Dynamic programming method and system for power grid normal and emergency states based on Hippo algorithm

Abstract

The present application discloses a method and system for dynamic planning of power grid flat and emergency states based on the Hippo algorithm, which relates to the technical field of power grid emergency control. The method includes: constructing an initial Hippo status influence relationship based on the degree of disaster stability and disaster type according to historical emergency state data, and constructing an initial river position influence relationship based on the disaster source and terrain type; constructing a power grid emergency optimization objective function based on the highest participation complexity and the lowest probability of damage based on the Hippo optimization algorithm architecture; constructing a power grid flat valley optimization objective function based on the state transition relationship and the minimum flat valley fluctuation; obtaining the initial Hippo status and initial river position, and using the power grid emergency optimization objective function and the power grid flat valley optimization objective function to obtain a power grid dynamic planning strategy. The present application can ensure that emergency power supply control can be realized in a timely manner under emergency conditions, thereby improving the timeliness and stability of control.

Description

Technical Field

[0001] The present application relates to the technical field of power grid emergency control, and in particular to a method and system for dynamic planning of power grid emergency states based on the Hippo algorithm. Background Art

[0002] In recent years, the integration of distributed energy sources has helped alleviate carbon emissions from power grids. However, because distributed energy systems are typically located in open areas, they are less resilient to weather disasters and are more likely to be damaged. This leads to unstable supply during unusual weather conditions. Furthermore, during natural disasters, large-scale damage to the power system can lead to power shortages in affected areas. Medical facilities rely on small energy storage devices, hindering rescue efforts and potentially preventing personnel safety.

[0003] Related technologies use quantitative assessments of the resilience of power grids, and by establishing a correlation between the evolution of extreme disasters and the transition of equipment states, the component withdrawal situation in time-varying scenarios under disasters is clarified. However, the affected scope of some disasters cannot be estimated, and each time a disaster occurs, the damage to the components caused by the disaster is simulated, making it impossible to regulate the power grid in a timely manner. The regulation efficiency is low, and emergency regulation is accompanied by rapid switching of power grid equipment, which can easily cause damage to power grid equipment. Rapid switching can also cause power grid imbalance problems, affecting the stability of the power grid.

[0004] The patent application, "Distribution Network Fault Reconfiguration Method Based on an Improved Hippo Optimization Algorithm," with publication number CN118868026A and publication date October 29, 2024, specifically discloses the use of connected components to assess fault recovery, pre-determining loads that can be restored through reconstruction, and reducing the complexity of the problem. Next, encoding is performed using a minimal ring set to map the solution space of the distribution network fault reconstruction problem to the algorithm's search space. This encoding method significantly increases the proportion of valid solutions. The proposed encoding strategy is then used to improve the traditional Hippo optimization algorithm, adapting it to discrete optimization problems. Finally, the improved Hippo optimization algorithm is used to rapidly solve the distribution network fault reconstruction strategy, thereby achieving rapid power restoration. This solution uses the proposed encoding strategy to generate the Hippo optimization algorithm's initial population, improving its solution efficiency and thereby improving its efficiency in solving distribution network fault reconstruction problems. However, this solution only seeks the optimal recovery and reconstruction strategy and cannot address power outages caused by power station damage in emergency situations.

[0005] The patent application, "A Method and System for Energy Storage Capacity Planning Considering Emergency Power Supply Scenarios," has publication number CN119849881A and a publication date of April 18, 2025. The patent application specifically discloses the following: obtaining annual operating data and constructing an energy scheduling optimization model for energy storage in both normal and emergency power supply scenarios based on the annual operating data; establishing an energy storage capacity optimization model for emergency power supply scenarios based on the energy scheduling optimization model, taking into account the configuration parameters of the energy storage and the uncertainty caused by fluctuations in grid operating parameters; and inputting the relevant energy storage parameters into the energy storage capacity optimization model, which then outputs the energy storage capacity planning results. This solution only considers the energy storage device's power supply for emergency scenarios. However, in some disasters, such as strong winds that affect wind power generation, hydropower generation can still provide emergency power supply compensation. Furthermore, this solution still relies on the emergency switching of energy storage devices during emergency power supply situations, which can still lead to grid imbalances. Summary of the Invention

[0006] In response to the problems in the prior art of poor supply stability and easy damage to power stations caused by emergency switching, the present application provides a method and system for dynamic planning of power grid flat and emergency states based on the Hippo algorithm. By constructing an initial Hippo position, the iteration of Hippo optimization calculation is reduced, the efficiency of emergency planning is improved, and the continuous updating of the disaster source position during the optimization process is achieved through the influence relationship of the initial river position. At the same time, the flat valley state is adjusted by utilizing the timing of the transition from the flat valley state to the emergency state, the number of emergency switching times in the emergency state is reduced, and the stability of power station regulation is ensured. The resistance behavior in the Hippo optimization algorithm is utilized to simulate the power station's avoidance of disasters (such as shutdown), etc., thereby improving the accuracy of power station emergency planning.

[0007] To achieve the above-mentioned technical objectives, the present application provides a technical solution, which is a dynamic planning method for the flat and emergency state of a power grid based on the Hippo algorithm, comprising the following steps: constructing an initial Hippo status influence relationship for each power station based on the degree of disaster stability and the type of disaster according to historical emergency state data; constructing an initial river position influence relationship for each disaster source based on the disaster source and the terrain type according to historical emergency state data; constructing a power grid emergency optimization objective function based on the highest participation complexity and the lowest probability of damage based on the Hippo optimization algorithm architecture; obtaining a state transition relationship based on historical emergency state data and historical flat and valley state data, and constructing a power grid flat and valley optimization objective function based on the state transition relationship and the minimum flat and valley fluctuation; calling the initial Hippo status influence relationship and the initial river position influence relationship according to the current regional state data to obtain the initial Hippo status and the initial river position, and using the power grid emergency optimization objective function and the power grid flat and valley optimization objective function to obtain a power grid dynamic planning strategy.

