An electric load prediction method under extreme weather

Abstract

The application discloses an electric load prediction method under extreme weather, and belongs to the technical field of electric load prediction. The method comprises the following steps: determining the electric power transmission sources in the influence range of extreme weather, analyzing whether renewable energy power generation is contained in the electric power transmission sources, and evaluating whether the extreme weather will cause an influence on the power generation facilities to predict total power generation power; and dividing the range influenced by the extreme weather into multiple different regions according to the connection relationship of the power grid. The application can improve the prediction accuracy of power generation, the risk resistance of the power grid and the power distribution efficiency by combining extreme weather prediction and geographic information to predict total power generation power, dividing regions according to the connection relationship of the power grid, combining conventional and post-disaster reconstruction load to predict total electric load demand, and finally identifying the load condition and optimizing power distribution, so that the stable operation of the power grid under extreme weather and after disasters is ensured.

Description

Technical Field

[0001] This invention relates to the field of electricity load forecasting technology, and more specifically, to a method for forecasting electricity load under extreme weather conditions. Background Technology

[0002] Against the backdrop of frequent extreme weather events, power grid operation faces severe challenges, and traditional load forecasting and dispatching methods are no longer sufficient to meet the demands. Existing technologies often overlook the differentiated impacts of extreme weather on different power generation facilities when forecasting power generation, especially failing to fully consider the instability of renewable energy, which can easily lead to forecasting errors and consequently power shortages or waste. Regional divisions rely heavily on experience and do not take into account the physical connections and weak points of the power grid, which cannot guarantee the efficiency of power transmission within the region, nor can it prevent the chain reaction caused by faults in weak lines, resulting in insufficient resilience of the power grid. Load forecasting only focuses on regular electricity consumption or a single post-disaster phase, failing to cover the direct impact of disasters on regular electricity consumption and the additional demand throughout the entire post-disaster phase, making it difficult to provide accurate data throughout the entire cycle and to guarantee the electricity needs of key areas and people's livelihoods during extreme weather and post-disaster phases.

[0003] No effective solutions have yet been proposed to address the problems in the relevant technologies. Summary of the Invention

[0004] To address the problems in related technologies, this invention proposes a method for predicting electricity load under extreme weather conditions, in order to overcome the technical problems of large generation power prediction errors, unreasonable regional division, incomplete load prediction, and blind and inefficient power dispatch in existing related technologies.

[0005] Therefore, the specific technical solution adopted by the present invention is as follows:

[0006] A method for predicting electricity load under extreme weather conditions, comprising the following steps:

[0007] S1. Identify the sources of power transmission within the impact range of extreme weather, analyze whether they include renewable energy generation, and assess whether extreme weather will affect these power generation facilities in order to predict the total power generation capacity;

[0008] S2. Based on the power grid connection, the area affected by extreme weather is divided into several different regions;

[0009] S3. Predict whether extreme weather will cause physical damage to the area. If damage occurs, calculate the additional electricity load demand during reconstruction and combine this demand with the regular electricity load predicted based on population distribution and building type to obtain the total electricity load demand of the area under extreme weather and in the post-disaster phase.

[0010] S4. Based on the prediction results, identify areas with high load pressure and sufficient load, generate dynamic diagrams of power grid risk and future power flow of each line, and calculate the optimal power flow to transfer electricity from the surplus area to the strained area, with geographical distance and convenient power grid connection as the primary constraints.

[0011] In a preferred embodiment, determining the power transmission sources within the impact range of extreme weather, analyzing whether they include renewable energy generation, and assessing whether extreme weather will affect these power generation facilities to predict total power generation includes the following steps:

[0012] S11. Based on the forecast path, intensity and impact range of extreme weather, accurately delineate the affected area on the geographic information system and determine the power transmission sources in this area, including traditional thermal power plants, hydropower stations, as well as wind farms and photovoltaic power stations.

[0013] Extreme weather includes typhoons, lightning, heavy rain, hail, strong winds, dust storms, high temperatures, and cold waves;

[0014] S12. Based on the generated power supply list, the extent to which renewable energy power generation facilities are affected by extreme weather is analyzed, and this is combined with the basic power generation data of each facility. Through weighted calculation, the power generation during the entire extreme weather event is comprehensively predicted, and finally the total power generation is calculated.

