Power grid risk early warning method and system based on artificial intelligence

Abstract

The invention provides a power grid risk early warning method and system based on artificial intelligence, and belongs to the technical field of power systems, and the method comprises the steps: determining a first type of power access node and a second type of power access node in a to-be-monitored power grid system, determining the output support capability of the first type of power supply access nodes and the output fluctuation state of the second type of power supply access nodes, and constructing a plurality of risk observation areas according to the matching result of the output support capability of the first type of power supply access nodes and the output fluctuation state of the second type of power supply access nodes, and determining a comprehensive risk level of each risk observation area, determining a risk level of the to-be-monitored power grid system according to a combination result of the comprehensive risk levels of the plurality of risk observation areas, determining an early warning strategy matched with the risk level of the to-be-monitored power grid system, and executing the early warning strategy. The invention aims to improve the problems of low early warning accuracy and poor universality of the existing scheme.

Description

Technical Field

[0001] The present application relates to the technical field of power systems, and in particular to an artificial intelligence-based power grid risk early warning method and system. Background Art

[0002] Grid risk early warning refers to the early detection of potential risks that may cause grid failure or instability through monitoring and analysis of the grid's operating status, external environment, load demand, and power supply. It aims to send out signals through the early warning system before the grid fails or faces serious safety hazards, and take corresponding measures to reduce or avoid accidents, thereby ensuring the safe, stable and efficient operation of the grid.

[0003] In traditional power systems, power sources have high controllability, so risk assessment of power grids mainly focuses on load changes and line failures. However, with the large-scale grid connection of renewable energy sources such as wind power and photovoltaics, the controllability and stability of power sources are relatively low, so the existing risk warning methods have poor versatility and low accuracy. Summary of the invention

[0004] The embodiment of the present application provides a power grid risk early warning method and system based on artificial intelligence. The present application adopts the following technical solutions:

[0005] In a first aspect, a power grid risk early warning method based on artificial intelligence is provided, the method comprising:

[0006] Determine a first type of power access node and a second type of power access node in the power grid to be monitored, the first type of power access node being a power access node with stable output, and the second type of power access node being a power access node with stable output;

[0007] Determine the output support capability of the first type power access node and the output fluctuation state of the second type power access node;

[0008] Constructing a plurality of risk observation areas according to a matching result of an output support capability of a first type of power access node and an output fluctuation state of a second type of power access node;

[0009] Determine the comprehensive risk level of each risk observation area, and determine the risk level of the power grid system to be monitored based on the combined results of the comprehensive risk levels of multiple risk observation areas;

[0010] Determine the early warning strategy that matches the risk level of the power grid system to be monitored and implement the early warning strategy.

[0011] In an embodiment of the first aspect of the present application, determining a first type of power access node and a second type of power access node in a power grid to be monitored includes:

[0012] Sending request information to all power access nodes to be identified in the power grid system to be monitored;

[0013] Receiving an identification data packet fed back by the power access node to be identified after receiving the request information, the identification data packet at least including historical load data and a type identifier of the power access node to be identified;

[0014] According to the type identifier, the type of the power access node to be identified is determined.

[0015] In an embodiment of the first aspect of the present application, determining the output support capability of the first type power access node and the output fluctuation state of the second type power access node includes:

[0016] Determining the output support capacity of the first type of power access node according to the load demand curve in the historical load data;

[0017] The output fluctuation state of the second type power supply access node is determined according to the output fluctuation curve in the historical load data.

[0018] In an embodiment of the first aspect of the present application, the risk observation area includes a first type power access node and multiple second type power access nodes, and multiple risk observation areas are constructed according to the matching result of the output support capability of the first type power access node and the output fluctuation state of the second type power access node, including:

[0019] Determine a first set according to a matching result of an output support capability of a first type power access node and an output fluctuation state of a second type power access node;

[0020] Determine a second set according to a matching result of the location information of the first type of power access node and the location information of the second type of power access node;

[0021] The risk observation area is determined according to the intersection of the first set and the second set.

[0022] In an embodiment of the first aspect of the present application, determining the comprehensive risk level of each risk observation area includes:

[0023] Obtaining a static image of the first monitoring period of the second type power access node in each risk observation area, and determining the external risk level of the second monitoring period of each risk observation area according to the static image of the first monitoring period;

[0024] Obtaining load data of the first type power access node and the second type power access node in each risk observation area during the first monitoring period, and determining the internal risk level of the second monitoring period of each risk observation area according to the load data of the first monitoring period;

[0025] Based on the external risk level and internal risk level of the second monitoring period of each risk observation area, the comprehensive risk level of the second monitoring period of each risk observation area is determined.

[0026] In an embodiment of the first aspect of the present application, determining the external risk level of each risk observation area in the second monitoring period according to the static image of the first monitoring period includes:

[0027] Determining a first risk assessment score according to a matching result between the behavioral features in the static image and a preset first risk assessment strategy;

[0028] Determining a second risk assessment score according to a matching result between the environmental features in the static image and a preset second risk assessment strategy;

[0029] An external risk level of a second monitoring period of each risk observation area is determined based on the first risk assessment score and the second risk assessment score.

[0030] In an embodiment of the first aspect of the present application, determining the internal risk level of the second monitoring period of each risk observation area according to the load data of the first monitoring period includes:

[0031] Construct input features based on the load data of the first monitoring period, and input the input features into a preset risk prediction model to obtain an initial internal risk level of each risk observation area in the second monitoring period;

[0032] Determine the correction factor based on the adjacent risk observation area;

[0033] The initial internal risk level of the second monitoring period is corrected according to the correction factor to obtain the final internal risk level of the second monitoring period for each risk observation area.

[0034] In an embodiment of the first aspect of the present application, determining the risk level of the power grid system to be monitored according to the combined result of the comprehensive risk levels of multiple risk observation areas includes:

[0035] Determine the risk weight of each risk observation area, and determine the risk prediction results of each risk observation area based on the risk weight and comprehensive risk level of each risk observation area;

[0036] The risk prediction results of each risk observation area are combined to determine the risk level of the power grid system to be monitored.

