Electric power system risk intelligent assessment device based on big data and machine learning

Abstract

The invention discloses an electric power system risk intelligent assessment device based on big data and machine learning, which relates to the technical field of electric power system risk assessment and comprises a proportion analysis module, a data acquisition module, a climate analysis module, a risk calculation module, a result output module and a power supply adjustment module. According to the electric power system risk intelligent assessment device based on big data and machine learning, weather data are collected in real time through the Internet of Things sensor and the meteorological monitoring equipment, extreme weather and climate events are identified through analysis, and the risk of the electric power system is evaluated by combining the type and proportion of a power generation mode of the current electric power system. The method can comprehensively evaluate the operation and fault risks of the novel power system powered by the combination of multiple power generation modes under the current weather condition in real time, especially the risks under extreme weather and climate events, can generate a corresponding scheduling instruction based on the risk evaluation result, dynamically adjusts the proportion of each power generation mode in the power generation system, and improves the power generation efficiency. And system risks are reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of power system risk assessment, and in particular to a power system risk intelligent assessment device based on big data and machine learning. Background Art

[0002] The new power system is a concept proposed on the basis of the traditional power system, combining the current energy development trend and the application of new technologies. Compared with the traditional power system, the new power system pays more attention to the use of renewable energy and the intelligence and flexibility of the power system.

[0003] As the foundation of economic development and the support of life security, the stability and security of the power system are of vital importance. The new power system has introduced more renewable energy and distributed energy generation methods, making the structure of the power system more complex and the operation mode more diversified. Therefore, it is very necessary to conduct risk assessment on the new power system. Through risk assessment, potential risk points can be discovered and prevented in advance to ensure the stable operation of the power system.

[0004] At present, there are three common methods for power system risk assessment: failure mode and effects analysis (FMEA), fault tree analysis (FTA) and risk assessment model. Among them, FMEA lists the possible failure modes in the system and analyzes their impact on the system, thereby determining the key risk points of the system; FTA takes the undesirable events in the system as the top events, and uses logic gate symbols to refine the various causes that may cause the top events step by step in a tree-like manner until the basic cause, that is, the bottom event of the fault tree, is found. However, both methods rely on the experience and judgment of experts, and are prone to subjectivity and uncertainty, resulting in inaccurate assessment results. Therefore, the risk assessment model has gradually become the mainstream assessment method. It can more objectively reflect the risk status of the system by establishing a mathematical model to quantitatively analyze the power system and assess the risk level of the system.

[0005] However, in actual use, since new power systems usually adopt more environmentally friendly power generation methods, such as photovoltaic power generation, hydropower generation, wind power generation, etc., compared with traditional thermal power generation, these power generation methods are easily affected by weather and climate, especially extreme weather and climate events. The impact on new power systems is extremely significant. For example, continuous high temperature weather will cause the output power of photovoltaic modules to decrease, thereby affecting the amount of solar power generation; while low temperature and rainy weather will affect the power generation capacity of hydropower stations and the output of wind farms; in addition, extreme weather may also cause problems such as power grid equipment failure and line tripping, further exacerbating the operation risk of the power system. However, the existing power system risk assessment model often relies on specific algorithms and technologies, and has certain limitations. It is difficult to measure the impact of weather and climate, an uncontrollable factor, on the power system, and thus it is impossible to accurately assess the risk of the new power system. As a result, due to the incorrect assessment of system risks, there may be a shortage of power supply, or even power supply interruption, and it may also cause damage to physical equipment such as power generation equipment.

[0006] Therefore, there is an urgent need to improve this shortcoming. The present invention studies and improves the existing technology and its shortcomings, and provides an intelligent risk assessment device for a power system based on big data and machine learning. Summary of the invention

[0007] The purpose of the present invention is to provide an intelligent risk assessment device for a power system based on big data and machine learning to solve the problems raised in the above-mentioned background technology.

[0008] To achieve the above object, the present invention provides the following technical solutions:

[0009] An intelligent risk assessment device for a power system based on big data and machine learning, comprising:

[0010] A proportion analysis module, which is used to count the types of power generation modes and their proportions in the current power system, and assign corresponding weights to each power generation mode according to the proportion;

[0011] A data acquisition module, which is used to use IoT sensors and meteorological monitoring equipment to obtain weather data in real time, including but not limited to key meteorological parameters such as temperature, humidity, wind speed, wind direction, and precipitation, and transmit it to the model for processing;

[0012] A climate analysis module, which is used to analyze real-time weather information, identify and determine whether there are extreme weather and climate events that have a significant impact on various power generation methods;

[0013] A risk calculation module, which is used to consider the risk value of each power generation mode under extreme weather and climate events, and combine the weights of each power generation mode to calculate the total risk value of the power system;

[0014] A power supply adjustment module, which is used to optimize and adjust the power supply mode of the power generation system, specifically, when the total system risk value or the risk value of an individual power generation mode exceeds a threshold, dynamically adjust the proportion of each power generation mode in the power generation system to reduce system risk;

[0015] The result output module is used to generate a power system risk assessment result based on the calculated total risk value of the power system and the risk values ​​of all power generation modes, and output the assessment result to the power system operator or decision maker in the form of charts, reports, etc.

