Cockpit multi-scene switching output control method based on artificial intelligence

Abstract

The invention relates to the technical field of power system management, and discloses a cockpit multi-scene switching output control method based on artificial intelligence, which comprises the following steps: S1, real-time data acquisition and fusion: acquiring power load, equipment state and environment data through a power load sensor, an equipment state sensor and an environment data sensor, fusing the collected data and constructing a multi-dimensional state space to represent a current power system state; and S2, deep reinforcement learning and optimal decision path calculation: constructing a state space by using the collected real-time data, and selecting a most suitable display mode through a reinforcement learning training model to ensure that the system can make an optimal decision in different scenes. According to the multi-scene switching output control method driven by artificial intelligence, intelligent and dynamic adjustment of the display content is realized, the system response speed, the decision efficiency and the stability are improved, and the information display precision and the operator decision support capability in the power dispatching process are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of power system management, and in particular to an artificial intelligence-based cockpit multi-scenario switching output control method. Background Art

[0002] In modern society, as the demand for electricity continues to increase, the complexity of power system management is increasing. In order to ensure the stability and reliability of power supply, the power dispatching system needs to flexibly adjust the displayed information under different operating conditions so that operators can make decisions quickly and accurately.

[0003] Most existing power dispatching systems use fixed display templates and switch display content through manual intervention. Under normal circumstances, these systems can provide basic power load and equipment status information to help operators perform daily power dispatching work. For tasks such as equipment status monitoring and load forecasting, existing technologies can analyze the power system to a certain extent through data collection and preset models to provide actionable information. The advantages of these systems are simple structure and easy operation. They can support the basic needs of power dispatching to a certain extent, especially when the load and equipment status are stable, and can provide a more intuitive display interface and operation method.

[0004] However, the existing technology has some shortcomings when facing complex and dynamic power systems; the traditional cockpit display mode is fixed and cannot automatically adjust the display content according to changes in power load, equipment status and external environment. Especially in the case of load fluctuations or equipment failures, the traditional system display content updates lags and cannot respond in time. Manual switching of display modes not only increases the burden on operators, but also easily leads to information overload or decision-making delays; furthermore, the existing technology lacks an effective coordination mechanism. In a multi-operator environment, display content is prone to conflict, resulting in reduced work efficiency; in addition, the existing system does not fully consider the impact of uncertainty factors, such as load fluctuations, equipment failures, etc. on decision-making. The system cannot maintain stable operation in such an uncertain environment, thereby affecting the efficiency and accuracy of power dispatching. Summary of the invention

[0005] In view of the shortcomings of the prior art, the present invention provides a cockpit multi-scene switching output control method based on artificial intelligence, which solves the problems in the prior art of inflexible switching of cockpit display content, slow response speed, low decision-making efficiency and inability to effectively deal with uncertain factors.

[0006] To achieve the above objectives, the present invention is implemented through the following technical solutions: A cockpit multi-scene switching output control method based on artificial intelligence, comprising the following steps: S1. Real-time data collection and fusion: Collect power load, equipment status and environmental data through power load sensors, equipment status sensors and environmental data sensors, fuse the collected data and construct a multi-dimensional state space to represent the current power system status; S2. Deep reinforcement learning and optimal decision path calculation: Use the collected real-time data to build the state space, and use reinforcement learning to train the model and select the most appropriate display mode to ensure that the system can make the best decision in different scenarios; S3. Game theory and collaborative decision-making: With the participation of multiple operators, game theory optimization is performed based on the optimal decision path to ensure that the display modes selected by different operators can be coordinated and consistent; S4. Robust control and uncertainty optimization: Based on game theory optimization, robust control methods are used to handle the uncertainties of load fluctuations and equipment failures to ensure stable system operation and reduce decision-making errors. S5. Display content adjustment and scene switching: Based on a stable system, the status of the power system is monitored in real time, and the display content is adjusted dynamically and the display mode is switched to meet the needs of current power dispatching; S6. Performance evaluation and optimization: Compare real-time data and historical data, evaluate system performance, perform self-tuning based on the evaluation results, and optimize display content and decision paths.

[0007] Preferably, the real-time data collection includes: Obtain current power load data, including instantaneous load and load forecast; Real-time monitoring of the operating status of key equipment, including temperature, load, and equipment health status; External environmental factors such as weather changes and equipment failure rates are collected through sensors and monitoring equipment.

[0008] Preferably, the fusion step comprises: Extract features from collected load data, equipment status and environmental data, and process them through multi-sensor fusion technology; The fused data is converted into a multi-dimensional state space, where the state space includes load status, equipment status and environmental data, represented as a three-dimensional data combination.

[0009] Preferably, the deep reinforcement learning model training includes: Use a deep Q network training model to map the system's state and action to Q values ​​and learn the optimal strategy; The entire system is modeled as a Markov decision process, and the action in each state is selected through the MDP framework.

[0010] Preferably, the optimal decision path calculation includes: The maximum reward value of each state is calculated through the Bellman optimality equation, and the optimal display mode is selected in each state; According to the calculation results of the optimal decision path, the most appropriate display mode under the current power load and equipment status is selected, and dynamic display content switching is performed.

[0011] Preferably, the game theory optimization includes: In multi-operator scenarios, game theory models are used to optimize the decision-making of each operator and ensure the consistency of the content displayed by each operator; By solving the Nash equilibrium and optimizing the decisions among operators, the display modes selected by each operator can be coordinated with each other and reach the optimal state.

