Power plant load visual control method and system based on digital twinning

Abstract

The invention relates to the technical field of digital twinning, and discloses a power plant load visualization control method based on digital twinning, and the method comprises the steps: obtaining power plant entity data, constructing and integrating a power plant appearance three-dimensional model and an internal equipment three-dimensional model, and obtaining a static three-dimensional model; constructing a dynamic digital model based on a mechanism modeling method and a data-driven modeling method, and integrating the static three-dimensional model and the dynamic digital model to obtain a digital twinborn model; processing the real-time operation data and the historical operation data of the power plant based on a reinforcement learning algorithm and a deep learning algorithm to obtain a prediction result; performing abnormal state judgment on the prediction result and an early warning threshold value based on the static three-dimensional model to obtain a judgment result; according to a judgment result, evaluating the severity of the abnormal state by using a rule model so as to formulate countermeasures; and based on the digital twinborn model, a dispatching instruction is calculated by using countermeasures and a PID control algorithm, so that the operation state of the power plant is visually monitored and controlled.

Description

Technical Field

[0001] The present invention relates to the technical field of digital twins, and in particular to a method and system for visualizing control of power plant loads based on digital twins. Background Art

[0002] At present, as the core facilities to ensure the electricity supply to residents, the stable operation of power plants is crucial to the health of the national power grid and the normal operation of people's livelihood. Visual monitoring of power plants is an important means to ensure the efficient operation of the power system, but traditional visual monitoring methods often have some limitations.

[0003] In traditional power plant monitoring systems, surveillance cameras and data acquisition systems usually only provide static or delayed information. This means that data is often collected and processed only after an event occurs. This delay not only fails to reflect the operating status of the power plant in real time, but also fails to detect potential problems or abnormal conditions in a timely manner. Therefore, when a power plant fails or has equipment problems, the response speed is often slow and the handling measures are not quick enough, which may lead to power supply interruption or system damage, thus affecting residents' lives and industrial production.

[0004] More importantly, traditional systems often fail to achieve efficient emergency dispatch when facing various power generation modes (such as wind power, hydropower, thermal power, etc.). Different types of power generation have their own unique operating characteristics and emergency needs, and traditional monitoring systems often find it difficult to respond quickly and accurately to these diverse needs. For example, wind power generation is greatly affected by wind speed and needs to adjust the operating status in real time to optimize power generation efficiency; hydropower generation may need to deal with the challenges brought by changes in water flow; thermal power generation requires real-time monitoring of fuel supply and equipment operation. At present, as a guarantee of residents' electricity use, power plants are an important part of ensuring people's livelihood by transmitting electricity to the national grid, and visual monitoring of power plants is a necessary and necessary means, which plays an important role in the real-time control of power plants in the later stage. How to achieve efficient, accurate and reliable operation of the three-energy dispatching system of power plants is a technical problem that needs to be solved urgently. Summary of the invention

[0005] The present invention provides a power plant load visualization control method based on digital twins to solve the technical problem in the prior art that traditional three-dimensional models are inefficient in modeling complex objects, resulting in difficulties in power dispatching.

[0006] According to one aspect of the present invention, the present invention provides a power plant load visualization control method based on digital twins.

[0007] Acquire the entity data of the power plant to construct and integrate the 3D model of the power plant appearance and the 3D model of the internal equipment to obtain a static 3D model;

[0008] Based on the mechanism modeling method and the data-driven modeling method, a dynamic digital model is constructed, and the static three-dimensional model and the dynamic digital model are integrated to obtain a digital twin model;

[0009] Based on reinforcement learning algorithms and deep learning algorithms, the real-time and historical operating data of the power plant are processed to obtain prediction results;

[0010] Based on the static three-dimensional model, the prediction result is compared with the early warning threshold to identify the abnormal state of the power plant and obtain a judgment result;

[0011] According to the judgment result, the severity of the abnormal state is evaluated using a rule model to formulate a response measure;

[0012] Based on the digital twin model, the countermeasures and PID control algorithm are used to calculate scheduling instructions to monitor and control the operating status of the power plant.

[0013] Furthermore, the acquisition of the power plant entity data to construct and integrate the power plant appearance three-dimensional model and the internal equipment three-dimensional model to obtain a static three-dimensional model includes:

[0014] Acquire the entity data of the power plant in real time, and construct a three-dimensional model of the appearance of the power plant and a three-dimensional model of the internal equipment;

[0015] Integrating the three-dimensional model of the power plant appearance and the three-dimensional model of the internal equipment to obtain an integrated three-dimensional model;

[0016] Based on the network rendering technology, the integrated three-dimensional model is rendered to obtain the static three-dimensional model.

[0017] Furthermore, the integrating the static three-dimensional model and the dynamic digital model to obtain the digital twin model includes: integrating the static three-dimensional model and the dynamic digital model by using parameter coupling and state variable association to obtain the digital twin model.

[0018] Furthermore, the real-time operation data and historical operation data of the power plant are processed based on the reinforcement learning algorithm and the deep learning algorithm to obtain the prediction result;

[0019] Acquiring the real-time operation data and the historical operation data, and preprocessing the real-time operation data and the historical operation data to obtain a standardized data set;

[0020] Based on the deep learning algorithm, feature extraction is performed on the standardized data set to obtain a key feature set;

[0021] Constructing a prediction model according to the reinforcement learning algorithm;

[0022] Based on the key feature set, the prediction model is trained to obtain a trained prediction model;

[0023] Based on the trained prediction model, the real-time operation data is predicted to obtain a prediction result of the power plant operation status.

[0024] Furthermore, based on the static three-dimensional model, the prediction result is compared with the warning threshold to identify the abnormal state of the power plant, and the judgment result obtained includes:

[0025] Inputting the prediction results into the digital twin model, and dynamically generating real-time operating status data in the static three-dimensional model;

[0026] Calculating the warning threshold using the historical operation data based on statistical analysis;

[0027] Based on an abnormality judgment algorithm, the real-time operating status data is compared with the early warning threshold to identify the abnormal state of the power plant and obtain the judgment result.