[0016] Furthermore, the power generation methods of the power system include photovoltaic power generation, hydropower generation, wind power generation, geothermal power generation, biomass power generation, nuclear power generation, and solar thermal power generation, and any power system includes at least two or more power generation methods.

[0017] Furthermore, the proportion analysis module specifically includes the following submodules:

[0018] Data collection submodule: using big data analysis technology, collects data such as power generation and installed capacity of various power generation methods from multiple channels such as power system database and power generation enterprise information system;

[0019] Data processing submodule: cleans, verifies and integrates the collected data to eliminate outliers, missing values ​​and duplicate values ​​in the data, including pre-processing operations such as normalization and standardization of the data for subsequent analysis and calculation;

[0020] Proportion calculation submodule: Calculate the proportion of each power generation mode in the power system according to its power generation or installed capacity, and the proportion calculation is in percentage form;

[0021] Weight assignment submodule: assign corresponding weights according to the proportion of each power generation mode, combined with the actual situation of the power system and risk assessment requirements.

[0022] Furthermore, the data acquisition module receives data from IoT sensors and meteorological monitoring equipment in real time, verifies the received data, and then preprocesses the verified data, such as data cleaning, format conversion and normalization. The data acquisition module transmits the preprocessed weather data to the risk assessment model in real time through a defined data interface (such as API, database connection, etc.), and ensures data synchronization between the data acquisition module and the risk assessment model, so that the model can use the latest weather data for risk assessment in real time;

[0023] The IoT sensors are deployed near power facilities and include meteorological sensors such as temperature sensors, humidity sensors, wind speed and direction sensors, and rain gauges. Each meteorological sensor captures and converts meteorological parameters into electrical signals in real time, and transmits the data collected by the sensors to the data acquisition module wirelessly through the IoT communication protocol (such as LoRa, NB-IoT, Zigbee, etc.).

[0024] The meteorological monitoring equipment is installed in meteorological stations within the coverage area of ​​the power system to provide more comprehensive and accurate meteorological data. The data from the meteorological stations is transmitted to the data acquisition module via wired or wireless means and integrated and processed together with other sensor data.

[0025] Furthermore, the climate analysis module specifically includes the following submodules:

[0026] Feature extraction submodule: Extract key features related to extreme weather and climate events from weather data (these data include temperature, humidity, wind speed, wind direction, precipitation, and other dimensions), such as the highest temperature of high temperature events, the lowest temperature of low temperature events, and the precipitation of precipitation events. Through feature extraction, data can be simplified and key information can be highlighted, which is helpful for subsequent model training and prediction;

[0027] Model training submodule: Use historical weather data and known extreme weather and climate event data to train machine learning models (such as decision trees, random forests, support vector machines, neural networks, etc.) that can learn and identify patterns of extreme weather and climate events;

[0028] Real-time prediction submodule: Real-time weather data is input into the trained machine learning model to perform real-time prediction and identification of extreme weather and climate events. The model determines whether the current weather is an extreme weather and climate event based on the characteristics of the input data and the learned patterns, and gives corresponding prediction results, that is, extreme weather and climate events that may occur in the future.

[0029] Furthermore, the climate analysis module processes and analyzes the collected weather data based on a machine learning algorithm to identify extreme weather and climate events, including extreme high temperatures, extreme low temperatures, extreme precipitation, extreme drought, hail, strong winds, tornadoes, thunderstorms and tropical cyclones. Based on the two major risk indicators of power generation loss and equipment damage rate (grid stability), the impact of extreme weather events on each power generation method is as follows:

[0030] 1) Photovoltaic power generation: Relying on sunlight, extreme high temperatures reduce the output power of modules, and hail and strong winds damage photovoltaic modules;

[0031] 2) Hydropower generation: It depends on water flow. Extreme drought leads to reduced water volume, affecting power generation. Extreme precipitation leads to floods, posing a threat to hydropower station facilities.

