control system through the state machine control logic. After receiving the system fault repair recovery signal, the system automatically switches to normal mode, automatically releases redundant control, and restores resource allocation under normal conditions.

[0011] Preferably, the local optimization control based on edge computing is collaboratively completed by an edge computing unit, a local data processing module, a distributed control algorithm and a local optimization model; The edge computing unit is deployed at the equipment layer and directly connected to the equipment in the power plant. It can locally process and optimize the equipment operation data. The edge computing unit is composed of an industrial controller and equipped with an edge computing algorithm to analyze the equipment's operating efficiency and failure rate data in real time and adjust the equipment's working status. The local data processing module processes data at the device level, makes local decisions by analyzing data in real time, combines IoT technology, data processing chips, and local memory, and uses a data stream processing engine to process high-frequency data in real time; The system uses a distributed control algorithm to enable each edge node to independently perform control tasks and work with other nodes through the network to achieve global optimization. It uses a distributed optimal control algorithm and interval optimization method to ensure that the control instructions of each edge node achieve a balance between local optimization and global optimization. The local optimization model means that the system uses a data-driven model to perform local optimization based on the working conditions of local equipment, establishes a local optimization model for the equipment through a machine learning algorithm, predicts equipment failures through a data-driven state detection model, adjusts the operation strategy in advance, and updates and optimizes the equipment's operating parameters in real time.

[0012] Preferably, the energy efficiency optimization strategy is jointly executed by an energy efficiency management module, an equipment operation mode optimization algorithm, a distributed load prediction unit and an energy efficiency collaborative control strategy; The energy efficiency management module is located at the management level of the hierarchical architecture. It is responsible for monitoring the energy efficiency of the entire power plant and automatically adjusting the operating status of the equipment to achieve optimal energy efficiency based on real-time data and prediction results. It develops energy efficiency optimization algorithms based on AI technology, uses load prediction, operation data analysis and dynamic optimization models, and continuously adjusts the power and working status of the equipment to achieve energy-saving goals. The equipment operation mode optimization algorithm means that within the framework of energy efficiency optimization, the operation mode of each device will be automatically adjusted according to the energy consumption status. Through the dynamic adjustment algorithm, the equipment operation mode is automatically switched according to the load forecast and energy consumption data to ensure the best energy efficiency under different load conditions. The distributed load forecasting unit uses multiple computing nodes to predict the load demand of each region in real time, and then provides optimization suggestions for the energy efficiency management module. Through cloud computing and edge computing architecture, the load forecasting task is distributed to multiple computing nodes, and regional forecasting is performed by combining historical data and real-time data; The energy efficiency collaborative control strategy integrates multi-objective optimization algorithms to establish a collaborative mechanism between the central control system and edge computing nodes. By coordinating devices at different levels, the system can achieve optimal energy consumption on a global scale.

[0013] Preferably, the load forecasting model adopts an LSTM neural network model to perform long-term and short-term load trend forecasts based on historical load data.

[0014] Preferably, the priority-based task scheduling algorithm is a preemptive priority scheduling algorithm, which can dynamically preempt system resources occupied by low-priority tasks when high-priority tasks appear.

[0015] Preferably, the distributed control algorithm is a model predictive control algorithm, and each edge computing node independently performs local optimization tasks through the distributed control algorithm while considering global objectives.

[0016] Preferably, the energy efficiency management module uses a dynamic energy allocation algorithm to adjust the operating power of the equipment according to load prediction and equipment status to achieve energy saving goals.

[0017] Preferably, the control method comprises the following steps: S1. Based on the load forecasting module system, the system combines historical data with real-time operation data and uses the LSTM neural network model to predict future load conditions in real time, and feeds the prediction results back to the dynamic resource allocation control unit. The dynamic resource allocation control unit makes a preliminary resource allocation plan based on the load change trend, providing a basis for subsequent resource scheduling; S2. The device layer sensors send the collected data to the edge computing unit through the Internet of Things technology. The edge computing unit processes the collected data on the spot. Based on the real-time operation data, the edge computing unit uses the local optimization control algorithm to make real-time adjustments to the operation of the equipment to ensure that the equipment reaches the optimal operating state under the current load; S3, the fault detection module of the system analyzes the operating status of the equipment in real time by fusing data from different sensors and discovers potential faults. The fault isolation module immediately performs isolation operations, removes the faulty equipment from the control system, and starts the redundant control unit to take over the equipment task; S4. Based on the load forecast data provided by the distributed load forecasting unit, the system automatically adjusts the equipment operation mode through the equipment operation mode optimization algorithm, controls the power output and operation status of the equipment in real time, and ensures the optimal energy efficiency under different load conditions.