[0037] In the second aspect, based on the same inventive concept, a power grid risk early warning system based on artificial intelligence is provided, the system comprising:

[0038] A first determination module is used to determine a first type of power access node and a second type of power access node in the power grid to be monitored, the first type of power access node is a power access node with stable output, and the second type of power access node is a power access node with stable output;

[0039] A second determination module, used to determine the output support capability of the first type power access node and the output fluctuation state of the second type power access node;

[0040] A construction module, configured to construct a plurality of risk observation areas according to a matching result of an output support capability of a first type of power access node and an output fluctuation state of a second type of power access node;

[0041] A risk identification module is used to determine the comprehensive risk level of each risk observation area, and determine the risk level of the power grid system to be monitored based on the combined results of the comprehensive risk levels of multiple risk observation areas;

[0042] The execution module is used to determine the early warning strategy that matches the risk level of the power grid system to be monitored and execute the early warning strategy.

[0043] In an embodiment of the second aspect of the present application, the first determining module includes:

[0044] A sending submodule, used for sending request information to all power access nodes to be identified in the power grid system to be monitored;

[0045] A receiving submodule, configured to receive an identification data packet fed back by the power access node to be identified after receiving the request information, wherein the identification data packet at least includes historical load data and a type identifier of the power access node to be identified;

[0046] The identification submodule is used to determine the type of the power access node to be identified according to the type identifier.

[0047] In an embodiment of the second aspect of the present application, the second determining module includes:

[0048] A first determination submodule, configured to determine an output support capability of a first type of power access node according to a load demand curve in historical load data;

[0049] The second determination submodule is used to determine the output fluctuation state of the second type power supply access node according to the output fluctuation curve in the historical load data.

[0050] In an embodiment of the second aspect of the present application, a building block comprises:

[0051] A first matching submodule, configured to determine a first set according to a matching result of an output support capability of a first type power access node and an output fluctuation state of a second type power access node;

[0052] A second matching submodule, configured to determine a second set according to a matching result of the location information of the first type of power access node and the location information of the second type of power access node;

[0053] The combination submodule is used to determine the risk observation area according to the intersection of the first set and the second set.

[0054] In an embodiment of the second aspect of the present application, the risk identification module includes

[0055] The external risk identification submodule is used to obtain a static image of the first monitoring period of the second type power access node in each risk observation area, and determine the external risk level of the second monitoring period of each risk observation area according to the static image of the first monitoring period;

[0056] An internal risk identification submodule, used to obtain the load data of the first type power access node and the second type power access node in each risk observation area during the first monitoring period, and determine the internal risk level of the second monitoring period of each risk observation area according to the load data of the first monitoring period;

[0057] The comprehensive risk identification submodule is used to determine the comprehensive risk level of each risk observation area in the second monitoring period according to the external risk level and internal risk level of each risk observation area in the second monitoring period.

[0058] In an embodiment of the second aspect of the present application, the internal risk identification submodule includes:

[0059] A first evaluation unit, configured to determine a first risk assessment score according to a matching result between the behavior feature in the static image and a preset first risk assessment strategy;

[0060] A second evaluation unit, configured to determine a second risk assessment score according to a matching result between the environmental features in the static image and a preset second risk assessment strategy;

[0061] The comprehensive evaluation unit is used to determine the external risk level of the second monitoring period of each risk observation area according to the first risk assessment score and the second risk assessment score.

[0062] In an embodiment of the second aspect of the present application, the external risk identification submodule includes:

[0063] A first evaluation unit is used to construct input features according to the load data of the first monitoring period, and input the input features into a preset risk prediction model to obtain an initial internal risk level of each risk observation area in the second monitoring period;

[0064] A correction factor determination unit, used to determine the correction factor according to adjacent risk observation areas;

[0065] The second evaluation unit is used to correct the initial internal risk level of the second monitoring period according to the correction factor to obtain the final internal risk level of the second monitoring period of each risk observation area.

[0066] In an embodiment of the second aspect of the present application, the risk identification module further includes:

[0067] A determination submodule is used to determine the risk weight of each risk observation area, and determine the risk prediction result of each risk observation area according to the risk weight and comprehensive risk level of each risk observation area;

[0068] The combination submodule is used to combine the risk prediction results of each risk observation area to determine the risk level of the power grid system to be monitored.

[0069] In summary, the above-mentioned power grid risk early warning method and system based on artificial intelligence have the following technical effects:

[0070] By accurately identifying and classifying the power access nodes, the power access nodes are divided into two categories. For these two types of nodes, the system evaluates their output support capabilities and fluctuation states respectively, so as to more accurately understand their impact on the power grid. After completing these evaluations, the system will construct multiple risk observation areas based on the evaluation results. These observation areas represent different areas in the power grid system, and the risk level of each area will be comprehensively evaluated. By analyzing the comprehensive risk level of each risk observation area, the system can reflect the actual risk situation of the power grid in different areas in detail. This refined risk assessment method makes the risk understanding of the power grid system more comprehensive and in-depth. Subsequently, the system will match and execute the corresponding early warning strategy according to these comprehensive risk levels. In summary, the method of the present application effectively responds to the instability and uncontrollability brought about by renewable energy by refining risk analysis, improving assessment accuracy, and implementing dynamic response strategies. This method significantly improves the versatility and prediction accuracy of the power grid system, enabling it to better adapt to and respond to the complex and changing operating environment brought about by renewable energy. BRIEF DESCRIPTION OF THE DRAWINGS

[0071] Figure 1 A schematic diagram of the steps of an artificial intelligence-based power grid risk early warning method provided in an embodiment of the present application;

[0072] Figure 2 A schematic diagram of the functional modules of an artificial intelligence-based power grid risk warning system provided in an embodiment of the present application. DETAILED DESCRIPTION

[0073] The terms used in the following embodiments are only for the purpose of describing specific embodiments, and are not intended to be used as limitations to the present application. As used in the specification and appended claims of the present application, the singular expressions "one", "a kind of", "", "above", "the" and "this" are intended to also include expressions such as "one or more", unless there is a clear contrary indication in the context. It should also be understood that in the following embodiments of the present application, "at least one", "one or more" refer to one or more than two (including two). The character " / " generally indicates that the objects associated before and after are in a "or" relationship.

[0074] The technical solutions in the embodiments of the present application will be described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments.

[0075] In the following, the terms "first", "second", etc. are used only for convenience of description and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, a feature defined as "first", "second", etc. may explicitly or implicitly include one or more of the features. In the description of this application, unless otherwise specified, "plurality" means two or more. For example, a plurality of processing units refers to two or more processing units.