[0032] 3) Wind power generation: Relying on wind power, extreme low temperatures can cause wind turbine performance to decline, and strong winds, tornadoes, and thunderstorms can cause equipment damage;

[0033] 4) Geothermal power generation: relatively stable, but extreme geological activities (such as earthquakes) can cause damage to geothermal wells and power generation facilities;

[0034] 5) Biomass power generation: Relying on biomass raw materials, extreme weather may affect the supply and transportation of raw materials;

[0035] 6) Nuclear power generation: relatively stable, extreme weather (such as floods and hurricanes) poses a threat to nuclear power plant facilities;

[0036] 7) Solar thermal power generation: Relies on sunlight and high temperatures. Extremely high temperatures increase power generation efficiency, but extremely low temperatures cause system performance to decline.

[0037] The analysis results of the climate analysis module are output to the risk calculation module as an important basis for calculating the total risk value of the power system. At the same time, they are fed back to the power system operators or decision makers so that they can take timely measures to deal with the risks of potential extreme weather and climate events and conduct manual intervention.

[0038] Furthermore, the risk calculation module includes the following risk calculation formula:

[0039]

[0040] In the formula, S_risk represents the total risk value of the power system, n represents the number of types of power generation in the current system, i∈N+ and 1≤i≤n, R i represents the risk value of the i-th power generation mode, k i Represents the weight of the i-th power generation method.

[0041] Furthermore, the calculation formulas for the risk values ​​of the different power generation modes are as follows:

[0042] 1) Photovoltaic power generation is mainly affected by light intensity and temperature under extreme weather conditions. The risk value calculation formula is based on power generation loss. Photovoltaic power generation risk value = P (extreme weather conditions) × (expected power generation - actual power generation) × economic loss coefficient, where P (extreme weather conditions) is the probability of extreme weather conditions occurring, and the economic loss coefficient is the coefficient that converts power generation loss into economic value;

[0043] 2) Hydropower generation is affected by water volume, water level, water flow speed and possible flood disasters under extreme weather conditions. The risk value calculation formula takes these factors into consideration: Hydropower generation risk value = P (extreme precipitation / drought) × (expected power generation - actual power generation) × economic loss coefficient + P (flood) × flood disaster loss, where P (extreme precipitation / drought) and P (flood) are the probabilities of extreme precipitation, drought and flood disasters, respectively.

[0044] 3) Wind power generation is mainly affected by wind speed, wind direction and possible extreme wind disasters (such as tornadoes and thunderstorms) in extreme weather. The risk value calculation formula is based on equipment damage and power generation loss. Wind power generation risk value = P (extreme wind speed) × (equipment damage cost + power generation loss × economic loss coefficient), where P (extreme wind speed) is the probability of extreme wind speed occurring;

[0045] 4) Geothermal power generation is affected by geological activities (such as earthquakes, volcanic eruptions) and possible equipment failures in extreme weather conditions. The risk value calculation formula is constructed based on these factors: Geothermal power generation risk value = P (geological activity) × the impact of geological activities on power generation facilities × economic loss coefficient + P (equipment failure) × equipment failure loss;

[0046] 5) Biomass power generation is affected by raw material supply, equipment failure and possible fire in extreme weather. The risk value calculation formula is constructed based on these factors: Biomass power generation risk value = P (raw material supply shortage) × power generation loss caused by raw material shortage × economic loss coefficient + P (equipment failure) × equipment failure loss + P (fire) × fire loss, where P (geological activity) is the probability of geological activity occurring, and P (equipment failure) is the probability of equipment failure occurring;

[0047] 6) Nuclear power generation in extreme weather conditions mainly focuses on the possibility of nuclear accidents and their consequences. Due to the particularity of nuclear power generation, the risk value calculation requires highly professional and detailed evaluation. The risk value of nuclear power generation is calculated through the probabilistic safety assessment (PSA) model, which takes into account various failure modes and possible consequences of nuclear facilities. Among them, P (raw material shortage) is the probability of raw material shortage, P (equipment failure) is the probability of equipment failure, and P (fire) is the probability of fire.

[0048] 7) Solar thermal power generation is mainly affected by solar radiation intensity and temperature under extreme weather conditions. The risk value calculation formula is constructed based on the power generation loss, which is similar to the risk value calculation formula for photovoltaic power generation. The risk value of solar thermal power generation = P (solar radiation < minimum requirement) × power generation loss caused by insufficient solar radiation, where P (solar radiation < minimum requirement) is the probability of solar radiation < minimum requirement.