[0018] Preferably, the control method further comprises the following steps: S11. Through the priority-based task scheduling algorithm, the system allocates more computing resources and communication loans to high-load and critical devices according to the importance and current load of each device. The resource allocation plan is transmitted to each layer of devices through the multi-layer communication network; S21. The processed local data is fed back to the central control system by the edge computing unit. The central control system analyzes the data based on the global perspective. The central control system and the edge computing unit jointly optimize the control strategy through the distributed control algorithm to achieve the best global performance and ensure the efficient operation of local devices in specific scenarios. S31. The redundant control unit uses the reserved resources and control strategies to continue to execute the key tasks of the isolated equipment to ensure that the system functions are not affected. The system uses the fast recovery logic to reallocate resources according to the isolation of the faulty equipment to ensure that other equipment can still operate normally while waiting for the faulty equipment to be repaired. S41. The energy efficiency management module cooperates with the dynamic resource allocation control unit to coordinate the operation strategies of equipment at different levels to achieve global energy efficiency optimization. The energy efficiency performance of the equipment is fed back to the central control system through the edge computing unit. The system dynamically adjusts the energy efficiency optimization strategy based on the feedback data to form a closed-loop optimization and further improve the overall energy efficiency.

[0019] Compared with the prior art, the present invention has the following beneficial effects: 1. The present invention realizes the dynamic resource allocation and coordination mechanism to dynamically allocate resource loads of power plant equipment, solves the problems of unbalanced resource utilization, equipment overload or inefficient operation in traditional power plants, especially the problem of being unable to quickly respond to resource demands when the load changes drastically, optimizes resource allocation, reduces energy waste, and improves the operating efficiency of the overall system; 2. The present invention realizes the function of quickly identifying the equipment status and isolating the fault by designing a fault self-healing mechanism, which solves the problem of long downtime and high maintenance cost caused by delayed equipment fault identification in traditional power plant systems, improves the fault tolerance and operation reliability of the system, shortens the fault repair time, reduces the economic losses caused by equipment failure, and ensures the continuity and stability of power plant production; 3. The present invention realizes the monitoring and calculation adjustment functions of the equipment layer by designing a local optimization control based on edge computing, solves the problems of slow response speed of the equipment layer in traditional power plants, excessive reliance on central system decision-making, resulting in local optimization lag and difficulty in adapting to rapidly changing load demands, improves the response speed and real-time performance of local equipment, reduces the calculation and communication pressure of the central system, and optimizes the equipment operation efficiency; 4. The present invention realizes the optimal configuration function of energy utilization by designing an energy efficiency optimization strategy, solves the problems of energy waste and low equipment operation efficiency in power plants, especially the difficulty of traditional power plants in flexibly adjusting equipment power and energy consumption according to real-time demand, which leads to the problem of unreasonable energy utilization. It significantly reduces the overall energy consumption of power plants, improves energy utilization, reduces operating costs, and at the same time extends the service life of equipment, further improving the economic and environmental benefits of power plants. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 It is a schematic flow chart of the automatic control method of the present invention. DETAILED DESCRIPTION