[0012] Preferably, the robust control and uncertainty optimization include: Through the H∞ control method, a robust control strategy is designed to reduce the impact of uncertainty in the system and ensure the stability of the system under load fluctuations and equipment failures; The uncertainty factors of load fluctuation and equipment failure are modeled and optimized through stochastic control theory to ensure stable operation of the system in an uncertain environment.

[0013] Preferably, the display content adjustment includes: According to the current power load: Based on the real-time collected load data, the system dynamically determines the current load status and adjusts the priority of the cockpit display content according to the load size; Adjust the level of detail of the display based on the load and health of key equipment, including displaying equipment load and spare capacity when certain equipment is highly loaded; Adjust the display content based on environmental data to ensure that the displayed information is timely and targeted.

[0014] Preferably, the performance evaluation and tuning module includes: Evaluate system performance under different load, equipment status and environmental conditions based on real-time and historical data; By analyzing the evaluation results, the system automatically selects and adjusts the optimization strategy to improve the accuracy of the system's displayed content and the response efficiency; The system continuously optimizes the decision-making process by learning from historical data and real-time feedback, improving the performance of the model and the accuracy of the display content switching strategy.

[0015] Preferably, the multi-scenario automatic switching mechanism includes: Based on real-time power load, equipment status, and external environment data, the system automatically identifies the current power dispatch scenario and selects the appropriate display mode; Based on the scene recognition results, the system automatically switches the cockpit display mode, including displaying global equipment information in low-load scenarios and focusing on the load conditions and spare capacity of key equipment in high-load scenarios; Based on the model training results and real-time feedback, the multi-scene switching mechanism is continuously optimized so that it can adjust the display content in various complex power dispatching scenarios.

[0016] The present invention provides a cockpit multi-scene switching output control method based on artificial intelligence. It has the following beneficial effects: 1. The present invention combines artificial intelligence with game theory to achieve intelligent display content switching in multi-operator scenarios, ensuring that the display content of each operator is coordinated and consistent. By solving the Nash equilibrium, the decision path is optimized, display content conflicts are avoided, and the overall decision efficiency is improved. Compared with traditional solutions, the system greatly reduces the problem of repeated information display and coordination between operators, and optimizes operating efficiency and user experience.

[0017] 2. The present invention improves the stability under uncertain factors such as power load fluctuations and equipment failures through robust control and uncertainty optimization technology. By using H∞ control and random control, the system's response to external disturbances is minimized to ensure the continuous and stable operation of the system. This solution makes up for the lack of uncertainty processing mechanism in the prior art and enhances the robustness and reliability of the system.

[0018] 3. The present invention uses an adaptive performance evaluation and optimization mechanism, and the system can adjust the decision path and the switching strategy of the displayed content in real time. According to the comparison between real-time and historical data, the system can continuously optimize the decision, ensure the accurate matching of the displayed content and the decision path, and improve the display efficiency and response speed. This feedback optimization method avoids the inefficiency problem of relying on manual adjustment in traditional methods, and realizes the effective combination of automation and intelligence. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 The figure is a flow chart of the method of the present invention. DETAILED DESCRIPTION

[0020] The following will be combined with the drawings in the specification 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.

[0021] Please see attached Figure 1 The embodiment of the present invention provides a cockpit multi-scene switching output control method based on artificial intelligence, comprising the following steps: S1. Real-time data collection and fusion: Collect power load, equipment status and environmental data through power load sensors, equipment status sensors and environmental data sensors, fuse the collected data and construct a multi-dimensional state space to represent the current power system status; Real-time data collection and fusion capture system status and provide key input data for subsequent deep learning, decision calculation, game optimization and other modules, ensuring that the system can make appropriate decisions based on the latest system status.

[0022] In this embodiment, the core of step S1 is to collect multi-dimensional data such as power load, equipment status, external environment, etc., and transform them into state space through data fusion technology. This process is not only to improve the accuracy of data collection, but also to provide a reliable basis for subsequent decision-making, scene switching and optimization.

[0023] First, the system uses multiple sensors to obtain real-time data on power load, equipment status, and external environment. Generally, the main data collected by the system include: Power load data: This type of data mainly includes instantaneous load and load forecast information. Instantaneous load reflects the immediate demand of the power system at the current moment, while load forecast predicts the power demand trend in a certain period of time in the future based on historical data and algorithm models.

[0024] Instantaneous load :Indicates that the system is at time The load amount.

[0025] Load Forecasting :Load forecasting based on load history and external influencing factors (such as season, climate, etc.) is used to determine load changes in the future.

[0026] Equipment status data: Equipment status data includes the operating status of key equipment (such as substations, generator sets, etc.), such as temperature, load, and health status. Equipment status is very important for real-time monitoring of system stability and emergency response.

[0027] Equipment Load : Indicates the load status of the device at a certain moment, such as the device's workload, temperature, health status, etc. The device's load data can directly affect the selection of display content. Especially in high-load situations, the system needs to give priority to displaying the status information of related devices.

[0028] External environmental data: External environmental data includes factors that may affect power load and equipment performance, such as weather changes and equipment failure rates. Weather changes and equipment failure rates are important external factors that affect power system load fluctuations.

[0029] Environmental data : This type of data includes weather factors such as temperature, humidity, wind speed, as well as information such as equipment failure rate and external interference.