[0028] Furthermore, based on the judgment result, the severity of the abnormal state is evaluated using a rule model to formulate response measures, including:

[0029] According to the judgment result, a convolutional neural network is used to classify the abnormal state to obtain a fault type;

[0030] Based on a machine learning algorithm, the severity of the abnormal state is evaluated to obtain a fault severity level;

[0031] Constructing a fault simulation scenario according to the fault type and the fault severity level;

[0032] Based on the rule model, the fault simulation scenario is analyzed using a preset rule library to obtain preliminary measures;

[0033] Based on the expert knowledge base, the preliminary measures are optimized and adjusted to obtain countermeasures.

[0034] Furthermore, based on the digital twin model, the countermeasures and the PID control algorithm are used to calculate the dispatching instructions to monitor and control the operating state of the power plant, including:

[0035] Formulate a preliminary energy dispatch strategy based on the forecast results;

[0036] According to the countermeasures, the preliminary energy scheduling strategy is optimized using the PID control algorithm to obtain the optimal energy scheduling strategy;

[0037] According to the optimal energy scheduling strategy, the scheduling instruction is generated by using the rule model;

[0038] Inputting the dispatch instruction into the digital twin model to simulate the operating state and effect of the power plant after executing the dispatch instruction to obtain a simulation result;

[0039] Adopting the dispatch instruction to adjust the operating state of the power plant to obtain an execution result;

[0040] Perform deviation analysis on the execution result and the simulation result to obtain a deviation correction scheme;

[0041] According to the deviation correction scheme, the preliminary energy scheduling strategy is adjusted to update the scheduling instruction.

[0042] According to another aspect of the present invention, a power plant load visualization control system based on digital twin is provided, the system comprising:

[0043] A three-dimensional modeling unit is used to obtain the entity data of the power plant to construct and integrate the three-dimensional model of the power plant appearance and the three-dimensional model of the internal equipment to obtain a static three-dimensional model;

[0044] A digital modeling unit, used to construct a dynamic digital model based on a mechanism modeling method and a data-driven modeling method, and integrate the static three-dimensional model with the dynamic digital model to obtain a digital twin model;

[0045] A data prediction unit is used to process the real-time operation data and historical operation data of the power plant based on a reinforcement learning algorithm and a deep learning algorithm to obtain a prediction result;

[0046] A state identification unit, configured to compare the prediction result with a warning threshold based on the static three-dimensional model to identify an abnormal state of the power plant and obtain a judgment result;

[0047] A measure generating unit, configured to evaluate the severity of the abnormal state using a rule model according to the judgment result, so as to formulate a countermeasure;

[0048] A model control unit is used to calculate scheduling instructions based on the digital twin model using the countermeasures and PID control algorithm to monitor and control the operating status of the power plant.

[0049] According to another aspect of the present invention, there is provided an electronic device, comprising: at least one processor, and a memory communicatively connected to the at least one processor;

[0050] The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the method described in any one of claims 1 to 7.

[0051] According to another aspect of the present invention, a non-transitory computer-readable storage medium storing computer instructions is provided, wherein the computer instructions are used to cause the computer to execute the method according to any one of claims 1-7.

[0052] Compared with the prior art, the power plant load visualization control method based on digital twin in the embodiment of the present invention has the following beneficial effects:

[0053] The embodiment of the present invention constructs a static three-dimensional model and a dynamic digital model of the power plant, and integrates the static three-dimensional model and the dynamic digital model to generate a digital twin model, thereby realizing a dynamic digital twin of the power plant. Not only can the operating status of each energy unit in the virtual power plant be monitored in real time through dynamic bionic simulation of the power plant, but historical operating data and real-time operating data can also be used to simulate and predict possible faults in the power plant, thereby facilitating remote control of each energy unit in the power plant, so that managers can formulate and adjust power dispatch response measures in a timely manner, and realize remote operation and maintenance and fault handling. The powerful modeling ability and dynamic presentation ability of the digital twin provide a decision-making basis for load forecasting and peak-shaving dispatching of the power grid, and provide strong intelligent support for the precise electroplating and maintenance of the power plant, thereby improving the efficient use of energy and enhancing the system's risk resistance.

[0054] It should be understood that the contents described in this section are not intended to identify the key or important features of the embodiments of the present invention, nor are they intended to limit the scope of the present invention. Other features of the present invention will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] The accompanying drawings are used to better understand the present invention and do not constitute a limitation of the present invention.

[0056] Figure 1 is a flow chart of a power plant load visualization control method based on digital twins provided in an embodiment of the present invention;

[0057] Figure 2 is a flowchart of step S10 in the power plant load visualization control method based on digital twins provided in an embodiment of the present invention;

[0058] Figure 3 is a flow chart of step S30 in the power plant load visualization control method based on digital twins provided in an embodiment of the present invention;

[0059] Figure 4is a flow chart of step S40 in the power plant load visualization control method based on digital twins provided in an embodiment of the present invention;

[0060] Figure 5 is a flow chart of step S50 in the power plant load visualization control method based on digital twins provided in an embodiment of the present invention;

[0061] Figure 6 is a flow chart of step S60 in the power plant load visualization control method based on digital twins provided in an embodiment of the present invention;

[0062] Figure 7 is a schematic diagram of a power plant load visualization control system based on digital twins provided in an embodiment of the present invention;

[0063] Figure 8 Schematic diagram of an electronic device and a storage medium according to an embodiment of the present invention.

[0064] In the figure, 100, a power plant load visualization control system based on digital twin; 101, a three-dimensional modeling unit; 102, a digital modeling unit; 103, a data prediction unit; 104, a state recognition unit; 105, a measure generation unit; 106, a model control unit; 200, an electronic device; 201, a computing unit; 202, a ROM; 203, a RAM; 204, a bus; 205, an I / O interface; 206, an input unit; 207, an output unit; 208, a storage unit; 209, a communication unit. DETAILED DESCRIPTION

[0065] The following is a description of exemplary embodiments of the present invention in conjunction with the accompanying drawings, including various details of the embodiments of the present invention to facilitate understanding, which should be considered as merely exemplary. Therefore, it should be appreciated by those of ordinary skill in the art that various changes and modifications may be made to the embodiments described herein without departing from the scope of the present invention. Similarly, for clarity and conciseness, descriptions of well-known functions and structures are omitted in the following description.