[0021] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions 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 of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0022] See also Figure 1 , an embodiment provided by the present invention: an electrical automation control system for a power plant based on hierarchical management, comprising a dynamic resource allocation coordination mechanism, a fault self-healing mechanism, a local optimization control based on edge computing, and an energy efficiency optimization strategy, wherein the resource output of the dynamic resource allocation coordination mechanism is used as the optimization resource input of the local optimization control based on edge computing, the energy efficiency optimization mechanism is used as a dynamic interaction layer between the dynamic resource allocation coordination mechanism and the local optimization control based on edge computing, and the fault self-healing mechanism is used as a result execution mechanism of the energy efficiency optimization strategy; The dynamic resource allocation coordination mechanism includes: a dynamic resource allocation control unit, a load prediction module and a priority-based task scheduling algorithm. The dynamic resource allocation coordination mechanism is used to adjust the computing resources and communication bandwidth allocation of each layer of the control system to ensure that key equipment responds first under high load conditions; The dynamic resource allocation control unit uses an embedded controller in conjunction with an artificial intelligence algorithm to predict system load trends and dynamically allocate system resources based on historical operating data and real-time data; The load forecasting module obtains the current load data of the power plant through real-time data acquisition equipment, and inputs the load data into the prediction models of the cloud and edge computing nodes to generate load forecasting results, which are fed back to the dynamic resource allocation control unit for dynamic adjustment; The priority-based task scheduling algorithm introduces a priority queue in the system scheduler to dynamically sort tasks according to their urgency, ensuring that high-priority tasks can get priority resources under high-load conditions; The load forecasting model adopts an LSTM neural network model to perform long-term and short-term load trend forecasts based on historical load data; The priority-based task scheduling algorithm is a preemptive priority scheduling algorithm, which can dynamically preempt the system resources occupied by low-priority tasks when high-priority tasks appear; Furthermore, the system monitors the real-time load conditions in each area of ​​the power plant through sensors at various levels. The load forecasting module uses the time series model and machine learning algorithm to forecast the load based on the monitoring data combined with the historical load conditions. Based on the load forecast results, the dynamic resource allocation unit will set priorities for different tasks or equipment; Critical tasks such as the operation of main equipment and power grid safety control are given higher priority, while non-critical tasks are scheduled to be performed during periods when resources are relatively idle. Based on the equipment priority, the system allocates computing resources, network bandwidth, and control signals in real time; In the case of high load or failure, the system reduces the resource allocation of non-priority tasks through dynamic resource adjustment logic to ensure that key equipment has sufficient resource support. In the case of equipment load changes or sudden failures, the system can quickly sense the abnormality and make adaptive adjustments to resource configuration. For example, if the load in a certain area suddenly increases, the dynamic resource allocation unit will respond immediately, increase the resource allocation of the equipment in that area, and at the same time reduce the resource allocation of other non-critical area equipment; The system can dynamically adjust task priorities based on real-time operating status and load forecasts to ensure that critical tasks can receive adequate resource support during peak load periods or when resources are tight, solving the problem that traditional power plant systems are not flexible in resource scheduling when loads fluctuate or equipment fails, resulting in critical equipment not being able to obtain adequate resource support during high loads. The dynamic resource allocation mechanism can effectively solve the problems of unreasonable resource scheduling and low utilization. The dynamic resource allocation mechanism achieves reasonable allocation and redistribution of resources through intelligent scheduling, improving the overall operating efficiency of the system.

[0023] See also Figure 1 , an embodiment provided by the present invention: an electrical automation control system of a power plant based on hierarchical management, the fault self-healing mechanism comprising: a multi-layer communication network, a fault detection and isolation module, a redundant control unit and a fast recovery logic; A multi-layer communication network means that in a layered system, the device layer, control layer and management layer are transmitted through independent communication networks. The power plant communication network based on industrial Ethernet and IEC 61850 protocol ensures the reliability and real-time performance of communication between different layers through the isolation of the physical layer and the logical layer. The fault detection and isolation module monitors the status of system equipment in real time, determines equipment abnormalities through multi-sensor data fusion technology, and processes the data collected by equipment sensors using multi-sensor data fusion algorithms. Once an abnormality is detected, an isolation command is issued through the SCADA system, and other equipment is reconfigured to take over the function. The redundant control unit immediately starts taking over the control task of the device after receiving the isolation instruction from the fault detection and isolation module. The introduction of redundant design ensures that the redundant control unit can seamlessly take over the control task when a single device fails. The fast recovery logic coordinates the state of the multi-layer control system through the state machine control logic. After receiving the system fault repair recovery signal, the system automatically switches to the normal mode, automatically releases the redundant control, and restores the resource allocation under the normal state; Furthermore, the fault self-healing mechanism monitors the status of equipment in various areas of the power plant, such as current, voltage, temperature, vibration, and power, in real time through a variety of sensors and monitoring devices distributed in various layers of the system, such as the equipment layer, control layer, and management layer. These sensors send monitoring data to the fault detection and isolation module through the Internet of Things technology. When the fault is confirmed, the fault detection and isolation module will be started immediately to isolate the faulty equipment from the control system to prevent the fault from further expanding or affecting other equipment. During the isolation process, the system automatically cuts off the power or communication connection of the faulty equipment through a circuit breaker or switch device. At the same time, the system starts the redundant control unit, which pre-stores the backup control strategy and resources. After the faulty equipment is isolated, the redundant unit will immediately take over the task of the equipment to ensure that the task will not be interrupted. The redundant control takeover process is seamless, avoiding the impact of the fault on the overall system operation; After the fault is isolated, the system uses the diagnostic module to conduct an in-depth analysis of the faulty equipment to determine the root cause of the fault. The diagnostic module can call historical fault data and current equipment status information to perform automatic or remote repair operations on the equipment, such as restarting and adjusting configuration parameters. If the equipment requires manual intervention, the system will notify the maintenance personnel to perform on-site repairs. After the faulty equipment is repaired and restored to normal working state, the system will automatically release the isolation and reintegrate the equipment into the system. At this time, the system re-allocates the required computing resources, communication bandwidth and task instructions to the device through a dynamic resource allocation mechanism to restore the normal operation of the equipment; the system monitors the operating status of the equipment in real time through a multi-layer sensor network, and uses intelligent algorithms to achieve early detection and diagnosis of faults. It can issue an early warning before the fault occurs, reduce the losses caused by sudden faults, and take over the tasks of the faulty equipment through redundant control units to ensure that during the equipment failure, the key tasks can continue to run and avoid system interruptions caused by faults; it solves the technical problems of fault detection lag, fault false alarm, fault diffusion affecting system operation and slow task interruption recovery in traditional power plant control systems.