[0030] After real-time data collection, the system will transform this information into a multi-dimensional state space through data fusion technology to represent the current state of the power system. In this process, feature extraction and data fusion are key steps to ensure that the information extracted from various sensors can be comprehensively utilized in subsequent decision-making.

[0031] In the data fusion stage, the data from multiple sensors are integrated into a unified state representation through fusion algorithms (such as Kalman filtering, particle filtering, etc.). For example, the system can use weighted averaging or Kalman filtering technology to fuse data from different sources to eliminate noise and improve data accuracy.

[0032] Through this fusion, the system can construct a multidimensional state space , the state space represents the overall situation of the current power system. Specifically, the state space The following dimensions are included: ; in: It is the state of the power load, including both instantaneous load data and load forecast data; It is the status of the equipment, reflecting the equipment's operating load, health status, etc. It is environmental data, covering external influencing factors such as weather changes and equipment failure rates.

[0033] By integrating this data, the system can obtain a real-time and accurate status representation, thereby providing strong support for subsequent decision-making, control, and display content switching.

[0034] In order to ensure that the system can use the collected data more effectively, feature extraction is a critical step. In this embodiment, the system will extract features from load data, equipment status data, and environmental data to filter out the most important information for current decision-making.

[0035] Load data feature extraction: For example, instantaneous load data The filtering algorithm removes high-frequency noise and extracts long-term trends. This can help the system identify load change trends more accurately and avoid inaccurate judgments due to short-term fluctuations.

[0036] Equipment status feature extraction: The temperature, load and health status of the equipment are analyzed through time series to extract the change pattern of the equipment status. For example, when the equipment load is high, the system can determine whether the equipment needs emergency maintenance based on historical data and equipment health status.

[0037] Environmental data feature extraction: Weather changes and changes in equipment failure rates are also extracted as features. Especially in the case of extreme weather or equipment failure, the system can automatically adjust the display content by learning from historical data to facilitate operators to respond to emergencies.

[0038] As an option, the Kalman filter algorithm is used in this embodiment for data fusion. Kalman filtering is a common recursive data processing method that can effectively reduce sensor noise and improve data accuracy. Specifically, the Kalman filter method estimates and updates the state through the following recursive formula: ; in: is the estimated value of the state, indicating the system at time The estimated status of It is the sensor measurement value, which represents the real-time data collected by the sensor; is the Kalman gain, which determines the weight between the sensor data and the system state estimate; is the observation matrix, which describes the relationship between sensor measurements and system states; Indicates that the system is at time The prior state estimate of .

[0039] Through Kalman filtering, the system is able to adjust state estimates in real time and eliminate noise, ensuring that subsequent decisions are made based on accurate data.

[0040] In a possible implementation, particle filtering technology can also be used as an alternative. Particle filtering can handle nonlinear and non-Gaussian noise situations and is suitable for more complex system state estimation. Especially when processing multi-dimensional sensor data, particle filtering can better capture subtle changes in the data. Through data fusion algorithms (such as Kalman filtering or particle filtering), the system can eliminate noise and ensure the accuracy of information. This provides accurate input data for subsequent deep reinforcement learning models, optimal decision path calculations, game theory optimization and other steps, ensuring that the system can make timely and accurate decisions based on real-time power load and equipment status.

[0041] S2. Deep reinforcement learning and optimal decision path calculation: Use the collected real-time data to build the state space, and use reinforcement learning to train the model and select the most appropriate display mode to ensure that the system can make the best decision in different scenarios; Using deep reinforcement learning models, especially deep Q networks (DQN), the state space is processed and the optimal decision path is calculated, and the most appropriate display mode is finally selected. Through this process, the system can dynamically adjust the display content to meet the display needs of the power system under different load and equipment conditions.

[0042] In this embodiment, the system first converts the power load data , Equipment status data and environmental data It is converted into a multidimensional state space and trained by a deep Q network (DQN). The deep Q network uses Q values ​​to evaluate the rewards of each state-action pair to select the optimal display mode. The update of Q values ​​and the calculation of the optimal path are based on the Q-learning algorithm and the Bellman optimality equation, and the decision is optimized recursively.

[0043] In general, the system is based on the current state at each moment. Select an action , the action corresponds to a display mode (for example, display equipment load, spare capacity, etc.). The Q value calculation formula is: ; in: Indicates that the system is in state Next select action The expected return obtained. This Q value is used to evaluate the long-term return of the current action; represents the immediate reward, that is, the system is in state Next action The reward is usually related to the relevance of the displayed content and the effectiveness of the display mode; Discount factor: A larger discount factor (closer to 1) means that the system focuses more on future returns, while a smaller discount factor means that the system focuses more on current returns. Indicates that the system is in the next state Next, choose the best action The maximum Q value obtained.

[0044] Through the Q-learning algorithm, the system optimizes the decision path by continuously updating the Q value, and finally selects the best display mode, so that the power system can make the most appropriate display decision under different load and equipment conditions.

[0045] In order to calculate the optimal decision path in each state, the system uses the Markov decision process (MDP) model. MDP is a mathematical model used to describe the decision process. It consists of a state space , Action Space , state transition probability , reward function and discount factor Through MDP, the system can select the optimal display mode in each state, thereby achieving the optimal switching of display content.