[0076] In addition, in the embodiments of the present application, "upper", "lower", "left" and "right" are not limited to being defined relative to the orientation of the components schematically placed in the drawings. It should be understood that these directional terms can be relative concepts, which are used for description and clarification relative to the components, and can change accordingly according to the change in the orientation of the components placed in the drawings. In the drawings, for the sake of clarity, the thickness of the layers and regions is exaggerated, and the size ratio relationship between the parts in the drawings does not reflect the actual size ratio relationship.

[0077] In the embodiments of the present application, unless otherwise clearly specified and limited, the term "connection" should be understood in a broad sense. For example, "connection" can be a fixed connection, a detachable connection, or an integral connection; it can be a direct connection or an indirect connection through an intermediate medium. In addition, the term "electrical connection" can be a direct electrical connection or an indirect electrical connection through an intermediate medium.

[0078] In the embodiments of the present application, the term "module" is generally a functional structure divided according to logic, and the "module" can be implemented by pure hardware, or by a combination of software and hardware. In the embodiments of the present application, "and / or" describes the association relationship of associated objects, indicating that three relationships can exist. For example, A and / or B can represent: A exists alone, B exists alone, and A and B exist at the same time.

[0079] In the embodiments of the present application, words such as "exemplary" or "for example" are used to indicate examples, illustrations or descriptions. Any embodiment or design described as "exemplary" or "for example" in the embodiments of the present application should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of words such as "exemplary" or "for example" is intended to present related concepts in a specific way.

[0080] The technical solution in this application will be described below in conjunction with the accompanying drawings.

[0081] In traditional power systems, the main reliance is on highly controllable power sources, such as thermal power, hydropower, and nuclear power. These power sources have relatively stable power output, and dispatchers can flexibly adjust the output power of the power sources according to the load changes and needs of the power grid, thereby ensuring the continuity and stability of power supply. Therefore, in such a system, the risk assessment of the power grid mainly focuses on controllable factors such as load changes and line failures.

[0082] With the large-scale access of renewable energy such as wind power and photovoltaic power generation, the operating environment of the power grid has undergone major changes. The characteristics of this type of energy are significantly different from traditional power sources. Due to the instability and difficulty of control of renewable energy, the output of power sources becomes uncontrollable. Therefore, traditional risk assessment and early warning methods are no longer reliable in terms of universality and accuracy.

[0083] Based on this, the inventive concept of the present application is proposed: by distinguishing and matching the characteristics of different types of power access nodes in the power grid, more accurate risk assessment and early warning can be achieved. Specifically, by dividing the power access nodes into the first type of power access nodes with stable output (such as traditional power generation sources) and the second type of power access nodes with large output fluctuations (such as wind power, photovoltaic and other renewable energy sources), their output support capabilities and fluctuation characteristics are evaluated respectively. Then, based on the matching results of these two types of power nodes, multiple risk observation areas are constructed to evaluate the comprehensive risk level of each observation area. According to the comprehensive risk levels of multiple risk observation areas, the risk level of the entire power grid system is determined, and corresponding early warning strategies are formulated accordingly. It can cope with the uncertainty and volatility problems brought about by the access of renewable energy to the power grid, improve the versatility and accuracy of existing power grid risk early warning methods, and thus ensure the safe and stable operation of the power grid.

[0084] Reference Figure 1 The embodiment of the present invention provides a power grid risk early warning method based on artificial intelligence, which is applied to a server and may specifically include the following steps:

[0085] S101: Determine a first type of power access node and a second type of power access node in a power grid to be monitored.

[0086] In this embodiment, first, according to the characteristics of each power source in the power grid system, the power access nodes are divided into the first type of power access nodes and the second type of power access nodes. Among them, the first type of power access nodes refer to power access nodes with relatively stable output, usually including traditional power generation units such as thermal power, hydropower, and nuclear power. These power sources have strong scheduling flexibility and output controllability. It is characterized by being able to provide relatively continuous and stable output support when the grid load changes, ensuring the smooth operation of the system.

[0087] The second type of power access nodes are those with large output fluctuations, mainly including wind power, photovoltaic and other renewable energy power stations. The output power of this type of power source is greatly affected by natural factors such as weather, sunshine, wind speed, etc., and has obvious intermittent and uncertain characteristics. Therefore, its power generation is difficult to accurately predict, which directly affects the stability of the power grid and the reliability of the dispatch plan.

[0088] The specific steps of determining that the server determines the first type of power access node and the second type of power access node in the power grid to be monitored may include:

[0089] S1011: Sending request information to all power access nodes to be identified in the power grid system to be monitored;

[0090] S1012: receiving an identification data packet fed back by the power access node to be identified after receiving the request information, where the identification data packet at least includes historical load data and a type identifier of the power access node to be identified;

[0091] S1013: Determine the type of the power access node to be identified according to the type identifier.

[0092] In the implementation of S1011 to S1013, first, the server sends a request message to all power access nodes to be identified in the monitoring power grid system to obtain detailed information of these power access nodes. The information of the power access node to be identified usually includes multiple parameters, such as the type of power access node, geographical location, and historical load data of the node. It should be noted that these parameters are not limited to the above content, but may also include other relevant information, such as the operating status of the node, adjustable capacity, access duration, etc. The specific content can be expanded according to actual needs, and this application does not limit this.

[0093] After receiving the request information sent by the server, the power access node to be identified will process its own relevant data, integrate historical load data, location coordinates, and type identification into an identification data packet, and prepare to feed it back to the server. To ensure data security and privacy during communication, the identification data packet can be encrypted to prevent data from being stolen or tampered with by the outside during transmission. Encryption can effectively ensure the security of key information of the power system.

[0094] After the server receives the identification data packet fed back by the power access node to be identified, the server can perform further data processing. First, the server can optionally decrypt the identification data packet to restore the original data. After the decryption is completed, the server will determine the type of each power access node based on the type identifier in the identification data packet.

[0095] S102: Determine the output support capability of the first type power access node and the output fluctuation state of the second type power access node.