[0024] See also Figure 1 , an embodiment provided by the present invention: an electrical automation control system for a power plant based on hierarchical management, wherein the local optimization control based on edge computing is collaboratively completed by an edge computing unit, a local data processing module, a distributed control algorithm and a local optimization model; The edge computing unit is deployed at the equipment layer and directly connected to the equipment in the power plant. It can locally process and optimize the equipment operation data. The edge computing unit is composed of an industrial controller and equipped with an edge computing algorithm to analyze the equipment's operating efficiency and failure rate data in real time and adjust the equipment's working status. The local data processing module processes data at the device level, makes local decisions by analyzing data in real time, combines IoT technology, data processing chips, and local memory, and uses a data stream processing engine to process high-frequency data in real time; The system uses a distributed control algorithm to enable each edge node to independently perform control tasks and work with other nodes through the network to achieve global optimization. It uses a distributed optimal control algorithm and interval optimization method to ensure that the control instructions of each edge node achieve a balance between local optimization and global optimization. The local optimization model means that the system uses the data-driven model to perform local optimization based on the working conditions of the local equipment, establishes the local optimization model of the equipment through the machine learning algorithm, predicts the failure of the equipment through the data-driven state detection model, adjusts the operation strategy in advance, and updates and optimizes the operation parameters of the equipment in real time; The distributed control algorithm is a model predictive control algorithm, and each edge computing node independently performs local optimization tasks through the distributed control algorithm while considering the global goal; Furthermore, at each equipment level of the power plant, sensor networks such as temperature sensors, pressure sensors, current and voltage sensors collect the operating parameters of the equipment in real time. These data are usually high-dimensional and high-frequency, and direct transmission to the central system will result in excessive communication load and delay; the raw data collected by the equipment is pre-processed by the edge computing unit to remove noise, filter redundant information, and compress the data. The edge computing unit will perform data aggregation, anomaly detection and simple status analysis to provide a basis for subsequent local optimization and decision-making; The edge computing unit deploys lightweight optimization algorithms such as genetic algorithms, fuzzy control algorithms, local linear regression, etc. These algorithms can make autonomous adjustment decisions based on the real-time data of local devices. For example, for information such as load, current, voltage, etc. of a certain device, the edge unit can adjust the output power of the device locally or switch the working mode to adapt to real-time demand and load changes; the edge computing unit forms a closed-loop feedback control loop, monitors the status of the device in real time and adjusts the control instructions according to the optimization algorithm output. For example, when the load of the generator changes, the edge unit can adjust its operating parameters in real time to improve efficiency or reduce energy consumption, with extremely fast response speed and no central system intervention; After performing local optimization control, the edge computing unit summarizes the processed data and optimization results. This information will be transmitted to the central control system as feedback data for global analysis and decision-making. At the same time, the edge computing unit also has an intelligent data filtering function, which only transmits valuable information to avoid communication bottlenecks; the edge computing unit also has fault detection capabilities. By analyzing the operating status of the equipment in real time, if abnormal fluctuations or parameter out-of-limit are detected, the edge unit will immediately isolate or downgrade the local fault of the device to ensure that the fault will not affect the safe operation of the equipment; By deploying edge computing units, the system can perform real-time optimization at the device level, reducing dependence on the central control system. The edge unit can adjust parameters according to the current operating status of the equipment to achieve autonomous control and optimization; it solves the problems of response delay, insufficient communication bandwidth, fault propagation and heavy computing load in traditional power plant control systems, significantly improves system response speed, improves equipment operating efficiency and energy efficiency, and enhances system reliability and fault tolerance.