[0015] In a preferred embodiment, the step of analyzing the extent to which renewable energy power generation facilities are affected by extreme weather based on the generated power list, combining this with the basic power generation data of each facility, and comprehensively predicting the power generation during the entire extreme weather event through weighted calculation, and finally calculating the total power generation includes the following steps:

[0016] S121. Based on the generated power supply list, analyze the differentiated impact of extreme weather on different power generation facilities, and set a quantitative output reduction factor for each renewable energy source.

[0017] S122. Combine the base power generation of each power generation facility under normal weather conditions with the output reduction factor to obtain the expected power generation of each facility during the duration of an extreme weather event.

[0018] S123. Calculate the predicted power generation of each facility based on its design output and reduction factor. Then, sum the predicted power generation of all facilities. Based on the intensity of extreme weather changes over time and the calculation of key time points, generate a regional total power generation prediction curve that dynamically changes over time. The calculation formula is as follows:

[0019] ;

[0020] in, The total number of power generation facilities, No. The design of each facility contributed to the effort. For the first Each facility at any time The reduction factor, 0 ≤ ≤ 1, This represents the total power generation capacity.

[0021] As a preferred embodiment, dividing the area affected by extreme weather into multiple different regions based on the grid connection relationship includes the following steps:

[0022] S21. Based on the geographic information of the power grid, extract all key nodes within the range affected by extreme weather, including power plants, substations, distribution centers and transmission lines connecting them, and construct a complete physical connection map of the power grid.

[0023] S22. Identify key hub nodes that undertake major power transmission tasks and connect multiple regions, as well as vulnerable lines affected by extreme weather, through network theory.

[0024] S23. Taking the core hub node as the center, the substations and load areas that are directly electrically connected are divided into an initial unit. At the same time, the areas of weak lines affected by extreme weather are pre-divided to form multiple preliminary areas that are relatively independent electrically but closely connected internally.

[0025] S24. Verify the initially divided areas to ensure that each area has the ability to operate independently under extreme weather conditions, and adjust the boundaries of the zones to achieve a balance between power supply and demand in each area.

[0026] As a preferred embodiment, the specific steps for predicting whether extreme weather will cause physical damage to the area, calculating the additional electricity load demand during reconstruction if damage occurs, and combining this demand with the conventional electricity load predicted based on population distribution and building type to derive the total electricity load demand of the area under extreme weather and in the post-disaster phase are as follows:

[0027] S31. By comprehensively utilizing meteorological forecast data, geological and hydrological information, and historical disaster data, determine whether the current extreme weather will exceed the defense standards of buildings and infrastructure in the area, and construct a predictive model to calculate the damage value.

[0028] S32. Based on the predicted physical damage values, the post-disaster process is divided into three stages: emergency rescue, temporary resettlement, and full recovery. At the same time, for each stage, the power consumption patterns of its key activities are analyzed to form a phased reconstruction load template.

[0029] S33. Based on the inherent attributes of the region, such as population density and building function, predict the basic conventional electricity load curve under normal conditions, and immediately correct it according to real-time extreme weather to obtain the conventional load directly affected by the current weather. At the same time, dynamically superimpose the corrected conventional load with the predicted post-disaster reconstruction load. When the current forecast disaster intensity has exceeded the defense standard, the conventional load affected by the weather is the main factor, and the reconstruction load of the corresponding stage is added to the gradually recovering conventional load to obtain a total electricity load demand curve from the occurrence of the disaster to the post-disaster recovery.

[0030] As a preferred implementation, the specific steps for comprehensively utilizing meteorological forecast data, geological and hydrological information, and historical disaster data to construct a predictive model, determine whether the current extreme weather will exceed the fortification standards of buildings and infrastructure in the area, and construct the predictive model to calculate the damage value are as follows:

[0031] S311. Obtain current extreme weather forecast data from meteorological and hydrological departments and quantify it into a unified intensity of disaster-causing factors. Simultaneously, the statutory defense standards corresponding to this area are extracted and quantified into comparable defense standard values. ;

[0032] S312, Intensity of disaster-causing factors Compared with the defense standard value In comparison, when This indicates that the intensity of the currently predicted disaster has exceeded the defense standard;