[0046] Specifically, under the MDP framework, the system recursively calculates the optimal value function of each state through the Bellman optimality equation , and then select the optimal action based on the function to ensure that the system can make the best decision when the power load and equipment status change.

[0047] The Bellman optimality equation is as follows: ; in: is the optimal value function, indicating that the system is in state The maximum long-term reward that can be obtained under each state. By recursively calculating this value, the system can identify the optimal decision path in each state; Indicates in status In this case, the system can choose an action, i.e., a display mode. For example, in a low-load scenario, the system may choose to display the overall distribution of device resources; in a high-load scenario, it may choose to display the load of the device; is an immediate reward, indicating that the system is in state Next action The immediate reward reflects the impact of the current decision on the overall performance of the system; is the discount factor, which indicates the impact of future rewards; is the state transition probability, indicating that the system is in state Next action After transfer to state In the power dispatching system, this usually reflects the changing pattern of load and equipment status; Status The optimal value function under , represents the maximum reward that the system can obtain in the next state.

[0048] Through the Bellman optimality equation, the system can calculate the optimal display mode in each state. At each moment, according to the current state , the system selects the optimal action Specifically, when the power load is high, the system will give priority to displaying information such as the load and spare capacity of key equipment; when the load is low, the system will choose to display global information such as resource distribution and equipment status.

[0049] As an option, when the system detects load fluctuations, based on the trained Q value, the system can adjust the priority of the displayed content, give priority to displaying equipment information with higher loads, and provide backup plans or emergency plans. This switching mechanism ensures that the cockpit can respond to changes in power load and equipment status in real time.

[0050] Through the Q-learning algorithm, the system can select the optimal display mode according to the current power load, equipment status and external environment data, and continuously calculate the optimal decision path through the Bellman optimality equation. The decision path of the system is calculated recursively to ensure that each display content switch can make the most appropriate response according to the latest power system status.

[0051] S3. Game theory and collaborative decision-making: With the participation of multiple operators, game theory optimization is performed based on the optimal decision path to ensure that the display modes selected by different operators can be coordinated and consistent; When multiple operators use the cockpit together, how to ensure that the selection of displayed content does not conflict and can effectively collaborate is a key issue to further improve system performance. Therefore, step S3 introduces game theory and collaborative decision-making mechanism to optimize decision-making coordination in a multi-operator environment.

[0052] In this embodiment, the goal of step S3 is to optimize the decision of each operator through the game theory model, so that multiple operators can coordinate and agree when selecting the display mode, avoid decision conflicts, and optimize the overall system performance. This process solves the Nash equilibrium to ensure that the decision of each operator is optimal under the strategies of other operators, thereby achieving overall optimization of the system.

[0053] In a multi-operator environment, each operator has different decision-making needs and chooses a presentation mode based on the current state. However, since multiple operators may choose different presentation modes, the system needs to optimize these choices so that the presentation content is as non-repetitive and consistent as possible. At the core of this process is a game theory model, through which the system optimizes the decisions of each operator.

[0054] In general, game theory models are used to measure the pros and cons of each operator's choice of display mode through a utility function. Each operator chooses a certain action (display mode) at a certain moment, and its utility (satisfaction) will be affected by the choices of other operators. Therefore, the system needs to find a stable strategy combination through game optimization to ensure that each operator's decision is both optimal and compatible with the choices of other operators.

[0055] The goal of the game is to ensure that the operators' display choices do not conflict by solving the Nash equilibrium. In game theory, Nash equilibrium refers to a strategy combination in which all operators have made the best choice, that is, no operator can obtain greater benefits by unilaterally changing his or her strategy. By solving the Nash equilibrium, the system can find a balance point so that each operator's decision can be optimized within the decision framework of other operators.

[0056] As an option, in this embodiment, the game theory model optimizes the display mode decision of each operator by solving the Nash equilibrium. Through this model, each operator will choose the display mode according to his or her own needs, and finally the game solution makes the decisions of all operators consistent, without conflict, and the overall performance of the system is maximized.

[0057] In the present invention, the core part of the game theory model is the utility function. Used to indicate that in a certain state, the operator selects a certain display mode The utility function reflects whether the operator's demand for display content is fully met, and the operator's goal is to choose a display mode that can maximize the utility function.

[0058] Specifically, the goal of game theory is to optimize the total utility of all operators, that is, to find the optimal strategy for each operator through the following equation: ; in: Indicates that the system is in state The optimal strategy under the system is the optimal display mode selected by each operator. This strategy ensures that the decisions of all operators are consistent with the overall goal of the system and avoids duplication or conflict of display content. Indicates The action selected by each operator is called the display mode. Each operator makes his own decision based on the current state and chooses the appropriate display mode. Different display modes will bring different utilities; Indicates Operators are selecting display mode The utility function is used to measure the satisfaction of the operator's needs for the displayed content and determine the operator's decision preference; Indicates the total number of operators participating in the game. In a multi-operator environment, the system needs to consider the needs of multiple operators at the same time and optimize the decision of each operator; represents the strategy space, i.e., the set of all possible display mode choices; It represents the strategy for solving the maximum utility function, that is, the system solves the game theory model to find the strategy combination that can maximize the total utility; is the total utility function, which represents the sum of the utilities of all operators.

[0059] In a multi-operator environment, the selection of display content needs to be optimized based on the needs of multiple operators and the overall goals of the system. By solving the Nash equilibrium, the system can find a stable strategy combination to ensure that all operators make consistent and coordinated decisions, thereby avoiding redundancy or conflict in display content.