[0049] Furthermore, the power supply adjustment module specifically includes the following submodules:

[0050] Risk monitoring submodule: real-time monitoring of the total risk value of the power generation system and the risk value of each power generation mode. When the risk value exceeds the preset threshold, the risk warning and adjustment mechanism are triggered;

[0051] Proportion adjustment submodule: dynamically adjust the proportion of each power generation mode in the power generation system according to the risk situation of the power system;

[0052] Instruction execution submodule: converts the results of the power generation mode ratio adjustment into specific dispatch instructions and executes them. The dispatch instructions include the start and stop of power generation facilities, output adjustment, etc. When executing the dispatch instructions, ensure the stable operation of the power generation system and take into account the constraints of the power grid;

[0053] Feedback optimization submodule: collects actual effect data after the execution of dispatch instructions, such as power generation, system stability, risk value changes, etc., evaluates the effectiveness of the adjustment strategy through comparative analysis, and provides feedback and optimization based on the evaluation results. The feedback and optimization process involves adjusting risk thresholds, optimizing adjustment algorithms or strategies, etc.;

[0054] Data communication submodule: used to communicate with other systems, such as power grid dispatching system, power generation facility control system, etc., to achieve data sharing and collaborative work.

[0055] Furthermore, the proportion adjustment submodule determines the best combination of power generation methods through optimization algorithms such as linear programming, nonlinear programming, dynamic programming, heuristic algorithms, etc. to balance the risks and benefits of the system, and the adjustment process includes increasing or decreasing the output of certain power generation methods, or starting / stopping certain power generation facilities.

[0056] The present invention provides an intelligent risk assessment device for power systems based on big data and machine learning, which has the following beneficial effects:

[0057] The present invention collects weather data in real time through Internet of Things sensors and meteorological monitoring equipment, identifies extreme weather and climate events through analysis, and combines the types of power generation methods and their proportions in the current power system to comprehensively and in real time evaluate the operation and failure risks of a new power system powered by a combination of multiple power generation methods under current weather conditions, especially the risks under extreme weather and climate events. Based on the risk assessment results, the present invention can generate corresponding dispatch instructions, dynamically adjust the proportion of each power generation method in the power generation system, reduce system risks, avoid shortages and interruptions in power supply, and reduce the possibility of damage to physical equipment such as power generation equipment, thereby improving the stability and safety of the new power system under extreme weather and climate events. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] Figure 1This is a general diagram of the logical architecture of the power system risk intelligent assessment device based on big data and machine learning of the present invention;

[0059] Figure 2 It is a schematic diagram of the operation flow of the proportion analysis module of the power system risk intelligent assessment device based on big data and machine learning of the present invention;

[0060] Figure 3 A risk value calculation logic diagram of the power system risk intelligent assessment device based on big data and machine learning of the present invention;

[0061] Figure 4 This is a schematic diagram of the operation flow of the power supply adjustment module of the power system risk intelligent assessment device based on big data and machine learning of the present invention. DETAILED DESCRIPTION

[0062] The following embodiments of the present invention are described in further detail in conjunction with the accompanying drawings and examples. The following examples are used to illustrate the present invention, but are not intended to limit the scope of the present invention.

[0063] like Figure 1-Figure 4 As shown, an intelligent risk assessment device for a power system based on big data and machine learning includes: a proportion analysis module, a data acquisition module, a climate analysis module, a risk calculation module, a result output module and a power supply adjustment module;

[0064] Proportion analysis module: Statistics the types of power generation methods and their proportions in the current power system, and assign corresponding weights to each power generation method according to the proportion; the power generation methods of the power system include photovoltaic power generation, hydropower generation, wind power generation, geothermal power generation, biomass power generation, nuclear power generation, and solar thermal power generation, and any power system contains at least two or more power generation methods; this module specifically includes the following sub-modules:

[0065] Data collection submodule: using big data analysis technology, collects data such as power generation and installed capacity of various power generation methods from multiple channels such as power system database and power generation enterprise information system;

[0066] Data processing submodule: cleans, verifies and integrates the collected data to eliminate outliers, missing values ​​and duplicate values ​​in the data, including pre-processing operations such as normalization and standardization of the data for subsequent analysis and calculation;

[0067] Proportion calculation submodule: Calculate the proportion of each power generation mode in the power system according to its power generation or installed capacity, and the proportion calculation is in percentage form;

[0068] Weight assignment submodule: assign corresponding weights according to the proportion of each power generation mode, combined with the actual situation of the power system and risk assessment requirements;