[0033] S313. Construct a prediction model and use the formula of the prediction model to calculate the physical damage value, where the prediction formula is:

[0034] :

[0035] in, This represents the physical damage value, ranging from 0 to 1. The intensity of disaster-causing factors is a vector, composed of quantitative indicators such as meteorological forecast data and geological and hydrological information. It represents the defense standard and is also a vector, indicating the design resistance of buildings and infrastructure in the area to various extreme weather conditions; These are all vulnerability parameters, obtained by analyzing historical disaster data. The parameters that control the overall rate of risk growth This reflects the nonlinear relationship between risk and the degree of exceeding the limit.

[0036] As a preferred implementation, the process of dividing the post-disaster process into three stages—emergency rescue, temporary resettlement, and full recovery—based on the predicted physical damage values, and analyzing the power consumption patterns of key activities in each stage to form a phased reconstruction load template, includes the following steps:

[0037] S321. Based on the predicted physical damage values, dynamically set the equipment, duration, and conversion nodes required for each stage;

[0038] S322. During the emergency rescue phase, analyze its core activities, including search and rescue, emergency lighting, medical emergency care, operation of drainage pumping stations, and operation of the emergency command center. Combine the typical equipment power and expected operating time of these activities to form a load template with safety as the core. During the temporary resettlement phase, analyze its core activities, including basic living electricity at centralized resettlement sites, continuous operation of medical points, and power supply for water supply facilities and temporary communication facilities. Based on this, form a load template with a long duration and stable load level.

[0039] S323. Connect the load templates of these three stages sequentially on the time axis, and scale the total load in each template proportionally according to the predicted physical damage values ​​to form a dynamic reconstruction load curve that starts from the post-disaster period and runs through the entire recovery cycle.

[0040] As a preferred implementation, the process of predicting the basic conventional electricity load curve under normal conditions based on the inherent attributes of the area, such as population density and building function, and immediately correcting it according to real-time extreme weather to obtain the conventional load directly affected by the current weather, while dynamically superimposing the corrected conventional load with the predicted post-disaster reconstruction load, and prioritizing the weather-affected conventional load when the currently predicted disaster intensity exceeds the design standard, and adding the reconstruction load of the corresponding stage with the gradually recovering conventional load to obtain a total electricity load demand curve from the occurrence of the disaster to post-disaster recovery, includes the following steps:

[0041] S331. Based on the inherent attributes of the region, such as population density, building ratio, and historical electricity consumption data, machine learning is used to predict the daily basic conventional electricity load curve under normal weather conditions.

[0042] S332. Integrate real-time extreme weather data and quantify the correction amount of weather factors to the baseline load through correlation analysis;

[0043] S333. Set the time point when the predicted disaster intensity exceeds the defense standard as the disaster critical point, and dynamically overlay the reconstruction load templates corresponding to the emergency rescue, temporary resettlement and full recovery stages with the corrected conventional load that is in the recovery and rise channel point by point to generate the total power load demand curve from the beginning of the disaster through the post-disaster emergency response.

[0044] As a preferred implementation, the process of identifying areas with high load pressure and sufficient load based on the prediction results, generating a dynamic diagram of grid risk and future power flow for each line, and calculating the optimal power flow to transfer electricity from surplus areas to strained areas, with geographical distance and grid connection convenience as primary constraints, includes the following steps:

[0045] S41. Compare the total power generation and total load demand of each region in real time, identify the load-short areas where the power generation is consistently less than the load demand and there is a power shortage, and the load-surplus areas where the power generation is greater than the load demand and there is surplus power, and quantify the amount and time distribution of their power surplus and deficit.

[0046] S42. Map the regional power surplus and deficit results onto the power grid geographic connection map to visually display the risk distribution; at the same time, based on the power grid topology and predicted power generation and load data, generate the expected power flow direction and magnitude of each line in the future through power flow calculation, and generate a dynamic power flow distribution map to visually display the bottlenecks and critical paths of power transmission.

[0047] S43. Taking the actual connection relationship of the power grid and the geographical electrical distance as the primary constraints, the regions with tight loads and regions with surplus power are initially paired and combined. Then, with the objectives of minimizing network losses, maximizing voltage stability, and maximizing safety margin, and taking into account safety constraints such as line capacity and voltage fluctuations, the optimal power transmission command between each pair of complementary regions is calculated.