[0025] See also Figure 1 , an embodiment provided by the present invention: an electrical automation control system for a power plant based on hierarchical management, wherein the energy efficiency optimization strategy is jointly executed by an energy efficiency management module, an equipment operation mode optimization algorithm, a distributed load prediction unit and an energy efficiency collaborative control strategy; The energy efficiency management module is located at the management level of the hierarchical architecture. It is responsible for monitoring the energy efficiency of the entire power plant and automatically adjusting the operating status of the equipment to achieve optimal energy efficiency based on real-time data and prediction results. It develops energy efficiency optimization algorithms based on AI technology, uses load prediction, operation data analysis and dynamic optimization models, and continuously adjusts the power and working status of the equipment to achieve energy-saving goals. The equipment operation mode optimization algorithm means that within the framework of energy efficiency optimization, the operation mode of each device will be automatically adjusted according to the energy consumption status. Through the dynamic adjustment algorithm, the equipment operation mode is automatically switched according to the load forecast and energy consumption data to ensure the best energy efficiency under different load conditions. The distributed load forecasting unit uses multiple computing nodes to predict the load demand of each region in real time, and then provides optimization suggestions for the energy efficiency management module. Through cloud computing and edge computing architecture, the load forecasting task is distributed to multiple computing nodes, and regional forecasting is performed by combining historical data and real-time data; The energy efficiency collaborative control strategy integrates multi-objective optimization algorithms to establish a collaborative mechanism between the central control system and edge computing nodes. By coordinating devices at different levels, the system can achieve optimal energy consumption on a global scale. The energy efficiency management module uses a dynamic energy allocation algorithm to adjust the operating power of the equipment according to load prediction and equipment status to achieve energy saving goals; Furthermore, the central control system is responsible for global energy efficiency scheduling and optimization, and adjusts the overall load distribution and resource usage of the system based on the feedback information of multiple edge computing units. For example, when some devices are in a low-load state, the system can reduce their operating power or switch them to energy-saving mode, while high-load devices are scheduled first. The system achieves overall improvement in energy efficiency through global resource allocation; through local and global optimization, the system can achieve intelligent collaboration between devices. For example, when the power demand of a certain device increases, the adjacent devices can assist the device to share part of the load within the power allowable range, thereby reducing the energy consumption pressure of a single device; the energy efficiency optimization strategy introduces a multi-mode operation mechanism for the device. When the load is low, the system will switch some devices to energy-saving mode or standby mode to reduce energy consumption; under high load conditions, the device will automatically switch to high-efficiency operation mode to ensure that the demand is met while running with minimum energy consumption; during the energy efficiency optimization process, the system will continuously monitor the energy consumption status of each device and continuously adjust the device operation strategy through sensors and feedback loops. For example, when the energy consumption of the generator is too high or the load is too low, the system will automatically reduce the power generation or adjust the operating parameters to reduce energy waste; the system will optimize its energy efficiency strategy by collecting and analyzing long-term operating data to adapt to changes in energy efficiency requirements under different working conditions. This adaptive mechanism is based on a machine learning model and can continuously improve the accuracy and effectiveness of energy efficiency optimization; The system can dynamically adjust the power output and operating mode of the equipment according to the real-time operating conditions of the equipment and predicted load demand, maximize energy utilization efficiency, and reduce energy waste. It solves the technical problems of energy waste, overload, inefficient operation, and uneven resource allocation in traditional power plant control systems, significantly reduces energy consumption, improves equipment operating efficiency, and optimizes overall operating costs.

[0026] See also Figure 1 The present invention provides an embodiment: a power plant electrical automation control method based on hierarchical management, the control method comprising the following steps: S1. Based on the load forecasting module system, the system combines historical data with real-time operation data and uses the LSTM neural network model to predict future load conditions in real time, and feeds the prediction results back to the dynamic resource allocation control unit. The dynamic resource allocation control unit makes a preliminary resource allocation plan based on the load change trend, providing a basis for subsequent resource scheduling; S2. The device layer sensors send the collected data to the edge computing unit through the Internet of Things technology. The edge computing unit processes the collected data on the spot. Based on the real-time operation data, the edge computing unit uses the local optimization control algorithm to make real-time adjustments to the operation of the equipment to ensure that the equipment reaches the optimal operating state under the current load; S3, the fault detection module of the system analyzes the operating status of the equipment in real time by fusing data from different sensors and discovers potential faults. The fault isolation module immediately performs isolation operations, removes the faulty equipment from the control system, and starts the redundant control unit to take over the equipment task; S4. Based on the load forecast data provided by the distributed load forecasting unit, the system automatically adjusts the equipment operation mode through the equipment operation mode optimization algorithm, controls the power output and operation status of the equipment in real time, and ensures the optimal energy efficiency under different load conditions.