[0008] Furthermore, the construction of the initial hippopotamus status impact relationship for each power station based on the geographical location and the disaster type according to the historical emergency status data includes: extracting the disaster type and the disaster stability of each power station from the historical emergency status data; constructing the temporal hierarchical relationship between each power station based on the disaster temporal hierarchy; using the disaster stability as the initial hippopotamus status judgment standard, establishing the correlation mapping between the initial hippopotamus status and the disaster type and temporal hierarchy relationship, and obtaining the initial hippopotamus status impact relationship of each power station.

[0009] Furthermore, the construction of the initial river position impact relationship for each disaster source based on the disaster source and the terrain type according to the historical emergency status data includes: obtaining the movement path and impact range of the disaster source from the historical emergency status data; using the movement path of the disaster source as the river path, establishing an association mapping between the initial river position and the impact range and terrain type, and obtaining the initial river position impact relationship of each disaster source.

[0010] Furthermore, the power grid emergency optimization objective function is constructed based on the Hippo optimization algorithm architecture according to the highest participation complexity and the lowest damage probability, including: calculating the participation complexity according to the number of power station types, and calculating the damage probability according to the entropy increase of the power station affected by the disaster.

[0011] Furthermore, the state transition relationship is obtained based on historical emergency state data and historical flat valley state data, and the grid flat valley optimization objective function is constructed based on the state transition relationship and the minimum flat valley fluctuation, including: obtaining historical flat valley state data of the disaster potential time segment according to the historical emergency state data, extracting conversion dimensions according to the historical flat valley state data of the disaster potential time segment, and constructing a state transition relationship based on the conversion dimensions and the disaster potential time segment.

[0012] Furthermore, constructing the grid valley optimization objective function based on the state transition relationship and the minimum valley fluctuation includes: calculating the valley fluctuation based on the state transition relationship, grid load fluctuation and power supply range fluctuation.

[0013] Furthermore, the initial hippopotamus status influence relationship and the initial river position influence relationship are called according to the current regional status data to obtain the initial hippopotamus status and the initial river position, and the power grid emergency optimization objective function and the power grid flat valley optimization objective function are used to obtain the power grid dynamic planning strategy, including: obtaining the predicted disaster potential time sequence and the predicted disaster source according to the current regional status data and the state conversion relationship; obtaining the initial hippopotamus status and the initial river position according to the predicted disaster source, the initial hippopotamus status influence relationship and the initial river position influence relationship; obtaining the power grid emergency planning strategy according to the power grid emergency optimization objective function based on the initial hippopotamus status and the initial river position; obtaining the power grid flat valley planning strategy according to the predicted disaster potential time sequence, the power grid emergency planning strategy and the power grid flat valley optimization objective function.

[0014] Furthermore, the disaster stability level is used as the criterion for judging the initial hippo status, and a correlation mapping between the initial hippo status and the disaster type and the temporal hierarchical relationship is established to obtain the initial hippo status influence relationship of each power station, including: obtaining the disaster stability level of all power stations under the same disaster type based on the temporal hierarchical relationship, and performing cluster analysis based on the disaster stability level to obtain outliers and cluster centers, taking the power stations corresponding to the outliers as small hippos, taking the power stations corresponding to the highest disaster stability level as hippo leaders, taking the power stations corresponding to the cluster centers other than the hippo leaders as male hippos, and taking the power stations corresponding to the cluster points surrounding the cluster centers as female hippos.

[0015] Furthermore, the performing of cluster analysis based on the disaster stability to obtain outliers and cluster centers also includes: performing a clustering of all power stations based on environmental similarity and power station type, and performing a secondary clustering of the first clustering results based on the disaster stability to obtain outliers and cluster centers of each clustering result.

[0016] Furthermore, the clustering of all power stations based on environmental similarity and power station type includes: calculating the longitude distance of the power station and the latitude distance of the power station based on the distance algorithm; obtaining the longitude distance weight of the power station, the latitude distance weight of the power station and the latitude weight of the regional type of the power station interval according to the disaster type and the power station type; obtaining environmental similarity according to the longitude distance weight of the power station, the latitude distance weight of the power station, the latitude weight of the regional type of the power station interval, the longitude distance of the power station, the latitude distance of the power station and the regional type of the power station interval.

[0017] Another technical solution provided by this application is a power grid normal and emergency state dynamic planning system based on the Hippo algorithm, which is used to implement the above method.

[0018] Another technical solution provided by the present application is that a computer program or instruction is stored in the storage medium, and when the computer program or instruction is executed by a processing device, the above method is implemented.