[0048] The beneficial effects of this invention are as follows:

[0049] 1. This invention combines extreme weather forecast information with geographic information to delineate affected areas and identify all power transmission sources. It also sets quantitative output reduction coefficients for the characteristics of renewable energy power generation facilities and performs weighted calculations using the facility's designed output as the weight, dynamically generating a regional total power generation prediction curve that changes over time. This approach not only fully considers the differentiated impact of extreme weather on different power generation facilities, avoiding errors caused by neglecting the instability of renewable energy due to a single forecasting method, but also allows grid dispatchers to have advance knowledge of power supply capacity during extreme weather events. This provides accurate and reliable power generation data support for subsequent power resource allocation and supply-demand balance analysis, reducing power shortages or waste caused by inaccurate power generation forecasts.

[0050] 2. This invention constructs a physical connection map using power grid geographic information, identifies core hub nodes and weak lines, then divides the initial units around the core hubs and segments the weak line areas. Finally, verification ensures that each area has independent operating capabilities and a balanced supply and demand. This division method not only ensures the efficiency and stability of power transmission within each area, avoiding the problem of inefficient power transmission due to a loose structure within the area, but also isolates weak line areas affected by extreme weather in advance, preventing the expansion of the fault range and the chain reaction when a weak line fails. At the same time, the independent operating capabilities and supply and demand balance of each area can reduce inter-regional power dependence and improve the overall power grid's resilience to extreme weather conditions.

[0051] 3. This invention constructs a predictive model by integrating meteorological, geological, hydrological, and historical disaster data to determine whether extreme weather exceeds the design standard and calculates the physical damage value. It then combines the landmark activities of different post-disaster stages to form a reconstruction load template. At the same time, it predicts the conventional electricity load based on the inherent attributes of the region and combines it with real-time weather correction. Finally, it dynamically superimposes the corrected conventional load and reconstruction load to generate a total electricity load demand curve. This method considers both the direct impact of extreme weather on conventional electricity consumption and comprehensively covers the additional electricity load demand at different post-disaster stages. It avoids the prediction deviation caused by focusing only on conventional load or the load of a single post-disaster stage. It can provide accurate load data for the power grid dispatch from the occurrence of a disaster to the post-disaster recovery cycle, ensuring that the power supply can meet the emergency needs during the disaster and adapt to the changes in electricity consumption at each stage of post-disaster reconstruction, ensuring the orderly progress of electricity supply for people's livelihood and reconstruction work.

[0052] 4. This invention accurately identifies areas with tight or surplus loads by comparing total power generation and total load demand in real time, quantifies power surplus and deficit, maps the results onto a grid geographic connection map, and generates a dynamic diagram of future power flow. Finally, it calculates the optimal power flow based on geographical distance and grid connection convenience as primary constraints, combined with network losses and voltage stability. This allocation method allows grid dispatchers to intuitively grasp the distribution of grid risks and power transmission bottlenecks, avoiding blind allocation due to unclear information. At the same time, it prioritizes geographical and connection convenience, while taking into account safety and efficiency goals. Under the premise of ensuring the safe and stable operation of the grid, it can achieve efficient power transmission from surplus to tight areas, minimize power waste, alleviate power supply pressure in areas with tight loads, and ensure power supply for key areas and people's livelihoods during extreme weather and post-disaster periods. Attached Figure Description

[0053] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0054] Figure 1 This is a flowchart of an electrical load forecasting method under extreme weather conditions according to an embodiment of the present invention. Detailed Implementation

[0055] To further illustrate the various embodiments, the present invention provides accompanying drawings, which are part of the disclosure of the present invention. These drawings are mainly used to illustrate the embodiments and can be used in conjunction with the relevant descriptions in the specification to explain the operating principles of the embodiments. With reference to these drawings, those skilled in the art should be able to understand other possible implementation methods and the advantages of the present invention. The components in the drawings are not drawn to scale, and similar component symbols are generally used to represent similar components.

[0056] According to an embodiment of the present invention, a method for predicting electrical load under extreme weather conditions is provided.