[0066] In order to make the purpose, technical scheme and advantages of the present application clearer, the specific embodiments of the present application are further described in detail below in conjunction with the accompanying drawings. It is understood that the specific embodiments described herein are only used to explain the present application, rather than to limit the present application. It should also be noted that, for the convenience of description, only the part related to the present application but not all the contents are shown in the accompanying drawings. Before discussing the exemplary embodiments in more detail, it should be mentioned that some exemplary embodiments are described as processes or methods depicted as flow charts. Although the flow chart describes each operation (or step) as a sequential process, many of the operations therein can be implemented in parallel, concurrently or simultaneously. In addition, the order of each operation can be rearranged. The process can be terminated when its operation is completed, but it can also have additional steps not included in the accompanying drawings. The process can correspond to a method, a function, a procedure, a subroutine, a subprogram, etc.

[0067] Terminology explanation section:

[0068] Digital twin: is a technology that associates virtual models of physical entities (such as machines, equipment, buildings, factories, etc.) with their physical world in real time. Through digital twin technology, physical entities can be simulated, analyzed and optimized in a virtual environment to help improve their performance, predict failures and optimize operations.

[0069] PID control algorithm: It is a classic control algorithm widely used in industrial control systems. It adjusts the output of the control system to make the output of the controlled object as close to the target value as possible, thereby achieving precise automatic control.

[0070] Navier-Stokes equation: It is the basic equation that describes fluid motion and is used to study velocity and pressure fields in fluid mechanics. It is based on Newton's second law and combines the laws of conservation of mass, conservation of momentum, and conservation of energy. The Navier-Stokes equation can be used to describe the motion behavior of continuous media such as gases and liquids, and is one of the foundations of modern fluid dynamics.

[0071] MATLAB Simul ink: is a powerful and widely used simulation and model design tool, which is mainly used for modeling, simulation and analysis of multi-domain dynamic systems. Simul ink is an extension module of MATLAB, which is tightly integrated with MATLAB and allows users to create and simulate complex system models through a graphical interface.

[0072] IQR method: It is a statistical method commonly used in data analysis to identify outliers in a data set. IQR represents the range between the 75th percentile (upper quartile, Q3) and the 25th percentile (lower quartile, Q1) in a data set.

[0073] MODBUS protocol: is a communication protocol widely used in industrial automation systems, especially for serial communication between control and monitoring devices. The MODBUS protocol was first developed by Modicon (now part of Schneider Electric) in 1979 to support communication between programmable logic controllers (PLCs) and other devices.

[0074] Parameter coupling: refers to the interdependence or mutual influence between multiple parameters in a system or model. This means that when one parameter is adjusted, the values ​​of other parameters may also be affected, and each parameter cannot be adjusted independently.

[0075] State variable association: refers to the relationship or dependency between multiple state variables in a system. This association is a key part of describing the dynamic behavior and state changes of the system, especially in control systems, physical modeling, economics, and other dynamic system analysis.

[0076] The K-means clustering algorithm is a classic unsupervised learning algorithm that is used to divide samples in a data set into KK mutually exclusive clusters. Its goal is to make the data points within each cluster as similar as possible, while the data points between different clusters are as different as possible. The K-means algorithm is widely used in data mining, image processing, market segmentation and other fields.

[0077] WebGL: is a browser-based 3D graphics rendering technology. It allows developers to create complex 3D graphics and interactive content directly on web pages without installing any plug-ins.

[0078] Uni 3D: is a widely used real-time 3D development platform that supports various types of content creation, from 2D and 3D games to virtual reality, augmented reality, movies, animation, etc.

[0079] Anomaly detection algorithms: used to identify abnormal data points or behaviors in a data set that are significantly different from normal patterns. In many application scenarios, such as network security, financial fraud detection, industrial monitoring, medical diagnosis, etc., anomaly detection algorithms can help identify potential problems or abnormal events so that timely measures can be taken. Such as Z-Score, K-nearest neighbor, support vector machine and isolation forest.

[0080] Bias analysis algorithms: are a class of algorithms used to detect and analyze biases or differences in data sets. Bias can refer to statistical deviations (e.g., systematic errors) or errors in model predictions. Bias analysis algorithms are widely used in fields such as data science, machine learning, statistics, and economics to help identify systematic errors in data or models and ensure the accuracy and fairness of results.

[0081] Figure 1 It is a flow chart of a power plant load visualization control method based on digital twin provided in an embodiment of the present invention.

[0082] like Figure 1 As shown, an embodiment of the present invention provides a power plant load visualization control method based on digital twins, the method comprising:

[0083] S10, obtaining the entity data of the power plant to construct and integrate the three-dimensional model of the power plant's appearance and the three-dimensional model of the internal equipment to obtain a static three-dimensional model; S20, based on the mechanism modeling method and the data-driven modeling method, construct a dynamic digital model, and integrate the static three-dimensional model and the dynamic digital model to obtain a digital twin model; S30, based on the reinforcement learning algorithm and the deep learning algorithm, process the real-time operation data and historical operation data of the power plant to obtain the prediction result; S40, based on the static three-dimensional model, judge the abnormal state of the prediction result and the warning threshold to obtain the judgment result; S50, based on the judgment result, use the rule model to evaluate the severity of the abnormal state to formulate countermeasures; S60, based on the digital twin model, use the countermeasures and PID control algorithm to calculate the scheduling instructions to monitor and control the operation status of the power plant.