[0019] The beneficial effects of the present application are as follows: the initial hippopotamus status influence relationship and the initial river position influence relationship are constructed through historical emergency status data, thereby reflecting the differences in the stability of power stations affected by different disaster types and the movement or impact of disaster sources under different terrain types, and then considering the overall power station's participation complexity during the disaster and the impact of the probability of damage on the stability of emergency power supply, the grid emergency optimization objective function is constructed, and the position update, resistance behavior, and escape behavior of the hippopotamus relative to the river in the hippopotamus optimization algorithm are used to simulate the relative position update between the power station and the disaster source, the defense behavior of the power station, and the shutdown behavior of the power station under disaster conditions, thereby improving the simulation accuracy of the impact on the actual power station behavior, and at the same time incorporating the consideration of the timing adjustment of the transition from the flat valley state to the emergency state. According to the state conversion timing and the regulation state of the demand in the emergency state, the flat valley stage adjustment is performed with the minimum flat valley fluctuation of the power grid as the goal, avoiding the problem of grid imbalance caused by too fast access, ensuring that emergency power supply regulation can be realized in time under emergency conditions, and improving the timeliness and stability of regulation. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 This is a flow chart of the dynamic planning method for power grid normal and emergency states based on the Hippo algorithm in this application.

[0021] Figure 2 This is a schematic diagram of the initial hippopotamus position in an embodiment of the present application. DETAILED DESCRIPTION

[0022] In order to make the purpose, technical solutions and advantages of this application more clear, the application is further described in detail below with reference to the accompanying drawings and examples. It should be understood that the specific implementation method described here is only an optimal embodiment of this application, which is only used to explain this application and does not limit the scope of protection of this application. All other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0023] like Figure 1 As shown in the embodiment 1 of the present application, a method for dynamic planning of power grid normal and emergency states based on the Hippo algorithm includes the following steps:

[0024] Based on historical emergency status data, the initial hippo status impact relationship for each power station is constructed based on the degree of disaster stability and disaster type;

[0025] Based on historical emergency data, the initial river location impact relationship for each disaster source is constructed based on the disaster source and terrain type;

[0026] Based on the Hippo optimization algorithm architecture, the power grid emergency optimization objective function is constructed according to the highest participation complexity and the lowest damage probability;

[0027] Based on historical emergency state data and historical valley state data, the state transition relationship is obtained, and the grid valley optimization objective function is constructed based on the state transition relationship and the minimum valley fluctuation.

[0028] According to the current regional status data, the initial hippopotamus status influence relationship and the initial river position influence relationship are called to obtain the initial hippopotamus status and the initial river position, and the power grid emergency optimization objective function and the power grid flat valley optimization objective function are used to obtain the power grid dynamic planning strategy.

[0029] In this embodiment, the initial hippopotamus status influence relationship and the initial river position influence relationship are constructed through historical emergency status data, thereby reflecting the differences in the stability of power stations affected by different disaster types and the movement or impact of disaster sources under different terrain types. Then, the participation complexity of the entire power station during the disaster and the influence of the probability of damage on the stability of emergency power supply are considered to construct the power grid emergency optimization objective function. The position update, resistance behavior, and escape behavior of the hippopotamus relative to the river in the hippopotamus optimization algorithm are used to simulate the relative position update between the power station and the disaster source, the defense behavior of the power station, and the shutdown behavior of the power station under the disaster situation, so as to improve the simulation accuracy of the impact of the actual power station behavior. At the same time, the timing adjustment of the transition from the flat valley state to the emergency state is taken into consideration. According to the state conversion timing and the regulation state of the demand in the emergency state, the flat valley stage adjustment is performed with the minimum flat valley fluctuation of the power grid as the goal to avoid the problem of grid imbalance caused by too fast access, ensure that the emergency power supply regulation can be realized in time under the emergency state, and improve the timeliness and stability of the regulation.

[0030] Specifically, based on historical emergency status data, the initial Hippo status impact relationship for each power station is constructed based on the degree of disaster stability and disaster type, including:

[0031] Extract the disaster type and disaster stability of each power station from historical emergency status data;

[0032] Construct the temporal hierarchical relationship between each power station based on the disaster-affected temporal hierarchy;

[0033] Taking the degree of disaster stability as the criterion for judging the initial hippo status, a correlation mapping between the initial hippo status and the disaster type and temporal hierarchy is established to obtain the impact relationship of the initial hippo status of each power station.

[0034] By obtaining the disaster stability of each power station when an emergency occurs in its own area, the ability of each power station to resist different disaster types is obtained, and the difference in disaster degree between different power stations is constructed based on the difference in disaster time sequence. It is visualized through a hierarchical model, and the initial hippopotamus status is allocated according to the behavior of the power station in different time sequences and different disaster types.

[0035] Specifically, in the Hippopotamus optimization algorithm, a hippopotamus group consists of female hippos, baby hippos, male hippos, and a hippopotamus leader. Baby hippos often show a tendency to deviate from the group, while female hippos are distributed around male hippos. Female hippos update their positions while exploring new areas while maintaining a certain distance from the dominant solution. After being expelled by the hippopotamus leader, male hippos need to establish a new group by competing or attracting females, thereby balancing global search and local search. The hippopotamus leader is the individual with the best median value in the current iteration. Figure 2 As shown in the figure, the degree of disaster stability is used as the criterion for judging the initial hippopotamus status, and a correlation mapping between the initial hippopotamus status and the disaster type and time series hierarchy is established. The influence relationship of the initial hippopotamus status of each power station is obtained, including:

[0036] Based on the temporal hierarchical relationship, the disaster stability of all power stations under the same disaster type is obtained, and cluster analysis is performed based on the disaster stability to obtain outliers and cluster centers. The power stations corresponding to the outliers are regarded as small hippos, the power stations corresponding to the highest disaster stability are regarded as hippo leaders, the power stations corresponding to the cluster centers other than the hippo leaders are regarded as male hippos, and the power stations corresponding to the cluster points surrounding the cluster centers are regarded as female hippos.