[0057] The present invention will now be further described in conjunction with the accompanying drawings and specific embodiments, such as... Figure 1 As shown, according to an embodiment of the present invention, a method for predicting electricity load under extreme weather conditions includes the following steps:

[0058] S1. Identify the sources of power transmission within the impact range of extreme weather, analyze whether they include renewable energy generation, and assess whether extreme weather will affect these power generation facilities in order to predict the total power generation capacity;

[0059] Furthermore, identifying the sources of power transmission within the impact area of ​​extreme weather, analyzing whether they include renewable energy generation, and assessing whether extreme weather will affect these power generation facilities, in order to predict total power generation capacity, includes the following steps:

[0060] S11. Based on the forecast path, intensity and impact range of extreme weather, accurately delineate the affected area on the geographic information system and determine the power transmission sources in this area, including traditional thermal power plants, hydropower stations, as well as wind farms and photovoltaic power stations.

[0061] S12. Based on the generated power supply list, the extent to which renewable energy power generation facilities are affected by extreme weather is analyzed, and this is combined with the basic power generation data of each facility. Through weighted calculation, the power generation during the entire extreme weather event is comprehensively predicted, and finally the total power generation is calculated.

[0062] Furthermore, based on the generated power inventory, the extent to which renewable energy power generation facilities are affected by extreme weather is analyzed, and this is combined with the basic power generation data of each facility. Through weighted calculation, the power generation during the entire extreme weather event is comprehensively predicted. Finally, the total power generation is calculated, including the following steps:

[0063] S121. Based on the generated power supply list, analyze the differentiated impact of extreme weather on different power generation facilities, and set a quantitative output reduction factor for each renewable energy source.

[0064] S122. Combine the base power generation of each power generation facility under normal weather conditions with the output reduction factor to obtain the expected power generation of each facility during the duration of an extreme weather event.

[0065] S123. Calculate the predicted power generation of each facility based on its design output and reduction factor. Then, sum the predicted power generation of all facilities. Based on the intensity of extreme weather changes over time and the calculation of key time points, generate a regional total power generation prediction curve that dynamically changes over time. The calculation formula is as follows:

[0066] ;

[0067] in, The total number of power generation facilities, No. The design of each facility contributed to the effort. For the first Each facility at any time The reduction factor, 0 ≤ ≤ 1, This represents the total power generation capacity.

[0068] S2. Based on the power grid connection, the area affected by extreme weather is divided into several different regions;

[0069] Furthermore, dividing the area affected by extreme weather into multiple different regions based on the grid connection relationships includes the following steps:

[0070] S21. Based on the geographic information of the power grid, extract all key nodes within the range affected by extreme weather, including power plants, substations, distribution centers and transmission lines connecting them, and construct a complete physical connection map of the power grid.

[0071] S22. Identify key hub nodes that undertake major power transmission tasks and connect multiple regions, as well as vulnerable lines affected by extreme weather, through network theory.

[0072] S23. Taking the core hub node as the center, the substations and load areas that are directly electrically connected are divided into an initial unit. At the same time, the areas of weak lines affected by extreme weather are pre-divided to form multiple preliminary areas that are relatively independent electrically but closely connected internally.

[0073] S24. Verify the initially divided areas to ensure that each area has the ability to operate independently under extreme weather conditions, and adjust the boundaries of the zones to achieve a balance between power supply and demand in each area.

[0074] It should be noted that by extracting key nodes to construct a complete connection diagram, and then using network theory to identify core hub nodes and weak lines, the division ensures that the key parts that undertake the main power transmission tasks are grasped, while also paying attention to vulnerable links that are easily affected. Then, the initial units are divided with the core hub as the center and the weak line area is segmented to form a preliminary area that is relatively independent in electrical structure and tightly connected internally. This ensures the efficiency and stability of power transmission within each area and avoids the expansion of the fault range when the weak line is affected by extreme weather. Finally, through verification, it is ensured that each area has the ability to operate independently under extreme weather conditions and the boundary is adjusted to achieve supply and demand balance. This not only allows each area to cope independently in extreme weather and reduces the risk of cascading failures caused by inter-area dependence, but also makes the power supply and demand of each area more reasonable.