[0096] In this embodiment, after determining the type of each power access node to be identified, the server needs to evaluate different attributes according to the specific type. Output support capability refers to the ability of the first type of power access node to provide power continuously and stably within a certain range. This type of power supply is usually highly adjustable and controllable, and can quickly adjust the power generation power according to the grid load demand to ensure the stability of the grid power supply. The output fluctuation state refers to the fluctuation of the power generation output of the second type of power access node with external conditions. Since the second type of power access node relies on natural resources, the output has significant intermittent and uncertainty, which is difficult to control manually. By evaluating the output support capability of the first type of power access node and the output fluctuation state of the second type of power access node, a more accurate scheduling strategy and risk warning means can be provided for the operation of the power grid. The specific evaluation steps may include:

[0097] S1021: Determine the output support capacity of the first type of power access node according to the load demand curve in the historical load data;

[0098] S1022: Determine an output fluctuation state of the second type power access node according to an output fluctuation curve in historical load data.

[0099] In the implementation of S1021 to S1022, the load demand curve reflects the load demand of the power grid in different time periods, that is, the power supply required by the power grid at a specific time. The output fluctuation curve reflects the output fluctuation of renewable energy power generation, which is greatly affected by meteorological conditions (such as wind speed, solar radiation intensity, etc.). Unlike traditional power sources, the power generation of renewable energy such as wind power and photovoltaic power is not linearly controllable, but fluctuates with changes in time and weather conditions.

[0100] When determining the output support capacity, by analyzing the peak period of the load demand curve, it can be determined whether the first type of power access node has sufficient output capacity to cope with the load peak in the power grid. During the load peak period, the power demand reaches the maximum, and the first type of power access node must have sufficient power output to support the smooth operation of the power grid. Therefore, by comparing the maximum output capacity of the power supply with the load demand peak, the support capacity of the power supply during the load peak period is determined. The valley period of the load demand curve corresponds to the time period with the lowest power demand. In this case, the first type of power access node should have the ability to adjust the output power to avoid excessive power generation when the load is low. This requires evaluating the ability of the power supply to maintain low output or flexibly adjust the output during low load periods to ensure the safe operation of the power grid during the load valley period. The volatility of the load demand curve reflects the speed and amplitude of the change of the load demand of the power grid. When evaluating the output support capacity of the first type of power access node, it is necessary to analyze whether the power supply can respond quickly to load changes and adjust the output power in time according to the real-time fluctuation of the load demand of the power grid. The response speed, adjustment flexibility and output stability of the power supply are key evaluation indicators. Through the long-term load demand curve (such as monthly or annual), it is possible to analyze whether the first type of power access node has the ability to continuously support the grid load demand over a long period of time. For example, if the load is higher than the average level for a long time, whether the power supply can maintain efficient and stable output for a long time.

[0101] When determining the output fluctuation state, by analyzing the output fluctuation curve, the change range of the power generation of the power source within a certain period of time can be determined. For example, the power generation of wind power and photovoltaic power will fluctuate violently with changes in wind speed or solar radiation intensity. The larger the fluctuation range, the greater the output instability of the power source. Based on these fluctuation characteristics, the maximum and minimum output power range of the power source can be evaluated to understand the impact of its volatility on the stability of the power grid. The output fluctuation curve can also show the fluctuation frequency of the power supply output, that is, the speed of power supply power change. For example, wind power may increase or decrease output significantly within a few minutes, while photovoltaic power generation may fluctuate frequently when clouds move quickly. By evaluating the fluctuation frequency, it can be determined whether the power generation fluctuation of the power source has a certain periodicity or suddenness, which helps to predict the impact of power supply fluctuations on the stability of the power grid. By analyzing the output fluctuation curve for a long time, the output uncertainty of the power source can be evaluated. For example, the power generation of wind power depends on the wind speed, which may vary in different time periods. Analyzing the historical data of the output curve can identify the predictable and unpredictable parts of the power supply output, helping the power grid dispatching system to better deal with the impact of its fluctuations on the power grid. By comparing the actual output curve with the predicted output model, the fluctuation uncertainty of the power supply can be further quantified.

[0102] S103: constructing a plurality of risk observation areas according to the matching results of the output support capability of the first type power access node and the output fluctuation state of the second type power access node.

[0103] In this embodiment, by combining the first type of power access node with stability and the second type of power access node without stability, a risk observation area with dynamic stability can be obtained. The purpose of constructing a risk observation area is to more effectively evaluate the operation risk of the power grid system and to perform zone monitoring and management according to the power supply and demand characteristics of different regions. The power access nodes in the power grid have different characteristics. The output of the first type of power access node is stable, while the output of the second type of power access node fluctuates greatly. Combining them can achieve dynamic balance in the local area. This combination helps to smooth the fluctuations caused by renewable energy and enhance the stability of the local power grid. By constructing a risk observation area, potential power supply risks can be identified according to the power supply configuration and load demand conditions in different regions. In each observation area, the stable power supply node can supplement the insufficient output of the fluctuating power supply node, while the fluctuating power supply node can provide flexibility and adjustment capabilities. Doing so will not only help reduce the impact of power fluctuations in the local area on the entire power grid system, but also improve the dynamic dispatching efficiency of the power grid, and provide early warning and respond to possible operation risks. The specific steps for constructing a risk observation area may include:

[0104] S1031: Determine a first set according to a matching result of an output support capability of a first type power access node and an output fluctuation state of a second type power access node;

[0105] S1032: Determine a second set according to a matching result of the location information of the first type power access node and the location information of the second type power access node;

[0106] S1033: Determine a risk observation area according to the intersection of the first set and the second set.

[0107] In the implementation methods of S1031 to S1033, each risk observation area may include a first type power access node and multiple second type power access nodes. By matching the characteristics of these two types of power access nodes, it is possible to find which stable power access nodes can provide sufficient support when wind power or photovoltaic power fluctuates. The matching process may involve evaluating the compatibility of the two in terms of power output, response speed, regulation capability, etc. For example, wind power nodes with large fluctuations can be matched with thermal power nodes with strong output support capabilities. After this matching analysis, a combination of power access nodes that can support each other can be determined. All these combinations constitute a first set, which contains stable power nodes that match their respective fluctuation characteristics. This set lays the foundation for the subsequent division of risk observation areas, because the complementarity between power sources helps reduce risks.