[0037] In this embodiment, cluster analysis is performed based on the degree of disaster stability of power stations under different time sequences and different disaster types. Different cluster centers are obtained based on the clustering results of different disaster stability levels. The cluster points are arranged around the cluster center to simulate the pattern of female hippos surrounding male hippos. The power station with the highest degree of disaster stability is defined as the initial optimal solution and is the initial hippo leader. Power stations that deviate from the cluster center, that is, power stations whose disaster stability levels do not match those of the other power stations, are defined as small hippos, thereby demonstrating the state of the small hippos being out of the group. By defining the initial hippo status of power stations based on their degree of disaster stability, the search dimension in the iterative optimization process is increased, and the exploration time of the algorithm iteration can be reduced by using the initial hippo status as the initial solution. The dynamic simulation of power station state changes using hippo behavior improves the accuracy and efficiency of power station emergency planning.

[0038] In other embodiments, the power stations may be clustered first according to environmental similarity and / or power station type, and then the different states of each power station may be further refined according to the clustering result and the degree of stability of the disaster.

[0039] In some cases, performing cluster analysis based on the degree of disaster stability to obtain outliers and cluster centers also includes:

[0040] A clustering is performed on all power stations based on environmental similarity, and a secondary clustering is performed on the primary clustering results based on the degree of disaster stability to obtain the outliers and cluster centers of each clustering result.

[0041] In this embodiment, the environmental similarity between the power stations is used to cluster the power stations once, and then a secondary clustering is performed on the clustering analysis results after the first clustering to obtain the clustering situation corresponding to the disaster stability of the power stations under similar environments, thereby incorporating the environmental dimension into the initial definition of the hippopotamus status. When the disaster stability under similar environments deviates from the large group, it is considered that the power station is far away from the group.

[0042] In other cases, cluster analysis based on the degree of disaster stability is performed to obtain outliers and cluster centers, including:

[0043] A primary clustering is performed on all power plants based on their types, and a secondary clustering is performed on the primary clustering results based on the degree of stability of the disaster to obtain the outliers and cluster centers of each clustering result.

[0044] In this case, all power plants are clustered once by power plant type, and then a second clustering is performed on the set of each power plant type, so as to show the different response states of different power plants to different disaster types, avoid the deviation of outlier selection due to the different behaviors of different types of power plants, further improve the accuracy of the initial definition of hippopotamus status, reduce the number of subsequent optimization iterations, and improve the accuracy of emergency planning strategy output.

[0045] In some other cases, performing cluster analysis based on the disaster stability to obtain outliers and cluster centers further includes:

[0046] A clustering is performed on all power plants based on environmental similarity and power plant type, and a secondary clustering is performed on the primary clustering results based on the disaster stability to obtain the outliers and cluster centers of each clustering result.

[0047] The environmental similarity is calculated based on the longitude and latitude of the power station and the type of area between power stations:

[0048] ;

[0049] ;

[0050] ;

[0051] ;

[0052] ;

[0053] in, Indicates environmental similarity; Represents the similarity weight of longitude and latitude; Indicates the interval area type weight; Indicates the longitude and latitude distance of the power station; Indicates the distance between area types; represents the relative position relationship between power station i and power station j; represents the difference in latitude arc between station i and station j, ; represents the latitude of station i in radians; represents the latitude of station j in radians; represents the difference in longitude arc between station i and station j, , represents the arc of longitude of station j; represents the arc of longitude of station i; represents the radius of the Earth; It indicates that the terrain of the g-th interval area is high; Indicates that the terrain of power station i is high; G indicates the total number of interval areas.

[0054] The degree of disaster stability is calculated based on the disaster-affected ratio, average recovery time, and power supply:

[0055] ;

[0056] ;

[0057] ;

[0058] ;

[0059] Among them, C represents the stability of the disaster; represents the weight of the disaster-affected proportion, represents the average recovery time weight, Indicates the power supply weight; Indicates the proportion of disasters; represents the mean recovery time, It represents the recovery time after the nth disaster, and N represents the total number of disaster recovery times; Indicates the power supply.

[0060] 、 、 、 、 It can be obtained through expert experience or by performing hierarchical analysis based on historical data. =0.6, =0.4, =0.5, =0.2, =0.3.

[0061] All power stations are clustered based on environmental similarity and power station type, and the power stations are initially divided according to their location and type to improve the sensitivity of outlier detection. The behavior of female hippos surrounding male hippos is used to simulate the synergistic effect of power stations of the same type, thereby improving the exploration efficiency and accuracy in the optimization calculation process.

[0062] In some other embodiments, calculating the environmental similarity based on the longitude and latitude of the power stations and the type of the area between the power stations further includes:

[0063] Calculate the longitude distance and latitude distance of the power station based on the distance algorithm;

[0064] Obtain the power station longitude distance weight, power station latitude distance weight, and power station interval area type latitude weight based on the disaster type and power station type;

[0065] The environmental similarity is obtained based on the power station longitude distance weight, the power station latitude distance weight, the power station interval area type latitude weight, the power station longitude distance, the power station latitude distance and the power station interval area type.