[0075] S3. Predict whether extreme weather will cause physical damage to the area. If damage occurs, calculate the additional electricity load demand during reconstruction and combine this demand with the regular electricity load predicted based on population distribution and building type to obtain the total electricity load demand of the area under extreme weather and in the post-disaster phase.

[0076] Furthermore, the specific steps for predicting whether extreme weather will cause physical damage to the area, calculating the additional electricity load demand during reconstruction if damage occurs, and combining this demand with the conventional electricity load predicted based on population distribution and building type to derive the total electricity load demand of the area under extreme weather conditions and in the post-disaster phase are as follows:

[0077] S31. By comprehensively utilizing meteorological forecast data, geological and hydrological information, and historical disaster data, determine whether the current extreme weather will exceed the defense standards of buildings and infrastructure in the area, and construct a predictive model to calculate the damage value.

[0078] Furthermore, by comprehensively utilizing meteorological forecast data, geological and hydrological information, and historical disaster data, a predictive model is constructed to determine whether the current extreme weather will exceed the fortification standards of buildings and infrastructure in the area, and the specific steps for constructing the predictive model to calculate the damage value are as follows:

[0079] S311. Obtain current extreme weather forecast data from meteorological and hydrological departments and quantify it into a unified intensity of disaster-causing factors. Simultaneously, the statutory defense standards corresponding to this area are extracted and quantified into comparable defense standard values. ;

[0080] S312, Intensity of disaster-causing factors Compared with the defense standard value In comparison, when This indicates that the intensity of the currently predicted disaster has exceeded the defense standard;

[0081] S313. Construct a prediction model and use the formula of the prediction model to calculate the physical damage value, where the prediction formula is:

[0082] :

[0083] in, This represents the physical damage value, ranging from 0 to 1. The intensity of disaster-causing factors is a vector, composed of quantitative indicators such as meteorological forecast data and geological and hydrological information. It represents the defense standard and is also a vector, indicating the design resistance of buildings and infrastructure in the area to various extreme weather conditions; These are all vulnerability parameters, obtained by analyzing historical disaster data. The parameters that control the overall rate of risk growth This reflects the nonlinear relationship between risk and the degree of exceeding the limit.

[0084] S32. Based on the predicted physical damage values, the post-disaster process is divided into three stages: emergency rescue, temporary resettlement, and full recovery. At the same time, for each stage, the power consumption patterns of its key activities are analyzed to form a phased reconstruction load template.

[0085] Furthermore, based on the predicted physical damage figures, the post-disaster process is divided into three stages: emergency rescue, temporary resettlement, and full recovery. For each stage, the power consumption patterns of key activities are analyzed to form a phased reconstruction load template, including the following steps:

[0086] S321. Based on the predicted physical damage values, dynamically set the equipment, duration, and conversion nodes required for each stage;

[0087] S322. During the emergency rescue phase, analyze its core activities, including search and rescue, emergency lighting, medical emergency care, operation of drainage pumping stations, and operation of the emergency command center. Combine the typical equipment power and expected operating time of these activities to form a load template with safety as the core. During the temporary resettlement phase, analyze its core activities, including basic living electricity at centralized resettlement sites, continuous operation of medical points, and power supply for water supply facilities and temporary communication facilities. Based on this, form a load template with a long duration and stable load level.

[0088] S323. Connect the load templates of these three stages sequentially on the time axis, and scale the total load in each template proportionally according to the predicted physical damage values ​​to form a dynamic reconstruction load curve that starts from the post-disaster period and runs through the entire recovery cycle.

[0089] S33. Based on the inherent attributes of the region, such as population density and building function, predict the basic conventional electricity load curve under normal conditions, and immediately correct it according to real-time extreme weather to obtain the conventional load under the direct influence of the current weather. At the same time, dynamically superimpose the corrected conventional load with the predicted post-disaster reconstruction load. When the current forecast disaster intensity has exceeded the defense standard, the conventional load affected by the weather is the main factor, and the reconstruction load of the corresponding stage is added to the gradually recovering conventional load to obtain a total electricity load demand curve from the occurrence of the disaster to the post-disaster recovery.