[0027] The control method further comprises the following steps: S11. Through the priority-based task scheduling algorithm, the system allocates more computing resources and communication loans to high-load and critical devices according to the importance and current load of each device. The resource allocation plan is transmitted to each layer of devices through the multi-layer communication network; S21. The processed local data is fed back to the central control system by the edge computing unit. The central control system analyzes the data based on the global perspective. The central control system and the edge computing unit jointly optimize the control strategy through the distributed control algorithm to achieve the best global performance and ensure the efficient operation of local devices in specific scenarios. S31. The redundant control unit uses the reserved resources and control strategies to continue to execute the key tasks of the isolated equipment to ensure that the system functions are not affected. The system uses the fast recovery logic to reallocate resources according to the isolation of the faulty equipment to ensure that other equipment can still operate normally while waiting for the faulty equipment to be repaired. S41, the energy efficiency management module cooperates with the dynamic resource allocation control unit to coordinate the operation strategies of devices at different levels to achieve global energy efficiency optimization. The energy efficiency performance of the equipment is fed back to the central control system through the edge computing unit. The system dynamically adjusts the energy efficiency optimization strategy based on the feedback data to form a closed-loop optimization and further improve the overall energy efficiency. Furthermore, in the hierarchical structure, the bottom-level device sensors will collect the operating status of the device in real time, and the data will be initially analyzed and filtered by the edge computing unit. The edge unit will perform preliminary local decisions based on real-time data, and upload important status information to the central control system for global optimization; The edge computing unit uses local optimization algorithms at the device level and combines real-time data to autonomously adjust the operating status of the device. For example, when the load on the generator is too high, the edge computing unit can instantly adjust its operating parameters to keep the device running at an efficient state.

[0028] In addition, local optimization also includes fault detection and preliminary isolation to ensure equipment safety; the central control system makes global resource scheduling and optimization decisions based on the status data uploaded by the device layer and the optimization results fed back by the edge computing unit. The system combines historical data, real-time status and prediction models to dynamically adjust load distribution, optimize power scheduling, and ensure maximum overall operating efficiency; Fault detection and self-healing mechanisms are embedded in the system. The device layer and edge computing units automatically identify abnormal situations by monitoring the operating status of each device in real time. Once a fault is detected, the edge unit immediately performs fault isolation and notifies the central system for global scheduling and repair. For minor faults, the system can mitigate the problem by adjusting operating parameters; for major faults, redundant equipment is started to maintain normal operation; energy efficiency optimization dynamically adjusts the power output of the equipment through real-time monitoring and scheduling of local equipment to ensure that the equipment operates in the most efficient way under different load conditions.

[0029] At the same time, the system uses predictive analysis to pre-adjust equipment status based on future load demand to avoid excessively high or low power operation and maximize energy efficiency.

[0030] Working principle: First, the control system manages and monitors the power plant equipment through a hierarchical structure. The underlying sensors and edge computing units collect the equipment's operating status data in real time, such as temperature, voltage, current, etc. The edge computing unit performs preliminary processing on the collected data, makes local optimization decisions, adjusts the equipment's operating status in real time to ensure its efficient operation, and transmits key data to the central control system for global analysis. The central control system then executes global optimization and scheduling strategies based on real-time data, historical operation records, and predictive analysis provided by the edge computing unit. It allocates the resources of the power plant through intelligent algorithms, dynamically allocates equipment loads, and optimizes overall power output and energy utilization. The process also incorporates energy efficiency optimization strategies to enable the system to operate at the highest energy efficiency under different load conditions, ensuring the rational use of resources and minimization of energy consumption; Finally, during the operation of the equipment, the system is embedded with a fault self-healing mechanism, which can monitor the health status of the equipment in real time. Once a fault occurs, the edge computing unit will quickly isolate the fault, start the backup equipment or adjust the operation strategy to ensure the continuous and stable operation of the system. Through global and local collaborative work, the system ensures the high efficiency, reliability and low energy consumption of the power plant's electrical automation control, maximizing the operating efficiency and economic benefits of the power plant.

[0031] It will be apparent to those skilled in the art that the invention is not limited to the details of the exemplary embodiments described above and that the invention can be implemented in other specific forms without departing from the spirit or essential features of the invention. Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description, and it is intended that all variations falling within the meaning and scope of the equivalent elements of the claims be included in the invention. Any reference numeral in a claim should not be considered as limiting the claim to which it relates.