[0066] In this embodiment, the environmental similarity is calculated specifically using the differences in longitude and latitude. For example, in terms of climate, regions at different latitudes receive different amounts of solar radiation, and regions at different longitudes are susceptible to atmospheric circulation and other factors, resulting in humidity differences. Distributed energy resources are affected differently, such as photovoltaic power stations, which are more susceptible to longitude differences, and wind power stations, which are more susceptible to atmospheric circulation factors. Different disaster types are also affected by differences in longitude and latitude. Different disaster types are weakened or enhanced to varying degrees by the type of intervening region. For example, plains have a weaker ability to weaken typhoons than hilly regions. Therefore, the longitude distance weight of the power station, the latitude distance weight of the power station, and the latitude weight of the intervening region type are calculated based on the power station type and the disaster type. For example, the degree of disaster weakening in historical disaster data can be used to determine the degree of weakening or enhancing different disasters in different intervening region types. Based on the differences in power supply of different power station types at different longitudes and latitudes, the longitude distance weight of the power station, the latitude distance weight of the power station, and the latitude weight of the intervening region type corresponding to the disaster type and power station type are calculated. Through targeted calculations, the environmental similarity is made more consistent with the actual situation, reducing the impact of the overall environment on the local environment and improving the adaptability of the results.

[0067] Based on historical emergency data, the initial river location impact relationship for each disaster source is constructed based on the disaster source and terrain type, including:

[0068] Obtain the movement path and impact range of disaster sources from historical emergency status data;

[0069] Taking the movement path of the disaster source as the river path, the correlation mapping between the initial river position and the impact range and terrain type is established to obtain the initial river position impact relationship of each disaster source.

[0070] In this embodiment, the movement path of the disaster source is simulated as a river path. Based on the river path, an association mapping is established with the affected range and terrain type. In this process, the terrain type can be hot-coded to facilitate digital processing. In the hippo optimization algorithm, the river attributes are mainly manifested as attraction attributes and resistance attributes. In this embodiment, the movement of the hippopotamus to the river position is converted into a relative position update. The situation where the disaster source is blocked by the terrain is mapped to the river's resistance attribute, and the situation where the power station is affected by the disaster source is mapped to the river's attraction attribute. This allows the disaster source to update its position based on the pre-acquired movement path. At each update time sequence, the movement distance and direction of the river are adjusted according to the direction and distance of the movement path and the "resistance" brought by the terrain type to achieve prediction of the movement of the disaster source. After the river position is updated, the relative position of each hippopotamus to the river is immediately calculated, and the initial hippo position and initial river position are obtained through the initial hippo position influence relationship and the initial river position influence relationship, reducing the number of iterative calculations and improving calculation efficiency.

[0071] Based on the Hippo optimization algorithm architecture, the power grid emergency optimization objective function is constructed according to the highest participation complexity and the lowest damage probability, including:

[0072] The participation complexity is calculated according to the number of power station types, and the damage probability is calculated according to the entropy increase of the power station affected by the disaster.

[0073] Different types of power plants differ in their operating principles, equipment configurations, and disaster resilience. When more types of power plants are present in a grid, they are less likely to be forced to shut down simultaneously due to the same disaster, owing to differences in disaster response mechanisms and equipment tolerance. This reduces the probability of synchronized shutdowns among power plants. The greater the number of different power plant types, the lower the probability of synchronized shutdowns, thus strengthening the overall resilience. The entropy increase of power plants affected by disasters is used to calculate the probability of damage. This is calculated by combining the entropy increase of power plant components due to the impact of their mitigation actions and the impact of the disaster source on the entropy increase of power plant components.

[0074] The calculation of the damage probability based on the entropy increase of the power station includes:

[0075] Obtain historical action data of the power plant, and calculate the entropy increase impact relationship of components corresponding to the power plant conversion action based on the historical action data of the power plant;

[0076] Obtain historical environmental data of the power plant and calculate the environmental entropy increase impact relationship based on the historical environmental data of the power plant;

[0077] The damage probability is calculated based on the relationship between component entropy increase and the relationship between environmental entropy increase.

[0078] During the iterative calculation process of the Hippo optimization algorithm, the environmental impact caused by the relative position of each power station and the disaster source is calculated. Based on the environmental impact, the environmental entropy increase influence relationship and the component entropy increase influence relationship are calculated to obtain the damage probability of the current power station.

[0079] The objective function of power grid emergency optimization is:

[0080] ;

[0081] ;

[0082] in, represents the grid emergency optimization objective function, Indicates the number of power station types. represents the probability of damage, Indicates the weight of the number of power station types, represents the damage probability weight.

[0083] The power plant type number weight and the damage probability weight can be calculated through expert experience or historical data. In some cases, the power plant type number weight and the damage probability weight are divided into two cases. When in a conservative situation, the weight of =0.3, =0.7, when in the aggressive strategy, that is, ensuring power supply as much as possible =0.6, =0.4.

[0084] The state transition relationship is obtained based on historical emergency state data and historical valley state data. The grid valley optimization objective function is constructed based on the state transition relationship and the minimum valley fluctuation. The function includes:

[0085] According to the historical emergency status data, the historical flat and valley state data of the disaster potential time series are obtained, the conversion dimension is extracted according to the historical flat and valley state data of the disaster potential time series, and the state conversion relationship is constructed based on the conversion dimension and the disaster potential time series.