[0060] Specifically, the application of the game model in a multi-operator environment can optimize the strategy of each operator in selecting display content, so that each operator's display content can be coordinated with the display content of other operators. For example, when an operator chooses to display the equipment load, the system will optimize the choices of other operators through game theory to avoid repeatedly displaying the same information and ensure that each operator obtains the most effective information they need.

[0061] In one possible implementation, the game theory model can be solved by a genetic algorithm or a dynamic programming method of game theory. The system iteratively optimizes the strategy combination of each operator and eventually finds a display content selection that all operators can accept, thereby improving the decision-making efficiency of the system.

[0062] By solving the Nash equilibrium, the system can coordinate the display content selection among multiple operators so that the decisions of all operators can cooperate with each other. In a multi-operator scenario, the display content of operators may overlap, so the system needs to avoid duplication of display information through game optimization.

[0063] For example, when the power load is high, multiple operators may need to view the load information of key equipment. Through game theory optimization, the system can ensure that when each operator chooses to display information such as load conditions and spare capacity, other operators will not repeat the same information, but choose to display complementary information, thereby improving system efficiency.

[0064] By introducing game theory optimization, the system can ensure that the decisions of multiple operators on display content will not conflict, but will be coordinated. Each operator's display mode selection not only meets his own needs, but also takes into account the needs of other operators, and finally achieves the global optimal decision. In this way, the system can provide timely and accurate information display in complex power dispatching and equipment management scenarios, helping operators make effective decisions.

[0065] As an option, game theory optimization can also improve the system's responsiveness. When operators choose to display information, the system can quickly optimize the display content based on real-time load and equipment status, and ensure that each operator gets the information they need. This process not only improves decision-making efficiency, but also reduces information overload.

[0066] By solving the Nash equilibrium, the system can ensure that multiple operators choose the same display content, avoid conflicts, and maximize system utility. The application of game theory effectively improves the decision-making efficiency in a multi-operator environment, ensuring that each operator selects display content according to actual needs while maintaining the overall coordination of the system.

[0067] S4. Robust control and uncertainty optimization: Based on game theory optimization, robust control methods are used to handle the uncertainties of load fluctuations and equipment failures to ensure stable system operation and reduce decision-making errors. In the power system, there are often some unforeseen factors and uncertainties, such as fluctuations in power load, changes in equipment failure rates, and the impact of the external environment. These uncertainties may affect the stability of the system and the accuracy of decision-making. Therefore, step S4 is committed to enhancing the system's adaptability to these uncertainties through robust control and uncertainty optimization, thereby ensuring that the system can continue to operate stably in an uncertain environment and maintain efficient decision-making and display content switching.

[0068] In this embodiment, step S4 helps the system reduce the impact caused by external disturbances in the decision-making process of power dispatch by applying robust control methods (such as H∞ control) and uncertainty optimization methods (such as stochastic control theory). Through these optimization techniques, the system can handle uncertain factors such as power load fluctuations and equipment failures, ensuring that the power system can remain stable in the face of different disturbances and uncertain environments, and make efficient and accurate decisions in such environments.

[0069] In general, the operation of the power system is affected by a variety of uncertain factors. These factors come not only from the system itself, such as equipment aging, load fluctuations, etc., but also from external environmental factors, such as weather changes, equipment failures, etc. These uncertain factors often increase the difficulty of prediction and control of the system. Therefore, the present invention solves these uncertainty problems by introducing a robust control method to ensure that the system can still maintain system stability and make optimal decisions under the influence of these uncertain factors.

[0070] As an option, the system uses H∞ control method for robust control. The goal of H∞ control is to design a controller that optimizes the performance of the system in the face of external disturbances and internal uncertainties, and minimizes the impact of uncertainty on the system output. The H∞ control method is not only suitable for processing noise within the system, but also can cope with external disturbances and unknown uncertainties.

[0071] The goal of the H∞ control problem is to make the output response of the system as unaffected by uncertainties as possible through control design. H∞ control achieves this goal by minimizing the H∞ norm. The H∞ norm represents the maximum gain of the system to disturbances, and the H∞ control method requires that this norm is less than a given threshold. , thus ensuring that the system remains stable in the face of external disturbances. The specific mathematical expression is: ; in: Represents the transfer function of the system, that is, the relationship function between input and output. In the power system, the transfer function describes the system's ability to respond to factors such as load fluctuations and changes in equipment status; Represents the H∞ norm of the system, that is, the maximum gain of the system. It measures the system's ability to respond to external disturbances. The larger the norm, the stronger the system's response to disturbances and the worse the system's robustness. is the optimization goal, which indicates the maximum degree of uncertainty that the system can tolerate. By designing the control strategy, the system hopes to Control at a minimum value to ensure system stability and minimize the risks brought by uncertainty.

[0072] By adjusting the parameters of the controller, the H∞ control method can ensure that the system can maintain system stability with smaller disturbance effects under uncertain conditions such as load fluctuations and equipment failures, thereby ensuring the accuracy of power dispatching.

[0073] In this embodiment, in addition to robust control, stochastic control theory is also used to further optimize the decision-making process of the system. Stochastic control theory mainly solves the random and uncertain factors in the system, which include not only equipment failures but also the uncertainty of power demand.