[0108] Geographical location is crucial for dispatching, energy transmission and risk management in power grid operation. The goal of matching is to find power nodes that are geographically close to each other to improve the efficiency of the power grid and reduce losses in energy transmission. First, the geographical location information of the power nodes needs to be determined. By matching the location information of the first type of power access nodes and the second type of power access nodes, find those node combinations that are close to each other. Geographically close nodes usually mean that the energy transmission loss between them is small and the dispatching flexibility is high, so they are more suitable for risk observation as a whole. After matching, the node combinations that are geographically close are divided into the second set. The combinations of these nodes are spatially interrelated and can be managed as a region. At this time, the nodes in the second set are mainly grouped based on geographical factors.

[0109] The intersection represents power nodes that meet both the output characteristic matching requirements (first set) and the geographical proximity requirements (second set). In other words, these nodes are not only complementary in characteristics, but also close in space, suitable for risk monitoring and management as a whole. The combination of nodes in the intersection will form a risk observation area. Within an observation area, power fluctuations can be adjusted by nearby stable power sources to reduce the impact on the overall grid stability. In addition, the combination of nodes with close geographical locations makes power dispatch and transmission within the region more efficient, reducing the system's energy transmission loss and response delay.

[0110] S104: Determine the comprehensive risk level of each risk observation area, and determine the risk level of the power grid system to be monitored based on the combined results of the comprehensive risk levels of multiple risk observation areas.

[0111] In this implementation, after completing the division of risk observation areas, it is next necessary to conduct detailed prediction and identification of the comprehensive risk level of each risk observation area. After completing the comprehensive risk assessment of each risk observation area, the risk levels of multiple risk observation areas are further integrated according to the risk levels of these observation areas to determine the overall risk level of the entire power grid system to be monitored. This process can effectively capture the impact of local risks on the global situation, thereby forming an accurate assessment of the overall operating status of the power grid system. In this way, a reliable risk warning can be provided for the safety of power grid operation, ensuring that corresponding preventive measures are taken in time before risks occur, thereby ensuring the stable and efficient operation of the power grid.

[0112] The steps to determine the comprehensive risk level of each risk observation area include:

[0113] S1041: Acquire a static image of the second type power access node in each risk observation area during the first monitoring period, and determine the external risk level of each risk observation area during the second monitoring period according to the static image of the first monitoring period;

[0114] S1042: Obtain load data of the first type power access nodes and the second type power access nodes in each risk observation area during the first monitoring period, and determine the internal risk level of each risk observation area during the second monitoring period according to the load data of the first monitoring period;

[0115] S1043: Determine the comprehensive risk level of each risk observation area in the second monitoring period according to the external risk level and the internal risk level of each risk observation area in the second monitoring period.

[0116] In the implementation of S1041 to S1043, the comprehensive risk of the risk observation area can be determined by two parts. The first part is the external risk of the risk observation area. Due to the existence of the second type of power access node, the possibility of the existence of external risk is greatly increased. The second part is the internal risk of the risk observation area.

[0117] The static images of the second type of power access nodes (such as wind power, photovoltaic and other renewable energy nodes) in each risk observation area during the first monitoring cycle are obtained through monitoring equipment (such as cameras, sensors, etc.). These images provide external information about the power grid environment during this time period, especially the working status of wind power and photovoltaic power generation equipment. By analyzing these static images, some key information about the external environment can be obtained, such as weather conditions (such as cloudy, wind speed, sunshine, etc.), and the operating status of power generation equipment (whether it is operating normally, whether there is external damage, etc.). These environmental and equipment status information can directly affect the power generation output of the second type of power access node. The external risk level is determined based on the fluctuation of these external environmental conditions. For example, when the static image shows that the wind speed decreases or the light is insufficient, the external risk level may rise, indicating that there may be a risk of fluctuation in the power supply of the power grid in the next monitoring cycle.

[0118] The server obtains the load data of the first monitoring cycle from the first type of power access nodes and the second type of power access nodes in each risk observation area. These load data include the actual output power, load fluctuations, and the degree of matching between load and output of each power node during the time period. Based on these load data, the server can evaluate the supply and demand balance of each power node during the first monitoring cycle. For example, if the load demand is high and the power output is insufficient, or the power fluctuation is large but the load demand is stable, the internal risk level will increase. The internal risk level is a risk assessed based on the supply and demand balance reflected by the load data and the performance of each power source. The load data can reveal the performance of the power supply equipment in matching its own output with the grid load demand, thereby quantitatively evaluating the internal risk.

[0119] The server combines the external risk level and the internal risk level to determine the comprehensive risk level of each risk observation area in the second monitoring cycle.

[0120] The external risk level is assessed based on environmental factors (such as weather conditions, external equipment status, etc.), while the internal risk level is assessed based on the degree of matching between power nodes and loads. The comprehensive risk level combines these two aspects of risk information to more comprehensively reflect the overall risk that each risk observation area may face in the next monitoring cycle.

[0121] The comprehensive risk level can be combined with external and internal risk conditions through certain weights or algorithms. For example, if the external risk level is very high (such as bad weather conditions), but the internal risk is low (such as good load balance), the comprehensive risk level may be medium risk. Conversely, if both internal and external risks are high, the comprehensive risk level will be higher. It should be noted that the embodiment of the present application does not limit the algorithm for determining the comprehensive risk level based on the external risk level and the internal risk level.

[0122] The embodiment of the present application uses the external environment of the second type of power supply node as a reference factor for risk prediction. By incorporating these external environmental factors into the risk prediction model, the system can more accurately evaluate and predict the potential risks of these nodes, thereby providing more detailed and effective risk warnings. This method not only improves the comprehensiveness and accuracy of risk prediction, but also optimizes the ability of the power grid system to respond to complex and dynamic environmental changes, ensuring the stability and reliability of power grid operation.

[0123] In a feasible implementation, determining the external risk level of each risk observation area in the second monitoring period according to the static image of the first monitoring period includes:

[0124] Determining a first risk assessment score according to a matching result between the behavioral features in the static image and a preset first risk assessment strategy;

[0125] Determining a second risk assessment score according to a matching result between the environmental features in the static image and a preset second risk assessment strategy;

[0126] An external risk level of a second monitoring period of each risk observation area is determined based on the first risk assessment score and the second risk assessment score.