[0086] In this embodiment, historical flat valley state data for the disaster potential time series segment is obtained based on historical emergency state data. For example, when a gradually occurring disaster such as a flood or drought occurs, there must be a time series that is not in an emergency state, but its environmental characteristics are consistent with those of the emergency state. For example, before the occurrence of a flood emergency, there must be a flat valley time series of rainfall. This time series is used as the disaster potential time series segment to obtain the corresponding flat valley state data. Thus, based on the historical flat valley state data, conversion dimensions are extracted. For example, in this embodiment, the conversion dimensions are rainfall, river level depth, humidity, wind speed, ice thickness, temperature, and the corresponding maintenance time. A mapping relationship between the conversion dimensions, the disaster potential time series segment, and the disaster source is constructed, that is, the state conversion relationship. At this time, the disaster potential time series and disaster source matching the current dimension data are retrieved according to the state conversion relationship, and used as the target time series for the current Pinggu state adjustment and the disaster source for emergency adjustment. For example, when the average rainfall in a certain area is 30mm, lasting for 6 hours, and the depth of a certain horizontal surface exceeds the warning depth, it is considered that there will be a flood disaster. The time when the flood disaster may occur is calculated based on the current data, and the location where the horizontal surface depth exceeds the warning depth is used as the disaster source, and pre-planning of emergency state control and pre-control warning of Pinggu state are carried out.

[0087] The grid valley optimization objective function based on the state transition relationship and minimum valley fluctuation is constructed as follows:

[0088] The flat-valley fluctuations are calculated based on the state transition relationship, grid load fluctuations and power supply range fluctuations.

[0089] The target timing for adjusting the current flat and valley state is obtained based on the state transition relationship. The grid load fluctuation and power supply range fluctuation are calculated based on the target timing and emergency planning strategy to obtain the optimal adjustment strategy under the flat and valley state.

[0090] The objective function of grid valley optimization is:

[0091] ;

[0092] in, represents the grid flat valley optimization objective function, Q represents the total number of regulated power stations output by the grid emergency planning strategy, Indicates that the target timing is Load fluctuation caused by the adjustment of power station q, Indicates that the target timing is The range fluctuations caused by the adjustment of the power station q.

[0093] In this embodiment, range fluctuation refers to the fluctuation in the power supply range during the power station adjustment process. When the power station is adjusted in the valley state toward the state output of the emergency planning strategy, both the load fluctuation of the power station adjustment and the power supply range problem during the power station adjustment process are considered. Under normal conditions, the power supply range of each power station usually has a certain stability. However, in the event of an emergency, the power station adjustment not only involves adjustments to the power station power and shutdown, but also involves the use of stable power stations for emergency power supply. In this case, the power supply range also needs to be adjusted accordingly. During the transition from the valley state to the emergency state, there is no guarantee that an emergency state will occur. Therefore, large-scale power supply range adjustments will actually reduce the user experience. In particular, users have a stronger demand for stable power supply in abnormal weather such as extreme cold or extreme heat. Therefore, the optimal adjustment strategy in the valley state is output based on the two-way adjustment of load fluctuation and power supply range. This allows the valley adjustment to reduce the load fluctuation caused by the emergency switching of the emergency adjustment and to ensure normal power consumption in the valley state as much as possible. In some cases, the importance of load fluctuation and power supply range can also be adjusted by using weight coefficients. For example, when power station equipment is susceptible to damage due to emergency switching and there are a large number of flexible loads in the current power supply area that can be used as an adjustment range, the weight coefficient of load fluctuation is greater than the weight coefficient of power supply range. Of course, different weight coefficients can also be constructed for each power station equipment, and then a comprehensive calculation is performed to improve accuracy.

[0094] According to the current regional status data, the initial hippopotamus status influence relationship and the initial river position influence relationship are called to obtain the initial hippopotamus status and the initial river position, and the power grid emergency optimization objective function and the power grid flat valley optimization objective function are used to obtain the power grid dynamic planning strategy including:

[0095] Obtain the predicted disaster potential time series and predicted disaster source based on the current regional status data and state conversion relationship;

[0096] Obtain the initial hippopotamus status and initial river position according to the predicted disaster source, the initial hippopotamus status impact relationship and the initial river position impact relationship;

[0097] Obtaining a power grid emergency planning strategy based on the initial hippopotamus status and the initial river position according to the power grid emergency optimization objective function;

[0098] The grid valley planning strategy is obtained based on the predicted disaster potential time series, grid emergency planning strategy and grid valley optimization objective function.

[0099] In this embodiment, the state data includes at least environmental data, topographic data, and power plant operation data. The power plant operation data includes at least power supply, component damage data, and power supply topology. The predicted disaster potential time series and predicted disaster source are retrieved based on the matching results between the current regional state data and the state transition relationship. The initial hippopotamus position and initial river position are obtained based on the matching results between the predicted disaster source and the initial hippopotamus position influence relationship and the initial river position influence relationship.

[0100] Based on the initial hippopotamus position and the initial river position, the hippopotamus optimization iterative calculation according to the power grid emergency optimization objective function includes the following steps:

[0101] Initialize the hippo population and assign an initial position to each hippopotamus based on the initial river position and the power station location;

[0102] Calculate the fitness value of each hippopotamus according to the power grid emergency optimization objective function and the initial hippopotamus status, and update the hippopotamus status;

[0103] When hippopotamus individuals enter the vicinity of a disaster source, escape behavior is triggered (to avoid power outages at power stations);

[0104] When there is no significant change in the Pareto frontier for multiple consecutive generations, or the maximum number of iterations Tmax is reached, the iteration is terminated and the Pareto optimal solution set is output.

[0105] Among them, the Pareto optimal solution set is the power grid emergency planning strategy. In this embodiment, the power supply constraint and the power supply range constraint are constructed for the power grid emergency optimization objective function. The power supply constraint is obtained based on the minimum power demand in the current area, and the power supply range constraint is obtained based on the current disaster prevention area. That is, the power grid emergency planning needs to meet the minimum power demand in the current area and at least meet the power supply demand in the disaster prevention area.