Claims

1. A power plant electrical automation control system based on hierarchical management, including a dynamic resource allocation coordination mechanism, a fault self-healing mechanism, a local optimization control based on edge computing, and an energy efficiency optimization strategy, characterized in that: The resource output of the dynamic resource allocation coordination mechanism is used as the optimization resource input of the local optimization control based on edge computing, the energy efficiency optimization mechanism is used as the dynamic interaction layer between the dynamic resource allocation coordination mechanism and the local optimization control based on edge computing, and the fault self-healing mechanism is used as the result execution mechanism of the energy efficiency optimization strategy; The dynamic resource allocation coordination mechanism includes: a dynamic resource allocation control unit, a load prediction module and a priority-based task scheduling algorithm. The dynamic resource allocation coordination mechanism is used to adjust the computing resources and communication bandwidth allocation of each layer of the control system to ensure that key equipment responds first under high load conditions; The dynamic resource allocation control unit uses an embedded controller in conjunction with an artificial intelligence algorithm to predict system load trends and dynamically allocate system resources based on historical operating data and real-time data; The load forecasting module obtains the current load data of the power plant through real-time data acquisition equipment, and inputs the load data into the prediction models of the cloud and edge computing nodes to generate load forecasting results, which are fed back to the dynamic resource allocation control unit for dynamic adjustment; The priority-based task scheduling algorithm introduces a priority queue into the system scheduler, and dynamically sorts tasks according to their urgency, ensuring that high-priority tasks can obtain resources first under high-load conditions.

2. According to claim 1, a power plant electrical automation control system based on hierarchical management is characterized in that: The fault self-healing mechanism includes: a multi-layer communication network, a fault detection and isolation module, a redundant control unit and a fast recovery logic; A multi-layer communication network means that in a layered system, the device layer, control layer and management layer are transmitted through independent communication networks. The power plant communication network based on industrial Ethernet and IEC 61850 protocol ensures the reliability and real-time performance of communication between different layers through the isolation of the physical layer and the logical layer. The fault detection and isolation module monitors the status of system equipment in real time, determines equipment abnormalities through multi-sensor data fusion technology, and processes the data collected by equipment sensors using multi-sensor data fusion algorithms. Once an abnormality is detected, an isolation command is issued through the SCADA system, and other equipment is reconfigured to take over the function. The redundant control unit immediately starts taking over the control task of the device after receiving the isolation instruction from the fault detection and isolation module. The introduction of redundant design ensures that the redundant control unit can seamlessly take over the control task when a single device fails. The fast recovery logic coordinates the status of the multi-layer control system through the state machine control logic. After receiving the system fault repair recovery signal, the system automatically switches to normal mode, automatically releases redundant control, and restores resource allocation under normal conditions.

3. The power plant electrical automation control system based on hierarchical management according to claim 1 is characterized in that: The local optimization control based on edge computing is collaboratively completed by the edge computing unit, the local data processing module, the distributed control algorithm and the local optimization model; The edge computing unit is deployed at the equipment layer and directly connected to the equipment in the power plant. It can locally process and optimize the equipment operation data. The edge computing unit is composed of an industrial controller and equipped with an edge computing algorithm to analyze the equipment's operating efficiency and failure rate data in real time and adjust the equipment's working status. The local data processing module processes data at the device level, makes local decisions by analyzing data in real time, combines IoT technology, data processing chips, and local memory, and uses a data stream processing engine to process high-frequency data in real time; The system uses a distributed control algorithm to enable each edge node to independently perform control tasks and work with other nodes through the network to achieve global optimization. It uses a distributed optimal control algorithm and interval optimization method to ensure that the control instructions of each edge node achieve a balance between local optimization and global optimization. The local optimization model means that the system uses a data-driven model to perform local optimization based on the working conditions of local equipment, establishes a local optimization model for the equipment through a machine learning algorithm, predicts equipment failures through a data-driven state detection model, adjusts the operation strategy in advance, and updates and optimizes the equipment's operating parameters in real time.