[0084] The embodiment of the present invention realizes the dynamic digital twin of the power plant by constructing a static three-dimensional model and a dynamic digital model of the power plant, integrating the static three-dimensional model and the dynamic digital model to generate a digital twin model, and not only can the operating status of each energy unit in the virtual power plant be monitored in real time through the dynamic bionic simulation of the power plant, but also the historical operation data and real-time operation data can be used to simulate and predict the possible fault conditions in the power plant, which is convenient for remote control of each energy unit of the power plant, so that managers can formulate and adjust the response measures of power dispatch in time, realize remote operation and maintenance and fault handling, and the powerful modeling ability and dynamic presentation ability of the digital twin provide a decision-making basis for the load forecasting and peak-shaving dispatching of the power grid, and provide strong intelligent support for the precise electroplating and maintenance of the power plant, thereby improving the efficient use of energy and enhancing the risk resistance of the system. Through intelligent data analysis and control strategies, the system can significantly improve the operating efficiency of the power plant, reduce the downtime caused by abnormalities or failures, and thus reduce operating costs. The prediction and early warning mechanism helps the power plant to identify and solve potential problems in advance, avoiding the high processing costs of emergencies.

[0085] Among them, the 3D modeling link constructs twin 3D appearance scenes and internal equipment, laying a solid foundation for the virtual power plant. In order to accurately replicate the virtual environment of each physical system of the power plant, the key is to build a high-quality 3D model as a digital mirror of the physical entity. In the digital twin application under the network environment, it is necessary to carefully balance the 3D modeling precision and loading efficiency in the browser. Digital modeling, as an intelligent engine for 3D interaction and motion control, ensures that all dynamic changes are under control. By obtaining the entity data of the power plant and building a static 3D model, the appearance and internal equipment of the power plant can be intuitively visualized. This 3D model not only provides a real scene reproduction of the power plant, but also provides a strong foundation for subsequent operations and analysis. The integration of static 3D models and dynamic digital models enables the operation of the power plant to be simulated in a virtual environment, thus providing a comprehensive digital twin environment for operators.

[0086] It should be noted that the dynamic digital model constructed based on the mechanism modeling method and the data-driven modeling method can accurately simulate the various dynamic behaviors of the power plant. This model can reflect the actual operating characteristics of the power plant, thereby providing higher accuracy in the simulation and prediction process. The application of reinforcement learning and deep learning algorithms enables the system to conduct in-depth analysis of the real-time operating data and historical data of the power plant, provide accurate prediction results, and help to foresee potential problems and take preventive measures. The generation of scheduling instructions based on the digital twin model and PID control algorithm makes the operation control of the power plant more accurate and efficient. The PID control algorithm can adjust the operating parameters of the power plant in real time to optimize the operating status, reduce energy consumption and improve the overall efficiency of the power plant. By combining with countermeasures, the system can quickly calculate the best control strategy after detecting an abnormality to ensure that the power plant can maintain a stable operating state under various circumstances.

[0087] Figure 2 It is a flowchart of step S10 in the power plant load visualization control method based on digital twin provided in an embodiment of the present invention.

[0088] like Figure 2 As shown, in an optional embodiment of the present invention, the entity data of the power plant is obtained to construct and integrate the three-dimensional model of the power plant appearance and the three-dimensional model of the internal equipment to obtain a static three-dimensional model, including: S11, real-time acquisition of the entity data of the power plant, construction of the three-dimensional model of the power plant appearance and the three-dimensional model of the internal equipment; S12, data integration of the three-dimensional model of the power plant appearance and the three-dimensional model of the internal equipment to obtain an integrated three-dimensional model; S13, based on network rendering technology, rendering the integrated three-dimensional model to obtain a static three-dimensional model.

[0089] Specifically, in order to obtain the entity data of the power plant, a laser scanner can be used to perform a three-dimensional scan of the power plant building structure with a scanning accuracy of up to 1 mm. At the same time, a wireless sensor network is used to collect equipment operating parameters in real time, with a transmission frequency of 10 times per second. When constructing a three-dimensional model of the power plant's exterior, 3ds Max software is used for modeling, and high-definition photos taken from multiple perspectives are used for model texture mapping to ensure the visual realism of the model. For the internal equipment of the power plant, AutoCAD or SolidWorks software is used for parametric modeling. The accuracy of the parts of the equipment model can reach 1 mm, and the stress conditions of the equipment are simulated and calculated through finite element analysis. When integrating the internal and external models of the power plant, a spatial coordinate matching algorithm is used to achieve precise alignment of the model by setting key points, and the matching error is controlled within 1 cm.

[0090] Based on network rendering technology, the model can be rendered using a ray tracing-based rendering engine, with a rendering resolution of 4K and a stable frame rate of 60FPS, ensuring the high-quality display effect of the model. For the dynamic behavior simulation of the equipment, a multi-body dynamics algorithm is used to solve the Newton-Euler equations to achieve a high degree of simulation of the equipment movement, with a simulation accuracy of more than 95%. At the same time, the machine learning algorithm is used to train historical operation data, and the health status of the equipment is intelligently diagnosed through support vector machines and decision tree models, with an accuracy of up to 90% for fault prediction. Finally, the power plant is displayed panoramically through the independently developed digital twin platform. Users can explore every corner of the power plant immersively through VR devices, and remotely control the equipment through gesture interaction, which can greatly improve the level of intelligent management of the power plant.

[0091] In an optional embodiment of the present invention, integrating the static three-dimensional model and the dynamic digital model to obtain the digital twin model includes: using parameter coupling and state variable association to integrate the static three-dimensional model and the dynamic digital model to obtain the digital twin model.

[0092] Specifically, the specific implementation process of dynamic digital modeling is: using the mechanism modeling method, through thermodynamics and fluid mechanics equations, such as the Navier-Stokes equation, to establish a preliminary dynamic digital model and determine the basic structure and parameters of the model. Next, through the data-driven modeling method, using machine learning algorithms such as random forests or support vector machines to optimize and adjust the preliminary dynamic model, and obtain supplementary data for the preliminary dynamic digital model to determine the accuracy of the model. Adjust the model parameters through multiple iterative training to improve the accuracy and adaptability of the model.