[0106] After outputting a grid emergency planning strategy based on the predicted disaster source, the predicted disaster potential time series is obtained based on the predicted disaster source. Using the grid valley optimization objective function, a valley planning strategy is output that, given the predicted disaster potential time series, ensures transition to the grid emergency planning strategy, minimizes load shedding fluctuations, and minimizes the impact on the current valley power supply. Power plant adjustments are made before the emergency state is entered to avoid grid fluctuations caused by emergency shedding. Simultaneously, the predicted river positions are continuously updated using the initial river position influence relationship, and the corresponding grid emergency planning strategy is output to accommodate disaster uncertainty. It is understood that once the current emergency state is entered, the grid valley optimization objective function is no longer invoked for dynamic valley state optimization. Instead, all computing resources are focused on emergency optimization calculations. The initial hippo position influence relationship and the initial river position influence relationship are utilized to reduce the computational difficulty and random inaccuracies of the initial solution, further improving planning efficiency.

[0107] As a second embodiment of the present application, a power grid normal and emergency state dynamic planning system based on the Hippo algorithm includes:

[0108] A data acquisition unit is used to obtain status data, wherein the status data at least includes historical emergency status data, historical flat valley status data and current regional status data;

[0109] An initial solution calculation unit is used to construct an initial hippopotamus status impact relationship and an initial river position impact relationship based on the disaster stability, disaster type, disaster source, and terrain type according to historical emergency status data;

[0110] An optimization construction unit, used to construct a power grid emergency optimization objective function and a power grid flat-valley optimization function based on the Hippo optimization algorithm architecture;

[0111] The dynamic programming unit is used to output the power grid dynamic programming strategy according to the initial hippopotamus status influence relationship, the initial river position influence relationship, the power grid emergency optimization objective function and the power grid flat valley optimization function.

[0112] In this embodiment, the data acquisition unit is connected to the initial solution calculation unit, the optimization construction unit and the dynamic planning unit. The initial solution calculation unit and the optimization construction unit are connected to the dynamic planning unit. The dynamic planning unit receives the current regional status data from the data acquisition unit, calls the hippopotamus status influence relationship and the initial river status influence relationship to obtain the initial hippopotamus status and the initial river status, and calculates the power grid emergency planning strategy according to the power grid emergency optimization objective function. The power grid emergency planning strategy and the power grid flat valley optimization function are used to predict the flat valley state and output the flat valley planning strategy under the disaster potential time sequence, thereby realizing pre-control and calculating pressure differentiation, and improving the accuracy and efficiency of emergency planning.

[0113] A data acquisition unit is used to obtain status data, wherein the status data at least includes historical emergency status data, historical flat valley status data and current regional status data;

[0114] An initial solution calculation unit is used to construct an initial hippopotamus status impact relationship and an initial river position impact relationship based on the disaster stability, disaster type, disaster source, and terrain type according to historical emergency status data;

[0115] An optimization construction unit, used to construct a power grid emergency optimization objective function and a power grid flat-valley optimization function based on the Hippo optimization algorithm architecture;

[0116] The dynamic programming unit is used to output the power grid dynamic programming strategy according to the initial hippopotamus status influence relationship, the initial river position influence relationship, the power grid emergency optimization objective function and the power grid flat valley optimization function.

[0117] As a third embodiment of the present application, a computer-readable storage medium is used to store a computer program or instruction. When the computer program or instruction is executed by a processing device, the above-mentioned dynamic planning method for the emergency state of the power grid based on the Hippo algorithm is implemented. The computer-readable storage medium can be any available medium that can be stored by a computing device or a data storage device such as a data center containing one or more available media. 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 (for example, a solid-state drive), etc.

[0118] The specific implementation method described above is a preferred implementation method of the power grid emergency state dynamic planning method and system based on the Hippo algorithm of this application, and is not intended to limit the specific implementation scope of this application. The scope of this application includes but is not limited to this specific implementation method. Any equivalent changes made in accordance with the shape and structure of this application are within the scope of protection of this application.

Claims

1. A dynamic programming method for power grid normal and emergency states based on the Hippo algorithm, characterized by: The steps include: Based on historical emergency status data, the initial hippo status impact relationship for each power station is constructed based on the degree of disaster stability and disaster type; Based on historical emergency data, the initial river location impact relationship for each disaster source is constructed based on the disaster source and terrain type; Based on the Hippo optimization algorithm architecture, the power grid emergency optimization objective function is constructed according to the highest participation complexity and the lowest damage probability; Based on historical emergency state data and historical valley state data, the state transition relationship is obtained, and the grid valley optimization objective function is constructed based on the state transition relationship and the minimum valley fluctuation. According to the current regional status data, the initial hippopotamus status influence relationship and the initial river position influence relationship are called to obtain the initial hippopotamus status and the initial river position, and the power grid emergency optimization objective function and the power grid flat valley optimization objective function are used to obtain the power grid dynamic planning strategy; Among them, the degree of stability of the disaster was used as the criterion for judging the initial hippo status, and the movement path of the disaster source was used as the river path.

2. The method for dynamic programming of power grid normal and emergency states based on the Hippo algorithm according to claim 1, characterized in that: The initial hippopotamus status impact relationship for each power station is constructed based on historical emergency status data, geographical location, and disaster type, including: Extract the disaster type and disaster stability of each power station from historical emergency status data; Construct the temporal hierarchical relationship between each power station based on the disaster-affected temporal hierarchy; Taking the degree of disaster stability as the criterion for judging the initial hippo status, a correlation mapping between the initial hippo status and the disaster type and temporal hierarchy is established to obtain the impact relationship of the initial hippo status of each power station.