[0127] In this embodiment, when determining the external risk level of the second monitoring period of the risk observation area, first, the server obtains static images of the second type of power access nodes in each risk observation area in the first monitoring period. These images can come from monitoring cameras or sensors. The server analyzes the behavioral features in the static images. These behavioral features may include the operating status of the equipment, operating behavior, and any abnormal activities (such as equipment failure, maintenance activities, etc.). The server matches these features with the preset first risk assessment strategy to evaluate the potential impact of these behaviors on external risks. Behavioral features, for example, whether the rotation state of the wind turbine of the wind power equipment is normal, whether the angle of the photovoltaic panel is appropriate, etc. Based on the matching results of the behavioral features and the strategy, the system calculates a score to quantify the degree of influence of the equipment behavior on the external risk. Abnormal behavior characteristics usually result in a higher risk assessment score. The environmental features in the static image are matched with the preset second risk assessment strategy. Environmental features include the impact of weather conditions, light intensity, temperature, etc. on the operation of the equipment. Environmental features, such as the light intensity of the photovoltaic system, the wind speed of the wind power system, etc. Based on the matching results of the environmental features and the strategy, a score is calculated to evaluate the degree of influence of environmental factors on external risks. Extremes in environmental factors result in higher risk assessment scores.

[0128] Finally, the first risk assessment score and the second risk assessment score are combined to comprehensively determine the external risk level of each risk observation area in the second monitoring period. This step obtains an overall external risk level by weighting the two risk assessment scores.

[0129] In a feasible implementation manner, determining the internal risk level of each risk observation area in the second monitoring period according to the load data of the first monitoring period includes:

[0130] Construct input features based on the load data of the first monitoring period, and input the input features into a preset risk prediction model to obtain an initial internal risk level of each risk observation area in the second monitoring period;

[0131] Determine the risk correction factor for the second monitoring cycle based on adjacent risk observation areas;

[0132] The initial internal risk level of the second monitoring period of the risk observation area is corrected according to the risk correction factor of the second monitoring period to obtain the final internal risk level of the second monitoring period of each risk observation area.

[0133] In this embodiment, when determining the external risk level of the second monitoring period of the risk observation area, first, the load data is collected from the first monitoring period. These data generally include the actual power output, load demand, load fluctuation, etc. of each power node and load node in the power grid. These load data are converted into input features, which may include the peak value, valley value, average value, load change rate, power factor, etc. of the load demand.

[0134] Input the constructed input features into the preset risk prediction model. The model can be a prediction model based on machine learning (such as regression model, neural network, decision tree, etc.) or a traditional model based on statistical analysis. After training, the model can predict the internal risk level based on the input features. The prediction result is the initial internal risk level, which represents the internal risk situation in the second monitoring period.

[0135] The mutual influence or environmental differences between the risk observation areas may lead to deviations in the initial predicted risk level. The risk correction factor adjusts the prediction results by considering the correlation between these areas. Therefore, the risk situation of adjacent risk observation areas is evaluated, which can be achieved by comparing the load data, risk level and other relevant factors of these areas. A risk correction factor is determined, which is a coefficient to adjust the predicted risk level based on the risk assessment results of the adjacent areas. The determination of the correction factor can be adjusted according to the environmental factors, load matching degree, operation conditions, etc. of the adjacent areas.

[0136] The initial internal risk level is revised using the determined risk correction factor. The revision process may include weighted adjustments or other calculation methods to reflect the impact of adjacent areas. The final internal risk level is the revised result, which takes into account the output of the forecast model and the impact of adjacent areas to provide a more accurate risk assessment. The final internal risk level more accurately reflects the actual operating risk of the grid system during the second monitoring period. By revising the initial forecast results to take into account the impact of adjacent areas, risk management and emergency response planning can be more accurately carried out.

[0137] In a feasible implementation, the risk level of the power grid system to be monitored is determined according to the combined results of the comprehensive risk levels of multiple risk observation areas, including:

[0138] S1044: Determine the risk weight of each risk observation area, and determine the risk prediction result of each risk observation area based on the risk weight and comprehensive risk level of each risk observation area;

[0139] S1045: Combine the risk prediction results of each risk observation area to determine the risk level of the power grid system to be monitored.

[0140] In the implementation methods of S1044 to S1045, the risk weight is a coefficient used to reflect the importance or influence of each risk observation area in the overall power grid risk assessment. Different risk observation areas may have different degrees of influence on the power grid system due to factors such as their location, the wrong number of power types, and load characteristics, so different weights need to be assigned. Risk weights can be assigned through historical data analysis, expert evaluation, or algorithmic models. For example, the weighted average method can be used to combine the comprehensive risk level of each risk observation area with its weight to form a specific risk weight value. For example, assuming that the comprehensive risk level of risk observation area A is 0.7 and the risk weight is 0.4, the risk prediction result is 0.7*0.4=0.28. The risk prediction results quantify the contribution of each risk observation area to the overall risk of the power grid system, and these results will be used to calculate the overall risk level of the power grid system.

[0141] Then, the risk prediction results of all risk observation areas are weighted and synthesized to obtain the overall risk level of the power grid system to be monitored. Common methods include weighted average, sum calculation, etc. The above combination results are used as the overall risk level of the power grid system to be monitored. This level reflects the total risk level of the power grid system after comprehensively considering the risk factors of each risk observation area. The overall risk level can be divided into several levels (for example, low, medium, and high) to facilitate risk management and decision-making. The specific classification criteria can be set according to the actual operation of the power grid and risk management requirements.

[0142] S105: Determine an early warning strategy that matches the risk level of the power grid system to be monitored, and execute the early warning strategy.

[0143] In this embodiment, first, the server matches the calculated overall risk level of the power grid system with the preset early warning strategy. Early warning strategies are usually operation guidelines or response measures set based on different risk levels, which are used to take appropriate actions when the power grid faces potential risks. For example, for low risk levels, early warning strategies may include routine monitoring and maintenance measures, while for high risk levels, more urgent emergency response measures may be required.

[0144] The server will then compare the risk level of the power grid system to be monitored with these early warning strategies and select the most appropriate strategy to deal with the current risk situation. The selected early warning strategy will include specific measures and operational steps, which may include adjusting power scheduling, strengthening equipment maintenance, increasing monitoring frequency, and launching emergency response plans.

[0145] After determining the appropriate early warning strategies, the server immediately executes them. This means putting the measures defined in the strategy into action, such as notifying relevant personnel, dispatching grid resources, adjusting load configuration, or conducting equipment inspections. The system may also send alarms or notifications through automated systems.