[0074] Specifically, the system uses a random process modeling method to model external disturbances (such as load fluctuations and equipment failures), and optimizes the control strategy through a random optimization algorithm. In power systems, load fluctuations and equipment failures usually occur randomly, so through stochastic control theory, this uncertainty can be accurately handled and the optimal response measures can be designed.

[0075] The system uses random processes to model uncertainty factors, ensuring that the system can adaptively adjust its decision-making strategy under random changes such as power load fluctuations or equipment failures. The goal of the system is to optimize the control strategy so that the system can maintain a stable and efficient operating state under uncertain environments.

[0076] In the optimization process of stochastic control, the system is usually optimized by the following formula: ; in: Represents the control input of the system, that is, the adjustment of the display content or the control strategy of the system. The control input will change over time. Adjust over time to cope with changes in the system; Represents the loss function of the system, which measures the system in a given state and control input The loss function reflects the system's operating efficiency and the quality of its decisions. The goal of the system is to minimize the loss function. Represents the square of the control input. In the optimal control problem, the control input is usually restricted. The square penalty term of the control input is used to control the control input of the system not to be too large, so as to ensure the stability and efficiency of the system. is the weight coefficient, which is used to balance the relationship between the size of the control input and the system loss; Represents the expected value, which is used to calculate the expectation of the loss function and usually represents the average performance of the system in the face of random perturbations; Indicates a time window, usually the time period from the start to the end of system operation; Indicates that the system is at time The state variables describe the load and power equipment status of the power system at a certain moment.

[0077] By minimizing the loss function, the system is able to make optimal decisions in the face of external disturbances and uncertainties. This optimization process ensures that the system can continue to operate stably under uncertainties such as power load fluctuations and equipment failures.

[0078] As an option, through robust control and stochastic control optimization, the system can remain stable in the face of multiple uncertainties and make each decision as efficient as possible. Through H∞ control and stochastic control theory, the system can predict and adaptively adjust factors such as load fluctuations and equipment failures, thereby improving the reliability and stability of the power dispatching system.

[0079] Through such optimization, the system can reduce errors and system performance degradation caused by external disturbances. For example, when the system load fluctuates sharply, the robust control strategy can be used to adjust the display content in time to ensure the real-time and accuracy of the display content.

[0080] Through H∞ control and stochastic control theory, the system can remain stable under uncertain factors such as load fluctuations and equipment failures, and reduce the impact of uncertainty on decision-making by optimizing control strategies.

[0081] S5. Display content adjustment and scene switching: Based on a stable system, the status of the power system is monitored in real time, and the display content is adjusted dynamically and the display mode is switched to meet the needs of current power dispatching; The dynamic characteristics of the power system mean that the priority of display content, display mode and information switching must be accurately adjusted according to the real-time power dispatching needs. The goal of step S5 is to dynamically adjust the display content and switch the display mode according to multiple factors such as the current load status of the power system, the health of the equipment and the external environment data.

[0082] In this embodiment, step S5 mainly monitors the system status in real time, combines the optimization algorithm, and adjusts the priority and display mode of the display content, thereby ensuring that the operator can receive the most relevant information at the critical moment and make accurate decisions. The system will flexibly switch the display content according to different scenarios (such as low load, high load, fault warning, etc.) to ensure the operator's response efficiency to power dispatch.

[0083] Generally speaking, in the power dispatching system, the adjustment of display content is closely related to the switching of scenes. The display content of the power system will be dynamically adjusted with the changes in power load, the different health conditions of equipment, and the influence of the external environment. For example, when the power load is low, the operator may pay more attention to the overall operating status of the equipment, while when the load is high, it is necessary to focus on the load, spare capacity and fault warning information of key equipment.

[0084] As an option, in this embodiment, the system classifies and identifies the current scene by integrating real-time power load data, equipment status, and environmental data. Based on the identified scene type, the system automatically selects the appropriate display mode to provide the operator with the most relevant information in any situation.

[0085] Specifically, scene recognition is performed through real-time load data, equipment health status, and external environment data. When power demand is low, the system automatically switches to display the status of all devices, including resource distribution, equipment load, and other information; when power demand is high, the system automatically highlights the load, spare capacity, and other information of key equipment, and provides decision support through emergency plans or fault warnings.

[0086] The adjustment of the display content needs to determine the priority of the display mode according to the real-time power system status. In order to ensure that the display content can be adjusted dynamically, the system calculates the priority of the display content based on multiple dimensions such as load, equipment status and environmental factors. The priority calculation model usually combines multiple factors and calculates the final priority through weighted sum. The calculation formula is as follows: ; in: Indicates the priority of the displayed content. The higher the priority, the more important the displayed content is in the current scenario and should be displayed first. , , They are the weight coefficients of power load, equipment status, and external environment data. The weight coefficients of different factors will affect their importance in the priority calculation; Represents the power load, that is, the system at time electricity demand; Indicates the device status, including the device's load, health status, temperature and other information; Represents external environmental data, such as weather, equipment failure rate and other external factors.

[0087] As an option, the display mode will automatically switch when the system recognizes load fluctuations or changes in equipment status. For example, when the load is low, the system will display the overall load status of the equipment, resource distribution and other information; when the load is high, the system will switch to display the load, spare capacity and fault warning information of key equipment. In these cases, the system will prioritize the display of relevant data and ensure that operators can receive key decision-making information.