3. The method for dynamic programming of power grid normal and emergency states based on the Hippo algorithm according to claim 1, characterized in that: The initial river position influence relationship for each disaster source is constructed based on the disaster source and terrain type according to the historical emergency status data, including: Obtain the movement path and impact range of disaster sources from historical emergency status data; Taking the movement path of the disaster source as the river path, the correlation mapping between the initial river position and the impact range and terrain type is established to obtain the initial river position impact relationship of each disaster source.

4. The method for dynamic planning of power grid normal and emergency states based on the Hippo algorithm according to claim 1, characterized in that: The power grid emergency optimization objective function constructed based on the Hippo optimization algorithm architecture according to the highest participation complexity and the lowest damage probability includes: The participation complexity is calculated according to the number of power station types, and the damage probability is calculated according to the entropy increase of the power station affected by the disaster.

5. The method for dynamic planning of power grid normal and emergency states based on the Hippo algorithm according to claim 1, characterized in that: The state conversion relationship is obtained based on historical emergency state data and historical valley state data, and the grid valley optimization objective function is constructed based on the state conversion relationship and the minimum valley fluctuation. According to the historical emergency status data, the historical flat and valley state data of the disaster potential time series are obtained, the conversion dimension is extracted according to the historical flat and valley state data of the disaster potential time series, and the state conversion relationship is constructed based on the conversion dimension and the disaster potential time series.

6. The method for dynamic planning of power grid normal and emergency states based on the Hippo algorithm according to claim 1, characterized in that: The grid flat valley optimization objective function constructed based on the state transition relationship and the minimum flat valley fluctuation includes: The flat-valley fluctuations are calculated based on the state transition relationship, grid load fluctuations and power supply range fluctuations.

7. The method for dynamic planning of power grid normal and emergency states based on the Hippo algorithm according to claim 1, characterized in that: The method of calling the initial hippopotamus status influence relationship and the initial river position influence relationship according to the current regional status data to obtain the initial hippopotamus status and the initial river position, and using the power grid emergency optimization objective function and the power grid flat valley optimization objective function to obtain the power grid dynamic programming strategy includes: Obtain the predicted disaster potential time series and predicted disaster source based on the current regional status data and state conversion relationship; Obtain the initial hippopotamus position and the initial river position according to the predicted disaster source, the initial hippopotamus position impact relationship and the initial river position impact relationship; Obtaining a power grid emergency planning strategy based on the initial hippopotamus status and the initial river position according to the power grid emergency optimization objective function; The grid valley planning strategy is obtained based on the predicted disaster potential time series, grid emergency planning strategy and grid valley optimization objective function.

8. The method for dynamic programming of power grid normal and emergency states based on the Hippo algorithm according to claim 2, characterized in that: The above-mentioned initial hippopotamus status evaluation criteria are based on the degree of disaster stability, and a correlation mapping between the initial hippopotamus status, disaster type, and time series hierarchy is established to obtain the initial hippopotamus status impact relationship of each power station, including: Based on the temporal hierarchical relationship, the disaster stability of all power stations under the same disaster type is obtained, and cluster analysis is performed based on the disaster stability to obtain outliers and cluster centers. The power stations corresponding to the outliers are regarded as small hippos, the power stations corresponding to the highest disaster stability are regarded as hippo leaders, the power stations corresponding to the cluster centers other than the hippo leaders are regarded as male hippos, and the power stations corresponding to the cluster points surrounding the cluster centers are regarded as female hippos.

9. The method for dynamic planning of power grid normal and emergency states based on the Hippo algorithm according to claim 8, characterized in that: The performing cluster analysis based on the disaster stability to obtain outliers and cluster centers also includes: A clustering is performed on all power plants based on environmental similarity and power plant type, and a secondary clustering is performed on the primary clustering results based on the disaster stability to obtain the outliers and cluster centers of each clustering result.

10. The method for dynamic planning of power grid normal and emergency states based on the Hippo algorithm according to claim 9, characterized in that: The clustering of all power stations based on environmental similarity and power station type includes: Calculate the longitude distance and latitude distance of the power station based on the distance algorithm; Obtain the power station longitude distance weight, power station latitude distance weight, and power station interval area type latitude weight based on the disaster type and power station type; The environmental similarity is obtained based on the power station longitude distance weight, the power station latitude distance weight, the power station interval area type latitude weight, the power station longitude distance, the power station latitude distance and the power station interval area type.

11. A power grid normal and emergency state dynamic planning system based on the Hippo algorithm, used to implement the method according to any one of claims 1 to 10, characterized in that: A data acquisition unit is used to obtain status data, wherein the status data at least includes historical emergency status data, historical flat valley status data and current regional status data; An initial solution calculation unit is used to construct an initial hippopotamus status impact relationship and an initial river position impact relationship based on the disaster stability, disaster type, disaster source, and terrain type according to historical emergency status data; An optimization construction unit, used to construct a power grid emergency optimization objective function and a power grid flat-valley optimization function based on the Hippo optimization algorithm architecture; The dynamic programming unit is used to output the power grid dynamic programming strategy according to the initial hippopotamus status influence relationship, the initial river position influence relationship, the power grid emergency optimization objective function and the power grid flat valley optimization function.

12. A computer-readable storage medium, characterized in that: The storage medium stores a computer program or instruction. When the computer program or instruction is executed by a processing device, the method according to any one of claims 1 to 10 is implemented.

Images