[0146] The artificial intelligence-based grid risk early warning method provided in the present application divides the power access nodes into two categories by accurately identifying and classifying the power access nodes. For these two types of nodes, the system evaluates their output support capabilities and fluctuation states respectively, so as to more accurately understand their impact on the power grid. After completing these evaluations, the system will construct multiple risk observation areas based on the evaluation results. These observation areas represent different areas in the power grid system, and the risk level of each area will be comprehensively evaluated. By analyzing the comprehensive risk level of each risk observation area, the system can reflect the actual risk situation of the power grid in different areas in detail. This refined risk assessment method makes the risk understanding of the power grid system more comprehensive and in-depth. Subsequently, the system will match and execute the corresponding early warning strategy according to these comprehensive risk levels. In summary, the method of the present application effectively responds to the instability and uncontrollability brought about by renewable energy by refining risk analysis, improving assessment accuracy, and implementing dynamic response strategies. This method significantly improves the versatility and prediction accuracy of the power grid system, enabling it to better adapt to and respond to the complex and changing operating environment brought about by renewable energy.

[0147] In the second aspect, based on the same inventive concept, refer to Figure 2 , shows an artificial intelligence-based power grid risk early warning system 200 provided in an embodiment of the present application, the system comprising:

[0148] A first determination module 201 is used to determine a first type of power access node and a second type of power access node in a power grid to be monitored, wherein the first type of power access node is a power access node with stable output, and the second type of power access node is a power access node with stable output;

[0149] A second determination module 202, configured to determine an output support capability of a first type power access node and an output fluctuation state of a second type power access node;

[0150] A construction module 203, configured to construct a plurality of risk observation areas according to a matching result of an output support capability of a first type of power access node and an output fluctuation state of a second type of power access node;

[0151] The risk identification module 204 is used to determine the comprehensive risk level of each risk observation area, and determine the risk level of the power grid system to be monitored based on the combined results of the comprehensive risk levels of multiple risk observation areas;

[0152] The execution module 205 is used to determine an early warning strategy that matches the risk level of the power grid system to be monitored and execute the early warning strategy.

[0153] In an embodiment of the present application, the first determination module includes:

[0154] A sending submodule, used for sending request information to all power access nodes to be identified in the power grid system to be monitored;

[0155] A receiving submodule, configured to receive an identification data packet fed back by the power access node to be identified after receiving the request information, wherein the identification data packet at least includes historical load data and a type identifier of the power access node to be identified;

[0156] The identification submodule is used to determine the type of the power access node to be identified according to the type identifier.

[0157] In an embodiment of the present application, the second determination module includes:

[0158] A first determination submodule, configured to determine an output support capability of a first type of power access node according to a load demand curve in historical load data;

[0159] The second determination submodule is used to determine the output fluctuation state of the second type power supply access node according to the output fluctuation curve in the historical load data.

[0160] In an embodiment of the present application, a building block comprises:

[0161] A first matching submodule, configured to determine a first set according to a matching result of an output support capability of a first type power access node and an output fluctuation state of a second type power access node;

[0162] A second matching submodule, configured to determine a second set according to a matching result of the location information of the first type of power access node and the location information of the second type of power access node;

[0163] The combination submodule is used to determine the risk observation area according to the intersection of the first set and the second set.

[0164] In an embodiment of the present application, the risk identification module includes

[0165] The external risk identification submodule is used to obtain a static image of the first monitoring period of the second type power access node in each risk observation area, and determine the external risk level of the second monitoring period of each risk observation area according to the static image of the first monitoring period;

[0166] An internal risk identification submodule, used to obtain the load data of the first type power access node and the second type power access node in each risk observation area during the first monitoring period, and determine the internal risk level of the second monitoring period of each risk observation area according to the load data of the first monitoring period;

[0167] The comprehensive risk identification submodule is used to determine the comprehensive risk level of each risk observation area in the second monitoring period according to the external risk level and internal risk level of each risk observation area in the second monitoring period.

[0168] In an embodiment of the present application, the internal risk identification submodule includes:

[0169] A first evaluation unit, configured to determine a first risk assessment score according to a matching result between the behavior feature in the static image and a preset first risk assessment strategy;

[0170] A second evaluation unit, configured to determine a second risk assessment score according to a matching result between the environmental features in the static image and a preset second risk assessment strategy;

[0171] The comprehensive evaluation unit is used to determine the external risk level of the second monitoring period of each risk observation area according to the first risk assessment score and the second risk assessment score.

[0172] In an embodiment of the present application, the external risk identification submodule includes:

[0173] A first evaluation unit is used to construct input features according to the load data of the first monitoring period, and input the input features into a preset risk prediction model to obtain an initial internal risk level of each risk observation area in the second monitoring period;

[0174] A correction factor determination unit, used to determine the correction factor according to adjacent risk observation areas;

[0175] The second evaluation unit is used to correct the initial internal risk level of the second monitoring period according to the correction factor to obtain the final internal risk level of the second monitoring period of each risk observation area.

[0176] In an embodiment of the present application, the risk identification module further includes:

[0177] A determination submodule is used to determine the risk weight of each risk observation area, and determine the risk prediction result of each risk observation area according to the risk weight and comprehensive risk level of each risk observation area;

[0178] The combination submodule is used to combine the risk prediction results of each risk observation area to determine the risk level of the power grid system to be monitored.

[0179] It should be noted that the specific implementation of the artificial intelligence-based power grid risk warning system 200 in the embodiment of the present application refers to the specific implementation of the artificial intelligence-based power grid risk warning method proposed in the first aspect of the aforementioned embodiment of the present application, and will not be repeated here.

[0180] The above embodiments can be implemented in whole or in part by software, hardware (such as circuits), firmware or any other combination. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When a computer instruction or computer program is loaded or executed on a computer, a process or function according to an embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable system. The computer program or instruction can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer program or instruction can be transmitted from one website site, computer, server or data center to another website site, computer, server or data center by wired (such as infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that contains one or more available media sets. The available medium can be a magnetic medium (for example, a floppy disk, a hard disk, a tape), an optical medium (for example, a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state hard disk.

[0181] It should be understood that the term "and / or" in this article is only a description of the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. A and B can be singular or plural. In addition, the character " / " in this article generally indicates that the associated objects before and after are in an "or" relationship, but it may also indicate an "and / or" relationship. Please refer to the context for specific understanding.