[0069] Data acquisition module: Use IoT sensors and meteorological monitoring equipment to obtain weather data in real time, including but not limited to key meteorological parameters such as temperature, humidity, wind speed, wind direction, and precipitation, and transmit them to the model for processing; IoT sensors are deployed near power facilities, including temperature sensors, humidity sensors, wind speed and direction sensors, rain gauges and other meteorological sensors, and each meteorological sensor captures and converts meteorological parameters into electrical signals in real time, and transmits the data collected by the sensor to the data acquisition module wirelessly through IoT communication protocols (such as LoRa, NB-IoT, Zigbee, etc.); and meteorological monitoring equipment is equipped in meteorological stations within the coverage area of ​​the power system to provide more comprehensive and accurate meteorological data, and the data from the meteorological stations is transmitted to the data acquisition module by wire or wireless means, and integrated and processed together with other sensor data;

[0070] In this embodiment, the data acquisition module receives data from IoT sensors and meteorological monitoring equipment in real time, verifies the received data, and then pre-processes the verified data, such as data cleaning, format conversion and normalization, etc. Then, the data acquisition module transmits the pre-processed weather data to the risk assessment model in real time through a defined data interface (such as API, database connection, etc.), and ensures data synchronization between the data acquisition module and the risk assessment model, so that the model can use the latest weather data for risk assessment in real time;

[0071] Climate analysis module: Analyzes real-time weather information, identifies and determines whether there are extreme weather and climate events that have a significant impact on various power generation methods; this module specifically includes the following sub-modules:

[0072] Feature extraction submodule: Extract key features related to extreme weather and climate events from weather data (these data include temperature, humidity, wind speed, wind direction, precipitation, and other dimensions), such as the highest temperature of high temperature events, the lowest temperature of low temperature events, and the precipitation of precipitation events. Through feature extraction, data can be simplified and key information can be highlighted, which is helpful for subsequent model training and prediction;

[0073] Model training submodule: Use historical weather data and known extreme weather and climate event data to train machine learning models (such as decision trees, random forests, support vector machines, neural networks, etc.) that can learn and identify patterns of extreme weather and climate events;

[0074] Real-time prediction submodule: inputs real-time weather data into the trained machine learning model to perform real-time prediction and identification of extreme weather and climate events. The model determines whether the current weather is an extreme weather and climate event based on the characteristics of the input data and the learned patterns, and gives the corresponding prediction results, that is, the extreme weather and climate events that may occur in the future;

[0075] In this embodiment, the climate analysis module processes and analyzes the collected weather data based on a machine learning algorithm to identify extreme weather and climate events. Extreme weather events include extreme high temperatures, extreme low temperatures, extreme precipitation, extreme drought, hail, strong winds, tornadoes, thunderstorms, and tropical cyclones. Based on the two major risk indicators of power generation loss and equipment damage rate (grid stability), the impact of extreme weather events on each power generation method is as follows:

[0076] 1) Photovoltaic power generation: Relying on sunlight, extreme high temperatures reduce the output power of modules, and hail and strong winds damage photovoltaic modules;

[0077] 2) Hydropower generation: It depends on water flow. Extreme drought leads to reduced water volume, affecting power generation. Extreme precipitation leads to floods, posing a threat to hydropower station facilities.

[0078] 3) Wind power generation: Relying on wind power, extreme low temperatures can cause wind turbine performance to decline, and strong winds, tornadoes, and thunderstorms can cause equipment damage;

[0079] 4) Geothermal power generation: relatively stable, but extreme geological activities (such as earthquakes) can cause damage to geothermal wells and power generation facilities;

[0080] 5) Biomass power generation: Relying on biomass raw materials, extreme weather may affect the supply and transportation of raw materials;

[0081] 6) Nuclear power generation: relatively stable, extreme weather (such as floods and hurricanes) poses a threat to nuclear power plant facilities;

[0082] 7) Solar thermal power generation: Relies on sunlight and high temperatures. Extremely high temperatures increase power generation efficiency, but extremely low temperatures cause system performance to decline.

[0083] The analysis results of the climate analysis module are output to the risk calculation module as an important basis for calculating the total risk value of the power system. At the same time, they are fed back to the power system operators or decision makers so that they can take timely measures to deal with the potential risks of extreme weather and climate events and conduct manual intervention;

[0084] Risk calculation module: Consider the risk value of each power generation mode under extreme weather and climate events, and combine the weight of each power generation mode to calculate the total risk value of the power system; the risk calculation formula is as follows:

[0085]

[0086] In the formula, S_risk represents the total risk value of the power system, n represents the number of types of power generation in the current system, i∈N+ and 1≤i≤n, R i represents the risk value of the i-th power generation mode, k irepresents the weight of the i-th power generation mode;

[0087] The calculation formula of risk value of different power generation modes is as follows:

[0088] 1) Photovoltaic power generation is mainly affected by light intensity and temperature under extreme weather conditions. The risk value calculation formula is based on power generation loss. Photovoltaic power generation risk value = P (extreme weather conditions) × (expected power generation - actual power generation) × economic loss coefficient, where P (extreme weather conditions) is the probability of extreme weather conditions occurring, and the economic loss coefficient is the coefficient that converts power generation loss into economic value;

[0089] 2) Hydropower generation is affected by water volume, water level, water flow speed and possible flood disasters under extreme weather conditions. The risk value calculation formula takes these factors into consideration: Hydropower generation risk value = P (extreme precipitation / drought) × (expected power generation - actual power generation) × economic loss coefficient + P (flood) × flood disaster loss, where P (extreme precipitation / drought) and P (flood) are the probabilities of extreme precipitation, drought and flood disasters, respectively.