[0088] Specifically, when the system detects a sharp change in power load or an abnormality in the equipment operation status, the display mode will switch according to the following rules: When power demand is low, the system displays the operating status and resource distribution of all equipment to help operators manage resources.

[0089] When power loads are high, the system will highlight the status of heavily loaded equipment, spare capacity, and potential failure warnings to help operators respond quickly to emergencies.

[0090] The priority adjustment of display content switching is based on scene recognition and real-time data analysis. The system will dynamically adjust the priority of display content according to the current power load, equipment status and environmental data. For some emergency scenarios, the system may automatically increase the priority of display content. For example, during peak load periods, the system will upgrade the load and spare capacity of key equipment to high-priority content for display so that operators can respond immediately.

[0091] The display mode switching in this embodiment depends not only on the power load and equipment status, but also on external environmental factors. By comprehensively analyzing these factors, the system can accurately identify the current power dispatching scenario and automatically adjust the display content according to the needs of different scenarios. Through this intelligent display content switching, the system can ensure that the operator can always obtain the most relevant information in various complex situations, thereby improving the decision-making efficiency and accuracy of power dispatching.

[0092] As an option, the system can provide more targeted decision support based on the priority of the displayed content. For example, when a device fails, the system will prioritize the display of the device's status, cause of failure, spare capacity, and emergency plan to help operators take quick action. This adaptive display content switching mechanism ensures that operators can make the most timely response in any environment.

[0093] The system calculates the priority of display content based on power load, equipment status and external environmental factors, and automatically determines the switching of display modes. Through this intelligent display content adjustment and scene switching, the system can ensure that operators always obtain the most relevant information, enhance the response efficiency and accuracy of decision-making, and thus optimize the performance of the power dispatching system.

[0094] S6. Performance evaluation and optimization: Compare real-time data and historical data, evaluate system performance, perform self-tuning based on the evaluation results, and optimize display content and decision paths; The main task of step S6 is to compare and evaluate the real-time data with the historical data, analyze the current performance of the system, and then self-optimize and adjust the display content and decision path, so as to improve the system's response speed, decision-making accuracy and relevance of the display content.

[0095] In general, system performance evaluation usually focuses on the following aspects: The accuracy of display content switching: that is, whether the system can accurately adjust the display content according to the real-time changes in load and equipment status.

[0096] Decision-making efficiency: Whether the system can quickly select and switch the most appropriate display mode to avoid redundant information and unnecessary switching.

[0097] System response speed: whether the system can respond promptly and adjust the display content in situations such as sudden changes in load and equipment failure.

[0098] As an option, the performance evaluation and optimization process in this embodiment analyzes the performance of the system in each scenario by comparing real-time data with historical data. For example, when the load is high, whether the display mode can give priority to displaying devices with large loads, and whether it can respond to power dispatch needs in a timely manner; when the equipment fails, whether the status information and emergency plans of the faulty equipment can be quickly called up, etc. Through these evaluations, the system will optimize the decision path, display mode and response strategy according to the evaluation results, thereby improving the overall performance of the system.

[0099] In this embodiment, the performance evaluation of the system is to calculate the comprehensive performance of the system by comprehensively considering three aspects: accuracy, response time and decision efficiency. The performance evaluation formula is as follows: ; in: Indicates the overall performance score of the system; Indicates the weight coefficient of accuracy in comprehensive performance; Indicates the accuracy of displayed content. Specifically, it measures whether the system can display the correct information at the right time; Indicates the weight coefficient of response time in comprehensive performance. Response time refers to the system's reaction speed to sudden events such as power load changes and equipment failures; Indicates the response time of display mode switching. It measures the time interval from when the system detects load change or equipment failure to when the display content is switched; Represents the weight coefficient of decision-making efficiency in comprehensive performance; Indicates decision efficiency, that is, whether the decision made by the system when switching display content is efficient.

[0100] Specifically, by evaluating the performance of the system, the system can adjust the decision path and display content switching strategy based on the evaluation results. The goal of optimizing the decision path is to reduce unnecessary display mode switching and ensure that each display content switch can help operators make effective decisions to the greatest extent. For example, during peak load, the system may over-display the status of some equipment, resulting in operators being unable to quickly obtain key information. In this case, the system can identify redundant displays through performance evaluation and reduce the switching frequency and number of displays through optimization algorithms, thereby improving the efficiency of display content switching.

[0101] As an option, the system can also self-regulate based on performance evaluation results. For example, the system may find that certain display content is over-displayed during low-load periods, resulting in reduced operator decision-making efficiency. At this time, the system can automatically adjust the display priority to reduce unnecessary information display when the load is low, thereby reducing information overload and improving decision-making efficiency.

[0102] In one possible implementation, the system uses a deep reinforcement learning (DQN) model to learn historical evaluation data and continuously optimize display mode selection and decision paths. With continuous feedback and self-adjustment, the system can better adapt to the dynamic changes in power load and equipment status, and continuously improve the accuracy of decision-making and the efficiency of display content switching.

[0103] Through performance evaluation, the system can not only optimize the switching of display content, but also adjust the priority of display content according to changes in power load and abnormal equipment status. For example, when the load is high, the system can automatically adjust the priority of display content according to the load situation to ensure that the display mode can effectively support the operator's decision-making. When the equipment fails or is abnormal, the system will quickly adjust the display mode to give priority to displaying the equipment's load, health status, and fault warnings to help operators take timely measures.