4. The power plant electrical automation control system based on hierarchical management according to claim 1 is characterized in that: The energy efficiency optimization strategy is jointly executed by the energy efficiency management module, the equipment operation mode optimization algorithm, the distributed load prediction unit and the energy efficiency collaborative control strategy; The energy efficiency management module is located at the management level of the hierarchical architecture. It is responsible for monitoring the energy efficiency of the entire power plant and automatically adjusting the operating status of the equipment to achieve optimal energy efficiency based on real-time data and prediction results. It develops energy efficiency optimization algorithms based on AI technology, uses load prediction, operation data analysis and dynamic optimization models, and continuously adjusts the power and working status of the equipment to achieve energy-saving goals. The equipment operation mode optimization algorithm means that within the framework of energy efficiency optimization, the operation mode of each device will be automatically adjusted according to the energy consumption status. Through the dynamic adjustment algorithm, the equipment operation mode is automatically switched according to the load forecast and energy consumption data to ensure the best energy efficiency under different load conditions. The distributed load forecasting unit uses multiple computing nodes to predict the load demand of each region in real time, and then provides optimization suggestions for the energy efficiency management module. Through cloud computing and edge computing architecture, the load forecasting task is distributed to multiple computing nodes, and regional forecasting is performed by combining historical data and real-time data; The energy efficiency collaborative control strategy integrates multi-objective optimization algorithms to establish a collaborative mechanism between the central control system and edge computing nodes. By coordinating devices at different levels, the system can achieve optimal energy consumption on a global scale.

5. The power plant electrical automation control system based on hierarchical management according to claim 1, characterized in that: The load forecasting model adopts an LSTM neural network model to perform long-term and short-term load trend forecasts based on historical load data.

6. The power plant electrical automation control system based on hierarchical management according to claim 1 is characterized by: The priority-based task scheduling algorithm is a preemptive priority scheduling algorithm, which can dynamically preempt system resources occupied by low-priority tasks when high-priority tasks appear.

7. The power plant electrical automation control system based on hierarchical management according to claim 3 is characterized by: The distributed control algorithm is a model predictive control algorithm, and each edge computing node independently performs local optimization tasks through the distributed control algorithm while considering the global goal.

8. The power plant electrical automation control system based on hierarchical management according to claim 4 is characterized by: The energy efficiency management module uses a dynamic energy allocation algorithm to adjust the operating power of the equipment according to load prediction and equipment status to achieve energy saving goals.

9. A power plant electrical automation control method based on hierarchical management, applicable to a power plant electrical automation control system based on hierarchical management as claimed in any one of claims 1 to 8, characterized in that: The control method comprises the following steps: S1. Based on the load forecasting module system, the system combines historical data with real-time operation data and uses the LSTM neural network model to predict future load conditions in real time, and feeds the prediction results back to the dynamic resource allocation control unit. The dynamic resource allocation control unit makes a preliminary resource allocation plan based on the load change trend, providing a basis for subsequent resource scheduling; S2. The device layer sensors send the collected data to the edge computing unit through the Internet of Things technology. The edge computing unit processes the collected data on the spot. Based on the real-time operation data, the edge computing unit uses the local optimization control algorithm to adjust the operation of the equipment in real time to ensure that the equipment reaches the optimal operating state under the current load; S3, the fault detection module of the system analyzes the operating status of the equipment in real time by fusing data from different sensors and discovers potential faults. The fault isolation module immediately performs isolation operations, removes the faulty equipment from the control system, and starts the redundant control unit to take over the equipment task; S4. Based on the load forecast data provided by the distributed load forecasting unit, the system automatically adjusts the equipment operation mode through the equipment operation mode optimization algorithm, controls the power output and operation status of the equipment in real time, and ensures the optimal energy efficiency under different load conditions.

10. The power plant electrical automation control method based on hierarchical management according to claim 9, characterized in that: The control method further comprises the following steps: S11. Through the priority-based task scheduling algorithm, the system allocates more computing resources and communication loans to high-load and key devices according to the importance and current load of each device. The resource allocation plan is transmitted to each layer of devices through the multi-layer communication network; S21. The processed local data is fed back to the central control system by the edge computing unit. The central control system analyzes the data based on the global perspective. The central control system and the edge computing unit jointly optimize the control strategy through the distributed control algorithm to achieve the best global performance and ensure the efficient operation of local devices in specific scenarios. S31. The redundant control unit uses the reserved resources and control strategies to continue to execute the key tasks of the isolated equipment to ensure that the system functions are not affected. The system uses the fast recovery logic to reallocate resources according to the isolation of the faulty equipment to ensure that other equipment can still operate normally while waiting for the faulty equipment to be repaired. S41. The energy efficiency management module cooperates with the dynamic resource allocation control unit to coordinate the operation strategies of equipment at different levels to achieve global energy efficiency optimization. The energy efficiency performance of the equipment is fed back to the central control system through the edge computing unit. The system dynamically adjusts the energy efficiency optimization strategy based on the feedback data to form a closed-loop optimization and further improve the overall energy efficiency.

Images