[0090] 3) Wind power generation is mainly affected by wind speed, wind direction and possible extreme wind disasters (such as tornadoes and thunderstorms) in extreme weather. The risk value calculation formula is based on equipment damage and power generation loss. Wind power generation risk value = P (extreme wind speed) × (equipment damage cost + power generation loss × economic loss coefficient), where P (extreme wind speed) is the probability of extreme wind speed occurring;

[0091] 4) Geothermal power generation is affected by geological activities (such as earthquakes and volcanic eruptions) and possible equipment failures under extreme weather conditions. The risk value calculation formula is constructed based on these factors: Geothermal power generation risk value = P (geological activity) × the impact of geological activities on power generation facilities × economic loss coefficient + P (equipment failure) × equipment failure loss, where P (geological activity) is the probability of geological activity occurring, and P (equipment failure) is the probability of equipment failure occurring;

[0092] 5) Biomass power generation is affected by raw material supply, equipment failure and possible fire in extreme weather conditions. The risk value calculation formula is constructed based on these factors: biomass power generation risk value = P (insufficient raw material supply) × power generation loss caused by raw material shortage × economic loss coefficient + P (equipment failure) × equipment failure loss + P (fire) × fire loss, where P (insufficient raw material supply) is the probability of insufficient raw material supply, P (equipment failure) is the probability of equipment failure, and P (fire) is the probability of fire.

[0093] 6) Nuclear power generation mainly focuses on the possibility of nuclear accidents and their consequences under extreme weather conditions. Due to the particularity of nuclear power generation, the risk value calculation requires highly professional and detailed evaluation. The risk value of nuclear power generation is calculated through the probabilistic safety assessment (PSA) model, which takes into account various failure modes and possible consequences of nuclear facilities;

[0094] 7) Solar thermal power generation is mainly affected by solar radiation intensity and temperature in extreme weather conditions. Its risk value calculation formula is based on power generation loss, which is similar to the risk value calculation formula for photovoltaic power generation. The risk value of solar thermal power generation = P (solar radiation < minimum requirement) × power generation loss caused by insufficient solar radiation, where P (solar radiation < minimum requirement) is the probability of solar radiation < minimum requirement.

[0095] Power supply adjustment module: optimizes and adjusts the power supply mode of the power generation system. Specifically, when the total risk value of the system or the risk value of an individual power generation mode exceeds the threshold, the proportion of each power generation mode in the power generation system is dynamically adjusted to reduce the system risk. This module specifically includes the following sub-modules:

[0096] Risk monitoring submodule: real-time monitoring of the total risk value of the power generation system and the risk value of each power generation mode. When the risk value exceeds the preset threshold, the risk warning and adjustment mechanism are triggered;

[0097] Proportion adjustment submodule: dynamically adjust the proportion of each power generation mode in the power generation system according to the risk situation of the power system;

[0098] Instruction execution submodule: converts the results of the power generation mode ratio adjustment into specific dispatch instructions and executes them. The dispatch instructions include the start and stop of power generation facilities, output adjustment, etc. When executing the dispatch instructions, ensure the stable operation of the power generation system and take into account the constraints of the power grid;

[0099] Feedback optimization submodule: collects actual effect data after the execution of dispatch instructions, such as power generation, system stability, risk value changes, etc., evaluates the effectiveness of the adjustment strategy through comparative analysis, and provides feedback and optimization based on the evaluation results. The feedback and optimization process involves adjusting risk thresholds, optimizing adjustment algorithms or strategies, etc.;

[0100] Data communication submodule: used to communicate with other systems, such as power grid dispatching system, power generation facility control system, etc., to achieve data sharing and collaborative work;

[0101] In this embodiment, the proportion adjustment submodule determines the best combination of power generation modes through optimization algorithms such as linear programming, nonlinear programming, dynamic programming, and heuristic algorithms to balance the risks and benefits of the system, and the adjustment process includes increasing or decreasing the output of certain power generation modes, or starting / stopping certain power generation facilities; the selection and application of the optimization algorithm are as follows:

[0102] Linear programming: Applicable when the objective function and constraints are linear. Application method: By building a linear programming model, factors such as power generation output, system risk, and economic benefits are taken into consideration to solve for the optimal combination of power generation methods;

[0103] Nonlinear programming: Applicable when there are nonlinear relationships in the objective function or constraints. Application: Dealing with more complex power system optimization problems, such as considering the nonlinear characteristics of generators, nonlinear losses of transmission networks, etc.