[0093] The optimized preliminary dynamic digital model continues to couple parameters, and uses optimization algorithms such as genetic algorithms or particle swarm optimization to adjust the parameters of the preliminary dynamic digital model, determine the correlation between the coupling parameters, and ensure the stability of the model. Through state variable association, use principal component analysis (PCA) or factor analysis to obtain the key variables of model integration and determine the integrity of the model. Using the digital twin method, using numerical simulation tools such as MATLAB Simul ink, the static three-dimensional model and the digital dynamic model are integrated to obtain the final form of the integrated model and judge the applicability of the model. According to the output of the model, use data analysis tools such as Tableau or Power BI to obtain key indicators of power plant operation, such as power generation efficiency, fuel consumption rate, etc., and determine the optimization direction of the model. Through these steps, it is ensured that the model is not only highly accurate, but also can effectively guide the optimized operation of the power plant.

[0094] Figure 3 is a flow chart of step S30 in the power plant load visualization control method based on digital twins provided in an embodiment of the present invention;

[0095] like Figure 3 As shown, in an optional embodiment of the present invention, the real-time operation data and historical operation data of the power plant are processed based on the reinforcement learning algorithm and the deep learning algorithm to obtain the prediction results including: S31, acquiring the real-time operation data and the historical operation data, and preprocessing the real-time operation data and the historical operation data to obtain a standardized data set; S32, based on the deep learning algorithm, performing feature extraction on the standardized data set to obtain a key feature set; S33, constructing a prediction model according to the reinforcement learning algorithm; S34, training the prediction model based on the key feature set to obtain a trained prediction model; S35, based on the trained prediction model, predicting the real-time operation data to obtain the prediction result of the power plant operation status.

[0096] Specifically, when acquiring the real-time and historical operation data of the power plant, key variables such as fuel supply, air circulation, power demand, and waste gas and wastewater treatment can be monitored in real time by deploying sensor networks and data acquisition systems. For example, the MODBUS protocol is used to extract operation data from PLC devices in real time, combined with SQL queries in the historical operation database to form a complete data set containing hourly data. Then, data preprocessing techniques such as Z-score standardization and IQR method are applied to clean and standardize the real-time and historical operation data sets, remove noise and outliers, and obtain a standard data set with a unified format to ensure the integrity and accuracy of data quality.

[0097] Data cleaning was performed using Python's Pandas library, outliers with more than 3 standard deviations were deleted, and missing data were filled using linear interpolation. Then, deep learning algorithms, such as convolutional neural networks (CNNs), were used to extract features from the preprocessed data, identify key feature variables that affect the operating status of the power plant, and construct key feature sets. Next, reinforcement learning algorithms, such as deep Q networks (DQNs), were used to build prediction models, input the extracted key feature sets, and train the prediction models to learn the changing patterns of the power plant's operating status. PyTorch was used to train using historical operating data from the past year, and the strategy was optimized to minimize the prediction error. Based on the trained prediction model, real-time operating data was input to predict the operating status of the power plant in the next few hours. Through digital modeling technology, the prediction results were dynamically simulated with the physical system of the power plant to generate a virtual power plant operating status mirror image, and the final prediction results of the power plant's operating status were obtained.

[0098] Among them, in order to verify the rationality of the prediction results, MATLAB Simul ink can be used for dynamic simulation, and the digital twin engine can be used to compare and analyze the simulated operating status with the actual power plant operating data to identify potential anomalies and risks. Through comparative analysis, it was found that the predicted load in certain periods deviated from the actual load by more than 5%, indicating that there may be equipment failure. According to the results of the comparative analysis, the parameters of the digital twin model are adjusted to optimize the prediction accuracy and reliability of the model. Through the optimized digital twin model, a detailed operating status forecast report for the power plant in the next few hours is generated to provide data support for the intelligent operation and maintenance and decision-making of the power plant. Finally, the generated report lists the predicted load, fuel consumption and emission levels every hour in detail, providing a basis for decision-making for management.

[0099] Figure 4 It is a flowchart of step S40 in the power plant load visualization control method based on digital twin provided in an embodiment of the present invention.

[0100] like Figure 4 As shown, in an optional embodiment of the present invention, based on the static three-dimensional model, the prediction result and the warning threshold are compared for abnormal state judgment, and the judgment result includes: S41, inputting the prediction result into the digital twin model, and dynamically generating real-time operation status data in the static three-dimensional model; S42, based on the statistical analysis method, using the historical operation data to calculate the warning threshold; S43, based on the abnormal judgment algorithm, comparing the real-time operation status data with the warning threshold to identify the abnormal state of the power plant and obtain a judgment result.

[0101] Specifically, in the process of acquiring real-time operation data of the power plant, sensors and data acquisition systems are deployed on key equipment such as generators, energy storage devices and grid interfaces. These sensors collect data once a second, and the generated data stream is transmitted to the central data processing system through the IoT gateway. In the digital twin engine, high-precision 3D modeling software is used to build a virtual environment of the power plant to ensure the synchronization of the model with the physical system. Real-time operation data is input into the digital twin engine through the API interface to update the status of the virtual model. Digital twin models such as state space models are used for dynamic analysis of real-time data to dynamically simulate the operating status of the power plant.

[0102] In the operation status monitoring of power plants, the prediction results are calculated by obtaining real-time operation data and historical operation data, and the digital twin model is used for calculation to obtain the real-time operation status data of the power plant, such as the output power of the generator set is 500MW, the cooling water temperature is 25 degrees Celsius, etc. Then, these data are loaded into the static 3D model of the power plant, and the real-time status of the power plant is visualized through dynamic display technology such as WebGL or Unity 3D to ensure the real-time and accuracy of the display. On this basis, relevant statistical information is extracted from the historical operation data, such as the average output power of the generator set in the past year is 480MW, and the average value of the cooling water temperature is 22 degrees Celsius. The warning threshold of the power plant operation is calculated through statistical analysis methods, such as regression analysis or time series analysis, for example, the threshold of output power is set to 450MW to 550MW, and the threshold of cooling water temperature is set to 20 degrees Celsius to 30 degrees Celsius.