[0182] In this application, "at least one" means one or more, and "more" means two or more. "At least one of the following" or similar expressions refers to any combination of these items, including any combination of single or plural items. For example, at least one of a, b, or c can be represented by: a, b, c, ab, ac, bc, or abc, where a, b, c can be single or multiple.

[0183] It should be understood that in the various embodiments of the present application, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0184] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0185] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described systems, systems and units can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0186] In the several embodiments provided in the present application, it should be understood that the disclosed systems, systems and methods can be implemented in other ways. For example, the system embodiments described above are only schematic. For example, the division of units is only a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, and the indirect coupling or communication connection of the system or unit can be electrical, mechanical or other forms.

[0187] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0188] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0189] If the function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the various embodiments of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk, and other media that can store program codes.

[0190] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A power grid risk early warning method based on artificial intelligence, applied to a server, characterized in that: The method comprises: Determine a first type of power access node and a second type of power access node in the power grid to be monitored, wherein the first type of power access node is a power access node with stable output, and the second type of power access node is a power access node with unstable output; Determining an output support capability of the first type of power access node and an output fluctuation state of the second type of power access node; constructing a plurality of risk observation areas according to a matching result of an output support capability of the first type of power access node and an output fluctuation state of the second type of power access node; Determine the comprehensive risk level of each risk observation area, and determine the risk level of the power grid system to be monitored based on the combined results of the comprehensive risk levels of multiple risk observation areas; Determine an early warning strategy that matches the risk level of the power grid system to be monitored, and execute the early warning strategy.

2. The power grid risk early warning method based on artificial intelligence according to claim 1 is characterized in that: The step of determining a first type of power access node and a second type of power access node in the power grid to be monitored comprises: Sending request information to all power access nodes to be identified in the power grid system to be monitored; Receiving an identification data packet fed back by the power access node to be identified after receiving the request information, wherein the identification data packet at least includes historical load data and a type identifier of the power access node to be identified; The type of the power access node to be identified is determined according to the type identifier.

3. The power grid risk early warning method based on artificial intelligence according to claim 1 is characterized in that: The determining the output support capability of the first type power access node and the output fluctuation state of the second type power access node includes: Determining the output support capacity of the first type of power access node according to the load demand curve in the historical load data; The output fluctuation state of the second-type power supply access node is determined according to the output fluctuation curve in the historical load data.

4. The power grid risk early warning method based on artificial intelligence according to claim 1 is characterized in that: The risk observation area includes one first-type power access node and multiple second-type power access nodes, and the multiple risk observation areas are constructed according to the matching result of the output support capability of the first-type power access node and the output fluctuation state of the second-type power access node, including: Determine a first set according to a matching result of an output support capability of the first type power access node and an output fluctuation state of the second type power access node; determining a second set according to a matching result of the location information of the first type of power access node and the location information of the second type of power access node; The risk observation area is determined according to the intersection of the first set and the second set.

5. The artificial intelligence-based power grid risk early warning method according to claim 1 is characterized in that: Determining the comprehensive risk level of each risk observation area includes: Acquire a static image of the first monitoring period of the second type power access node in each of the risk observation areas, and determine the external risk level of the second monitoring period of each of the risk observation areas according to the static image of the first monitoring period; Obtaining load data of a first-type power access node and a second-type power access node in each of the risk observation areas during a first monitoring period, and determining an internal risk level of a second monitoring period of each of the risk observation areas according to the load data of the first monitoring period; The comprehensive risk level of the second monitoring period of each risk observation area is determined according to the external risk level of the second monitoring period of each risk observation area and the internal risk level.

6. The artificial intelligence-based power grid risk early warning method according to claim 5 is characterized in that: The step of determining the external risk level of each risk observation area in the second monitoring period according to the static image of the first monitoring period includes: Determining a first risk assessment score according to a matching result between the behavioral features in the static image and a preset first risk assessment strategy; Determining a second risk assessment score according to a matching result between the environmental features in the static image and a preset second risk assessment strategy; An external risk level of a second monitoring period of each of the risk observation areas is determined according to the first risk assessment score and the second risk assessment score.

7. The power grid risk early warning method based on artificial intelligence according to claim 5 is characterized in that: The step of determining the internal risk level of each risk observation area in the second monitoring period according to the load data in the first monitoring period includes: Constructing input features according to the load data of the first monitoring period, and inputting the input features into a preset risk prediction model to obtain an initial internal risk level of each risk observation area in the second monitoring period; Determine the risk correction factor for the second monitoring cycle based on adjacent risk observation areas; The initial internal risk level of the risk observation area in the second monitoring period is corrected according to the risk correction factor of the second monitoring period to obtain the final internal risk level of each risk observation area in the second monitoring period.

8. The artificial intelligence-based power grid risk early warning method according to claim 1 is characterized in that: Determining the risk level of the to-be-monitored power grid system according to the combined result of the comprehensive risk levels of the plurality of risk observation areas comprises: Determine the risk weight of each risk observation area, and determine the risk prediction result of each risk observation area based on the risk weight and comprehensive risk level of each risk observation area; The risk prediction results of each risk observation area are combined to determine the risk level of the power grid system to be monitored.

9. An artificial intelligence-based power grid risk early warning system, characterized in that: The system comprises: A first determination module, used to determine a first type of power access node and a second type of power access node in the power grid to be monitored, wherein the first type of power access node is a power access node with stable output, and the second type of power access node is a power access node with stable output; A second determination module, configured to determine an output support capability of the first type power access node and an output fluctuation state of the second type power access node; A construction module, configured to construct a plurality of risk observation areas according to a matching result of an output support capability of the first type of power access node and an output fluctuation state of the second type of power access node; A risk identification module, used to determine the comprehensive risk level of each risk observation area, and determine the risk level of the power grid system to be monitored based on the combined results of the comprehensive risk levels of multiple risk observation areas; The execution module is used to determine an early warning strategy that matches the risk level of the power grid system to be monitored and execute the early warning strategy.

10. The artificial intelligence-based power grid risk early warning system according to claim 9, characterized in that: The first determining module includes: A sending submodule, used for sending request information to all power access nodes to be identified in the power grid system to be monitored; A receiving submodule, configured to receive an identification data packet fed back by the power access node to be identified after receiving the request information, wherein the identification data packet at least includes historical load data and a type identifier of the power access node to be identified; The identification submodule is used to determine the type of the power access node to be identified according to the type identifier.

Images