[0104] Dynamic programming: Applicable to multi-stage decision-making problems, such as reservoir scheduling, wind farm power forecasting, etc. Application method: By building a dynamic programming model, the optimization scheduling problem of the power system is decomposed into multiple sub-problems, and they are solved in sequence to finally obtain the global optimal solution;

[0105] Heuristic algorithm: Applicable to complex, nonlinear, and non-convex optimization problems, such as unit combination and economic dispatch. Application method: By simulating natural processes or biological behaviors, the approximate optimal solution of the problem is found to solve the problem of complex power generation mode combination.

[0106] Result output module: Generates power system risk assessment results based on the calculated total risk value of the power system and the risk values ​​of all power generation modes, and outputs the assessment results to power system operators or decision makers in the form of charts, reports, etc.

[0107] The embodiments of the present invention are given for the purpose of illustration and description, and are not intended to be exhaustive or to limit the invention to the disclosed forms. Many modifications and variations will be apparent to those of ordinary skill in the art. The embodiments are selected and described in order to better illustrate the principles and practical applications of the present invention and to enable those of ordinary skill in the art to understand the present invention and thereby design various embodiments with various modifications suitable for specific uses.

Claims

1. An intelligent risk assessment device for power system based on big data and machine learning, characterized in that: include: Proportion analysis module, risk calculation module and power supply adjustment module; The proportion analysis module is used to count the types of power generation methods and their proportions in the current power system, and assign corresponding weights to each power generation method according to the proportion; The risk calculation module is used to consider the risk value of each power generation mode under extreme weather and climate events, and combine the weight of each power generation mode to calculate the total risk value of the power system; The risk calculation formula is: In the formula, S_risk represents the total risk value of the power system, n represents the number of types of power generation in the current system, i∈N+ and 1≤i≤n, R i represents the risk value of the i-th power generation mode, k i represents the weight of the i-th power generation mode; The power supply adjustment module is used to dynamically adjust the proportion of each power generation method in the power generation system when the total system risk value or the risk value of an individual power generation method exceeds a threshold value, so as to reduce the system risk.

2. The power system risk intelligent assessment device based on big data and machine learning according to claim 1 is characterized in that: The proportion analysis module specifically includes the following submodules: The data collection submodule is used to collect the power generation and installed capacity data of various power generation modes from channels such as power system databases and power generation enterprise information systems using big data analysis technology; The data processing submodule is used to clean, verify and integrate the collected data, including normalization and standardization operations on the data; The proportion calculation submodule is used to calculate the proportion of each power generation mode in the power system according to the power generation or installed capacity of each power generation mode, and the proportion calculation is in the form of percentage; The weight assignment submodule is used to assign corresponding weights according to the proportion of each power generation mode, combined with the actual situation of the power system and risk assessment requirements.

3. The power system risk intelligent assessment device based on big data and machine learning according to claim 1 is characterized in that: The device also includes: The data acquisition module is used to obtain weather data in real time using IoT sensors and meteorological monitoring equipment, and transmit it to the climate analysis module for processing; Climate analysis module, used to analyze weather data acquired in real time, identify and determine whether there are extreme weather and climate events that have a significant impact on various power generation methods; The result output module is used to generate the power system risk assessment results based on the calculated total risk value of the power system and the risk values ​​of all power generation modes, and output the assessment results to the power system operators or decision makers in the form of charts and reports.

4. The power system risk intelligent assessment device based on big data and machine learning according to claim 1 is characterized in that: The power generation methods of the power system include at least two of photovoltaic power generation, hydroelectric power generation, wind power generation, geothermal power generation, biomass power generation, nuclear power generation, and solar thermal power generation.

5. The power system risk intelligent assessment device based on big data and machine learning according to claim 3 is characterized in that: The data acquisition module receives weather data from IoT sensors and meteorological monitoring equipment in real time, verifies the received weather data, and then preprocesses the verified weather data. The data acquisition module transmits the preprocessed weather data to the risk assessment model in real time through the data interface, and ensures data synchronization between the data acquisition module and the risk assessment model.