[0104] By introducing indicators such as accuracy, response time, and decision-making efficiency, the system can optimize the display mode switching strategy under different power loads, equipment status, and environmental changes, and continuously improve the system's performance. The performance evaluation and optimization mechanism not only enhances the system's adaptability, but also helps the system achieve efficient decision-making and display content switching in complex environments.

[0105] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A cockpit multi-scene switching output control method based on artificial intelligence, characterized in that: The following steps are involved: S1. Real-time data collection and fusion: Collect power load, equipment status and environmental data through power load sensors, equipment status sensors and environmental data sensors, fuse the collected data and construct a multi-dimensional state space to represent the current power system status; S2. Deep reinforcement learning and optimal decision path calculation: Use the collected real-time data to build the state space, and use reinforcement learning to train the model and select the most appropriate display mode to ensure that the system can make the best decision in different scenarios; S3. Game theory and collaborative decision-making: With the participation of multiple operators, game theory optimization is performed based on the optimal decision path to ensure that the display modes selected by different operators can be coordinated and consistent; S4. Robust control and uncertainty optimization: Based on game theory optimization, robust control methods are used to handle the uncertainties of load fluctuations and equipment failures to ensure stable system operation and reduce decision-making errors. S5. Display content adjustment and scene switching: Based on a stable system, the status of the power system is monitored in real time, and the display content is adjusted dynamically and the display mode is switched to meet the needs of current power dispatching; S6. Performance evaluation and optimization: Compare real-time data and historical data, evaluate system performance, perform self-tuning based on the evaluation results, and optimize display content and decision paths.

2. According to the artificial intelligence-based cockpit multi-scene switching output control method of claim 1, it is characterized in that: The real-time data collection includes: Obtain current power load data, including instantaneous load and load forecast; Real-time monitoring of the operating status of key equipment, including temperature, load, and equipment health status; External environmental factors such as weather changes and equipment failure rates are collected through sensors and monitoring equipment.

3. The method for controlling cockpit multi-scene switching output based on artificial intelligence according to claim 1 is characterized in that: The fusion step comprises: Extract features from collected load data, equipment status and environmental data, and process them through multi-sensor fusion technology; The fused data is converted into a multi-dimensional state space, where the state space includes load status, equipment status and environmental data, represented as a three-dimensional data combination.

4. The method for controlling cockpit multi-scene switching output based on artificial intelligence according to claim 1 is characterized in that: The deep reinforcement learning model training includes: Use a deep Q network training model to map the system's state and action to Q values ​​and learn the optimal strategy; The entire system is modeled as a Markov decision process, and the action in each state is selected through the MDP framework.

5. The method for controlling cockpit multi-scene switching output based on artificial intelligence according to claim 1 is characterized in that: The optimal decision path calculation includes: The maximum reward value of each state is calculated through the Bellman optimality equation, and the optimal display mode is selected in each state; According to the calculation results of the optimal decision path, the most appropriate display mode under the current power load and equipment status is selected, and dynamic display content switching is performed.

6. The method for controlling cockpit multi-scene switching output based on artificial intelligence according to claim 1 is characterized in that: The game theory optimization includes: In multi-operator scenarios, game theory models are used to optimize the decision-making of each operator and ensure the consistency of the content displayed by each operator; By solving the Nash equilibrium and optimizing the decisions among operators, the display modes selected by each operator can be coordinated with each other and reach the optimal state.

7. The method for controlling cockpit multi-scene switching output based on artificial intelligence according to claim 1 is characterized in that: The robust control and uncertainty optimization include: Through the H∞ control method, a robust control strategy is designed to reduce the impact of uncertainty in the system and ensure the stability of the system under load fluctuations and equipment failures; The uncertainty factors of load fluctuation and equipment failure are modeled and optimized through stochastic control theory to ensure stable operation of the system in an uncertain environment.

8. The method for controlling cockpit multi-scene switching output based on artificial intelligence according to claim 1 is characterized in that: The display content adjustment includes: According to the current power load: Based on the real-time collected load data, the system dynamically determines the current load status and adjusts the priority of the cockpit display content according to the load size; Adjust the level of detail of the display based on the load and health of key equipment, including displaying equipment load and spare capacity when certain equipment is highly loaded; Adjust the display content based on environmental data to ensure that the displayed information is timely and targeted.

9. The method for controlling cockpit multi-scene switching output based on artificial intelligence according to claim 1 is characterized in that: The performance evaluation and tuning module includes: Evaluate system performance under different load, equipment status and environmental conditions based on real-time and historical data; By analyzing the evaluation results, the system automatically selects and adjusts the optimization strategy to improve the accuracy of the system's display content and response efficiency; The system continuously optimizes the decision-making process by learning from historical data and real-time feedback, improving the performance of the model and the accuracy of the display content switching strategy.

10. The method for controlling cockpit multi-scene switching output based on artificial intelligence according to claim 1, characterized in that: The multi-scenario automatic switching mechanism includes: Based on real-time power load, equipment status, and external environment data, the system automatically identifies the current power dispatch scenario and selects the appropriate display mode; Based on the scene recognition results, the system automatically switches the cockpit display mode, including displaying global equipment information in low-load scenarios and focusing on the load conditions and spare capacity of key equipment in high-load scenarios; Based on the model training results and real-time feedback, the multi-scene switching mechanism is continuously optimized so that it can adjust the display content in various complex power dispatching scenarios.

Images