[0103] Then, the real-time operating status data is compared with the preset threshold, and the abnormal state in the power plant operation is identified using anomaly judgment algorithms, such as Z-score or anomaly detection algorithms in machine learning. For example, if the output power of the generator set drops to 440MW at a certain moment, it is identified as an abnormal state. If an abnormal state is identified, an abnormal report is generated through the information processing link. The report contains detailed information about the abnormality, such as the time when the abnormality occurred, the equipment involved, and possible cause analysis, such as excessive cooling water temperature causing reduced efficiency. Based on the information in the abnormal report, machine learning algorithms, such as decision trees or random forests, are used to further analyze the abnormal state and obtain possible solutions to the abnormal state, such as adjusting the cooling water flow or increasing the output of the standby generator set. Finally, the solution to the abnormal state is transmitted to the power plant control system through the information system to ensure that the power plant can adjust the operating status in time and resume normal operation.

[0104] Figure 5 It is a flowchart of step S50 in the power plant load visualization control method based on digital twin provided in an embodiment of the present invention.

[0105] like Figure 5As shown, in an optional embodiment of the present invention, based on the judgment result, the severity of the abnormal state is evaluated using a rule model to formulate countermeasures including:

[0106] S51, based on the judgment result, the convolutional neural network is used to classify the abnormal state to obtain the fault type; S52, based on the machine learning algorithm, the severity of the abnormal state is evaluated to obtain the fault severity level; S53, according to the fault type and fault severity level, a fault simulation scenario is constructed; S54, based on the rule model, a preset rule library is used to analyze the fault simulation scenario to obtain preliminary measures; S55, based on the expert knowledge base, the preliminary measures are optimized and adjusted to obtain countermeasures.

[0107] Specifically, in the operation monitoring of the power plant, the working parameters of each energy unit, such as temperature, pressure, flow, etc., are collected in real time through the sensor network, and combined with historical data, they are processed using the big data analysis platform. By using algorithms such as K-means clustering, these data are clustered and analyzed to identify the pattern of normal operating status. Next, data mining techniques such as support vector machines (SVM) are applied to detect anomalies in real-time monitoring data. For example, if the temperature of an energy unit exceeds the set threshold and the deviation from historical data exceeds two standard deviations, it is judged as an abnormal state.

[0108] If an abnormal state is detected, a convolutional neural network (CNN) is used to classify the abnormal data and determine the fault type, such as mechanical fault, electrical fault or sensor fault. Subsequently, the random forest algorithm is used to evaluate the severity of the fault and output the fault severity level, such as minor, moderate or severe. According to the fault type and severity level, the corresponding fault simulation scenario is generated, and the state when the fault occurs is simulated through the digital twin model, such as simulating the pressure change caused by a pipeline rupture. Then, the rule-based model is used in combination with the preset rule base to analyze the fault simulation scenario and formulate preliminary measures, such as adjusting the valve opening or changing the operating frequency of the pump. Through the expert knowledge base, the preliminary countermeasures are optimized and adjusted to obtain countermeasures to ensure their effectiveness and feasibility. The countermeasures are then converted into control instructions, and the power plant is remotely controlled through the digital twin control module, such as adjusting the output power of the generator to reduce the impact of the fault. Finally, the control effect is monitored. If the fault is still not effectively handled, the process of abnormal detection, classification, evaluation, simulation, analysis, optimization and control is repeated until the fault is successfully resolved.

[0109] It should be noted that the rule model is a model that makes decisions or judgments based on predefined rules and logical conditions. It is the core component of dynamic digital twin technology. It cleverly integrates the control algorithm with the three-dimensional model to achieve precise control of the motion state of the virtual twin and intuitive interaction in three-dimensional space. This combination not only gives life to the virtual twin, enabling it to act as a digital mirror of the physical system, dynamically simulate and display various behavioral changes based on data. Expert knowledge base A database or system built based on the knowledge and experience of domain experts, used to store, manage and apply knowledge in a professional field. This knowledge base usually includes a series of rules, facts, experience, best practices, case analysis and decision-making strategies, which can help the system make more accurate and effective judgments or decisions in complex or professional situations.

[0110] Figure 6 It is a flowchart of step S60 in the power plant load visualization control method based on digital twin provided in an embodiment of the present invention.

[0111] like Figure 6 As shown, in an optional embodiment of the present invention, based on the digital twin model, countermeasures and PID control algorithm are used to calculate the dispatching instructions to monitor and control the operating status of the power plant, including: S61, formulating a preliminary energy dispatching strategy based on the prediction results; S62, optimizing the preliminary energy dispatching strategy based on the countermeasures using the PID control algorithm to obtain the optimal energy dispatching strategy; S63, generating dispatching instructions based on the optimal energy dispatching strategy using a rule model; S64, inputting the dispatching instructions into the digital twin model to simulate the operating status and effect of the power plant after executing the dispatching instructions to obtain simulation results; S65, adjusting the operating status of the power plant using the dispatching instructions to obtain execution results; S66, performing deviation analysis on the execution results and the simulation results to obtain a deviation correction plan; S67, adjusting the preliminary energy dispatching strategy based on the deviation correction plan to update the dispatching instructions.

[0112] Specifically, according to the prediction results, a preliminary energy dispatching strategy for regulating the operating status of the power plant is formulated, and the optimization objectives under different preliminary energy dispatching strategies are calculated. It is assumed that there are three dispatching strategies: Strategy A focuses on maximizing power generation, Strategy B focuses on minimizing fuel consumption, and Strategy C balances power generation and fuel consumption. Through simulation calculations, the optimization target value of Strategy A is 600 MW of power generation, the fuel consumption of Strategy B is reduced to 900 tons per hour, and Strategy C controls fuel consumption to 950 tons under the condition of 550 MW of power generation. In response to the countermeasures, the PID control algorithm is used to compare the optimization target values ​​of each dispatching strategy, and Strategy C is determined to be the optimal energy dispatching strategy.