6. The power system risk intelligent assessment device based on big data and machine learning according to claim 5 is characterized in that: The IoT sensor is deployed near the power facilities, including a temperature sensor, a humidity sensor, a wind speed and direction sensor, and a rain gauge. The IoT sensor captures and converts meteorological parameters into electrical signals in real time, and wirelessly transmits the data collected by the IoT sensor to the data acquisition module through the IoT communication protocol.

7. The power system risk intelligent assessment device based on big data and machine learning according to claim 5 is characterized in that: The meteorological monitoring equipment is equipped in a meteorological station within the coverage area of ​​the power system, and the data of the meteorological station is transmitted to the data acquisition module via wired or wireless means.

8. The power system risk intelligent assessment device based on big data and machine learning according to claim 3 is characterized in that: The climate analysis module specifically includes the following submodules: The feature extraction submodule is used to extract key features related to extreme weather and climate events from weather data; A model training submodule is used to train a machine learning model that can learn and identify patterns of extreme weather and climate events using historical weather data and known extreme weather and climate event data; The real-time prediction submodule is used to input real-time weather data into the trained machine learning model to perform real-time prediction and identification of extreme weather and climate events. The machine learning model determines whether the current weather is an extreme weather and climate event based on the characteristics of the input data and the learned patterns, and gives corresponding prediction results, i.e., extreme weather and climate events that may occur in the future. The extreme weather events include extreme high temperatures, extreme low temperatures, extreme precipitation, extreme drought, hail, strong winds, tornadoes, thunderstorms and tropical cyclones. Based on the two major risk indicators of power generation loss and equipment damage rate, extreme weather events have different impacts on each power generation method.

9. The power system risk intelligent assessment device based on big data and machine learning according to claim 4 is characterized in that: The calculation formula for the risk value of each power generation mode in the risk calculation formula is as follows: 1) Photovoltaic power generation risk value = P (extreme weather conditions) × (expected power generation - actual power generation) × economic loss coefficient, where P (extreme weather conditions) is the probability of extreme weather conditions occurring, and the economic loss coefficient is the coefficient that converts power generation loss into economic value; 2) Hydropower risk value = P (extreme precipitation / drought) × (expected power generation - actual power generation) × economic loss coefficient + P (flood) × flood disaster loss, where P (extreme precipitation / drought) and P (flood) are the probabilities of extreme precipitation, drought and flood disasters, respectively; 3) Wind power generation risk value = P (extreme wind speed) × (equipment damage cost + power generation loss × economic loss coefficient), where P (extreme wind speed) is the probability of extreme wind speed occurring; 4) Geothermal power generation risk value = P (geological activity) × the impact of geological activity on power generation facilities × economic loss coefficient + P (equipment failure) × equipment failure loss, where P (geological activity) is the probability of geological activity occurring, and P (equipment failure) is the probability of equipment failure occurring; 5) Biomass power generation risk value = P (raw material shortage) × power generation loss caused by raw material shortage × economic loss coefficient + P (equipment failure) × equipment failure loss + P (fire) × fire loss, where P (raw material shortage) is the probability of raw material shortage, P (equipment failure) is the probability of equipment failure, and P (fire) is the probability of fire. 6) The risk value of nuclear power generation is calculated through a probabilistic safety assessment model; 7) Solar thermal power generation risk value = P (solar radiation < minimum requirement) × power generation loss caused by insufficient solar radiation, where P (solar radiation < minimum requirement) is the probability of solar radiation < minimum requirement.

10. The power system risk intelligent assessment device based on big data and machine learning according to claim 1, characterized in that: The power supply adjustment module specifically includes the following submodules: The risk monitoring submodule is used to monitor the total risk value of the power generation system and the risk value of each power generation mode in real time. When the risk value exceeds the preset threshold, the risk warning and adjustment mechanism are triggered; The proportion adjustment submodule is used to dynamically adjust the proportion of each power generation mode in the power generation system according to the risk situation of the power system; The instruction execution submodule is used to convert the result of the power generation mode proportion adjustment into a specific dispatch instruction and execute it. When executing the dispatch instruction, the stable operation of the power generation system is ensured, and the constraints of the power grid are taken into consideration; The feedback optimization submodule is used to collect the actual effect data after the scheduling instruction is executed, evaluate the effectiveness of the adjustment strategy through comparative analysis, and provide feedback and optimization based on the evaluation results; The data communication submodule is used to communicate with other systems to achieve data sharing and collaborative work; The proportion adjustment submodule determines the best combination of power generation modes through an optimization algorithm to balance the risks and benefits of the system, and the adjustment process includes increasing or decreasing the output of certain power generation modes, or starting / stopping certain power generation facilities.

Images