[0113] Based on the countermeasures, the PID control algorithm is used to calculate the optimal energy scheduling strategy: Assuming that the PID control parameters are Kp=5, Ki=1, and Kd=05, by adjusting these parameters, the scheduling strategy calculated is to increase the fuel supply by 10% during peak hours and reduce the fuel supply by 5% during off-peak hours. The generated scheduling instructions are input into the digital twin model of the power plant, and the operating status and effect of the power plant after executing the scheduling instructions are simulated. The simulation results show that the power plant's power generation increased to 480 MW during peak hours and decreased to 420 MW during off-peak hours, and the overall energy efficiency increased by 2%. The simulation results are compared with the execution results of the actual power plant after executing the scheduling instructions. If the simulation effect is not ideal, the PID control parameters can be adjusted and the preliminary scheduling strategy can be recalculated. If the simulation results meet expectations, the scheduling instructions are sent to the control system of the power plant, and the control system adjusts the equipment according to the instructions.

[0114] It should be noted that the dispatching instructions are sent to the control system of the power plant to perform energy distribution and load adjustment operations. After obtaining the execution results of the power plant after executing the dispatching instructions, data analysis techniques, such as principal component analysis (PCA), are applied to evaluate the execution results and determine whether the operating status of the power plant meets expectations. Through the three-dimensional model display system, the execution results and the virtual operation simulation results are analyzed for deviations, and the deviation correction plan is obtained using deviation analysis algorithms, such as mean square error (MSE). According to the deviation correction plan, the dispatching strategy is continuously adjusted, the dispatching instructions are updated, and a closed-loop control is formed to ensure the continuous optimization of the operating status of the power plant. The entire process is achieved through automated information technology without human intervention, ensuring the efficient and safe operation of the power plant.

[0115] Figure 7 is a schematic diagram of a power plant load visualization control system based on digital twins provided in an embodiment of the present invention;

[0116] like Figure 7 As shown, a power plant load visualization control system 100 based on digital twins includes:

[0117] The three-dimensional modeling unit 101 is used to obtain the entity data of the power plant to construct and integrate the three-dimensional model of the appearance of the power plant and the three-dimensional model of the internal equipment to obtain a static three-dimensional model;

[0118] The digital modeling unit 102 is used to construct a dynamic digital model based on the mechanism modeling method and the data-driven modeling method, and integrate the static three-dimensional model and the dynamic digital model to obtain a digital twin model;

[0119] The data prediction unit 103 is used to process the real-time operation data and historical operation data of the power plant based on the reinforcement learning algorithm and the deep learning algorithm to obtain a prediction result;

[0120] The state identification unit 104 is used to compare the prediction result with the warning threshold based on the static three-dimensional model to identify the abnormal state of the power plant and obtain a judgment result;

[0121] A measure generating unit 105 is used to evaluate the severity of the abnormal state using a rule model according to the judgment result to formulate a countermeasure;

[0122] The model control unit 106 is used to calculate the dispatching instructions based on the digital twin model using the countermeasures and the PID control algorithm to monitor and control the operating status of the power plant.

[0123] For the description of specific functions and examples of each module and submodule of the device in the embodiment of the present invention, reference can be made to the relevant description of the corresponding steps in the above method embodiment, which will not be repeated here.

[0124] In the technical solution of the present invention, the acquisition, storage and application of user personal information involved are in compliance with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0125] According to an embodiment of the present invention, the present invention also provides an electronic device, a readable storage medium and a computer program product.

[0126] Figure 8 A schematic block diagram of an example electronic device 200 that can be used to implement an embodiment of the present invention is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital assistants, cellular phones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present invention described and / or required herein.

[0127] like Figure 8 As shown, the device 200 includes a computing unit 201, which can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) 202 or a computer program loaded from a storage unit 208 into a random access memory (RAM) 203. In the RAM 203, various programs and data required for the operation of the device 200 can also be stored. The computing unit 201, the ROM 202, and the RAM 203 are connected to each other via a bus 204. An input / output (I / O) interface 205 is also connected to the bus 204.

[0128] A number of components in the device 200 are connected to the I / O interface 205, including: an input unit 206, such as a keyboard, a mouse, etc.; an output unit 207, such as various types of displays, speakers, etc.; a storage unit 208, such as a disk, an optical disk, etc.; and a communication unit 209, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 209 allows the device 200 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0129] The computing unit 201 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the computing unit 201 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, digital signal processors (DSPs), and any appropriate processors, controllers, microcontrollers, etc. The computing unit 201 performs the various methods and processes described above, such as a power generation prediction method. For example, in some embodiments, a power generation prediction method may be implemented as a computer software program, which is tangibly contained in a machine-readable medium, such as a storage unit 208. In some embodiments, part or all of the computer program may be loaded and / or installed on the device 200 via the ROM 202 and / or the communication unit 209. When the computer program is loaded into the RAM 203 and executed by the computing unit 201, one or more steps of a power generation prediction method described above may be performed. Alternatively, in other embodiments, the computing unit 201 may be configured to perform a power generation prediction method in any other appropriate manner (e.g., by means of firmware).

[0130] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0131] The program code for implementing the method of the present invention can be written in any combination of one or more programming languages. These program codes can be provided to a processor or controller of a general-purpose computer, a special-purpose computer or other programmable data processing device, so that the program code, when executed by the processor or controller, enables the functions / operations specified in the flow chart and / or block diagram to be implemented. The program code can be executed entirely on the machine, partially on the machine, partially on the machine as a stand-alone software package and partially on a remote machine, or entirely on a remote machine or server.

[0132] In the context of the present invention, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, device, or equipment. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices, or equipment, or any suitable combination of the foregoing. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0133] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user can provide input to the computer. Other types of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including voice input, or tactile input).

[0134] The systems and techniques described herein may be implemented in a computing system that includes back-end components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes front-end components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such back-end components, middleware components, or front-end components. The components of the system may be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), and the Internet.

[0135] A computer system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The relationship of client and server is generated by computer programs running on respective computers and having a client-server relationship with each other. The server may be a cloud server, a server of a distributed system, or a server combined with a blockchain.

[0136] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps described in the present invention can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solution disclosed in the present invention can be achieved, and this document does not limit this.

[0137] The above specific implementations do not constitute a limitation on the protection scope of the present invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the principles of the present invention should be included in the protection scope of the present invention.

[0138] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and substitutions can be made without departing from the technical principles of the present invention. These improvements and substitutions should also be regarded as the scope of protection of the present invention.

Claims

1. A power plant load visualization control method based on digital twin, characterized in that: Acquire the entity data of the power plant to construct and integrate the 3D model of the power plant appearance and the 3D model of the internal equipment to obtain a static 3D model; Based on the mechanism modeling method and the data-driven modeling method, a dynamic digital model is constructed, and the static three-dimensional model and the dynamic digital model are integrated to obtain a digital twin model; Based on reinforcement learning algorithms and deep learning algorithms, the real-time and historical operating data of the power plant are processed to obtain prediction results; Based on the static three-dimensional model, the prediction result is compared with the early warning threshold to identify the abnormal state of the power plant and obtain a judgment result; According to the judgment result, the severity of the abnormal state is evaluated using a rule model to formulate a response measure; Based on the digital twin model, the countermeasures and PID control algorithm are used to calculate scheduling instructions to monitor and control the operating status of the power plant.

2. The method according to claim 1, characterized in that The acquisition of the power plant entity data to construct and integrate the power plant appearance three-dimensional model and the internal equipment three-dimensional model to obtain a static three-dimensional model includes: Acquire the entity data of the power plant in real time, and construct a three-dimensional model of the appearance of the power plant and a three-dimensional model of the internal equipment; Integrating the three-dimensional model of the power plant appearance and the three-dimensional model of the internal equipment to obtain an integrated three-dimensional model; Based on the network rendering technology, the integrated three-dimensional model is rendered to obtain the static three-dimensional model.

3. The method according to claim 1, characterized in that The step of integrating the static three-dimensional model and the dynamic digital model to obtain a digital twin model includes: integrating the static three-dimensional model and the dynamic digital model to obtain the digital twin model by using parameter coupling and state variable association.

4. The method according to claim 1, characterized in that: The real-time operation data and historical operation data of the power plant are processed based on the reinforcement learning algorithm and the deep learning algorithm to obtain the prediction results including: Acquiring the real-time operation data and the historical operation data, and preprocessing the real-time operation data and the historical operation data to obtain a standardized data set; Based on the deep learning algorithm, feature extraction is performed on the standardized data set to obtain a key feature set; Constructing a prediction model according to the reinforcement learning algorithm; Based on the key feature set, the prediction model is trained to obtain a trained prediction model; Based on the trained prediction model, the real-time operation data is predicted to obtain a prediction result of the power plant operation status.

5. The method according to claim 1, characterized in that: The prediction result is compared with the early warning threshold based on the static three-dimensional model to identify the abnormal state of the power plant, and the judgment result obtained includes: Inputting the prediction results into the digital twin model, and dynamically generating real-time operating status data in the static three-dimensional model; Calculating the warning threshold using the historical operation data based on statistical analysis; Based on an abnormality judgment algorithm, the real-time operating status data is compared with the early warning threshold to identify the abnormal state of the power plant and obtain the judgment result.

6. The method according to claim 1, characterized in that According to the judgment result, the severity of the abnormal state is evaluated using a rule model to formulate response measures, including: According to the judgment result, a convolutional neural network is used to classify the abnormal state to obtain a fault type; Based on a machine learning algorithm, the severity of the abnormal state is evaluated to obtain a fault severity level; Constructing a fault simulation scenario according to the fault type and the fault severity level; Based on the rule model, the fault simulation scenario is analyzed using a preset rule library to obtain preliminary measures; Based on the expert knowledge base, the preliminary measures are optimized and adjusted to obtain countermeasures.

7. The method according to claim 1, characterized in that The method of calculating the dispatching instruction based on the digital twin model using the countermeasures and the PID control algorithm to monitor and control the operation status of the power plant includes: Formulate a preliminary energy dispatch strategy based on the forecast results; According to the countermeasures, the preliminary energy scheduling strategy is optimized using the PID control algorithm to obtain the optimal energy scheduling strategy; According to the optimal energy scheduling strategy, the scheduling instruction is generated by using the rule model; Inputting the dispatch instruction into the digital twin model to simulate the operating state and effect of the power plant after executing the dispatch instruction to obtain a simulation result; Adopting the dispatch instruction to adjust the operating state of the power plant to obtain an execution result; Perform deviation analysis on the execution result and the simulation result to obtain a deviation correction scheme; According to the deviation correction scheme, the preliminary energy scheduling strategy is adjusted to update the scheduling instruction.

8. A power plant load visualization control system based on digital twin, characterized in that: The system comprises: A three-dimensional modeling unit is used to obtain the entity data of the power plant to construct and integrate the three-dimensional model of the power plant appearance and the three-dimensional model of the internal equipment to obtain a static three-dimensional model; A digital modeling unit, used to construct a dynamic digital model based on a mechanism modeling method and a data-driven modeling method, and integrate the static three-dimensional model with the dynamic digital model to obtain a digital twin model; A data prediction unit is used to process the real-time operation data and historical operation data of the power plant based on a reinforcement learning algorithm and a deep learning algorithm to obtain a prediction result; A state identification unit, configured to compare the prediction result with a warning threshold based on the static three-dimensional model to identify an abnormal state of the power plant and obtain a judgment result; A measure generating unit, configured to evaluate the severity of the abnormal state using a rule model according to the judgment result, so as to formulate a countermeasure; A model control unit is used to calculate scheduling instructions based on the digital twin model using the countermeasures and PID control algorithm to monitor and control the operating status of the power plant.

9. An electronic device, characterized in that: include: at least one processor, and a memory communicatively connected to the at least one processor; The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the method described in any one of claims 1 to 7.

10. A non-transitory computer-readable storage medium storing computer instructions, wherein: The computer instructions are used to cause the computer to execute the method according to any one of claims 1-7.

Images