[0090] Furthermore, based on the inherent attributes of the region, such as population density and building function, the basic conventional electricity load curve under normal conditions is predicted and immediately corrected according to real-time extreme weather to obtain the conventional load directly affected by the current weather. Simultaneously, the corrected conventional load is dynamically superimposed with the predicted post-disaster reconstruction load. When the currently predicted disaster intensity exceeds the design standard, the conventional load affected by the weather is prioritized, and the reconstruction load at the corresponding stage is added to the gradually recovering conventional load to obtain a total electricity load demand curve from the occurrence of the disaster to post-disaster recovery, including the following steps:

[0091] S331. Based on the inherent attributes of the region, such as population density, building ratio, and historical electricity consumption data, machine learning is used to predict the daily basic conventional electricity load curve under normal weather conditions.

[0092] S332. Integrate real-time extreme weather data and quantify the correction amount of weather factors to the baseline load through correlation analysis;

[0093] S333. Set the time point when the predicted disaster intensity exceeds the defense standard as the disaster critical point, and dynamically overlay the reconstruction load templates corresponding to the emergency rescue, temporary resettlement and full recovery stages with the corrected conventional load that is in the recovery and rise channel point by point to generate the total power load demand curve from the beginning of the disaster through the post-disaster emergency response.

[0094] In practice, the meteorological department predicted that the maximum wind speed at the center of the typhoon would reach 52 m / s, and the wind protection standard for buildings in the area was 42 m / s, which could withstand a level 14 typhoon.

[0095] By quantifying the intensity of disaster-causing factors, it can be known that =52m / s, and the design standard value is =42m / s, calculate the degree of exceedance: =52 / 42-1≈0.238, substitute this into the prediction model, and the parameters in the model are calibrated using historical data. =1.2, =1.8, calculated to obtain ≈0.087, therefore it is known that there is an 8.7% probability of physical damage, and post-disaster reconstruction load forecasting needs to be initiated;

[0096] Based on the predicted extent of damage, the post-disaster phases are divided and the required equipment is estimated. The emergency rescue phase includes high-intensity loads such as drainage pumping stations (2500kW), emergency lighting (800kW), and medical rescue (500kW), forming a load template with a peak of approximately 3800kW. The temporary resettlement phase includes stable loads such as electricity for daily life at resettlement sites (2000kW), water supply facilities (600kW), and communication support (400kW), forming a load template with a peak of approximately 3000kW. The full recovery phase includes a peak load template of 3500kW for reconstruction construction and fluctuating loads for the gradual recovery of industry and commerce.

[0097] Based on population density and commercial distribution, the daily peak load under normal weather conditions is predicted to be 25MW. During the typhoon, the industrial and commercial load drops to 20% of the normal value, while the residential load rises to 120%. The corrected peak load is about 8MW. The corrected peak load curve is superimposed on the reconstruction load template along the time axis to generate the total load demand curve from the typhoon landfall to 45 days after the disaster. It shows that the load change gradually increases from the initial 8MW to about 12MW during the recovery period.

[0098] S4. Based on the prediction results, identify areas with high load pressure and sufficient load, generate dynamic diagrams of power grid risk and future power flow of each line, and calculate the optimal power flow to transfer electricity from the surplus area to the strained area, with geographical distance and convenient power grid connection as the primary constraints.

[0099] Furthermore, based on the prediction results, areas with high load pressure and sufficient load are identified, and a dynamic diagram of grid risk and future power flow for each line is generated. Using geographical distance and grid connection convenience as primary constraints, the optimal power flow to transfer electricity from surplus areas to strained areas is calculated, including the following steps:

[0100] S41. Compare the total power generation and total load demand of each region in real time, identify the load-short areas where the power generation is consistently less than the load demand and there is a power shortage, and the load-surplus areas where the power generation is greater than the load demand and there is surplus power, and quantify the amount and time distribution of their power surplus and deficit.

[0101] S42. Map the regional power surplus and deficit results onto the power grid geographic connection map to visually display the risk distribution; at the same time, based on the power grid topology and predicted power generation and load data, generate the expected power flow direction and magnitude of each line in the future through power flow calculation, and generate a dynamic power flow distribution map to visually display the bottlenecks and critical paths of power transmission.

[0102] S43. Taking the actual connection relationship of the power grid and the geographical electrical distance as the primary constraints, the regions with tight loads and regions with surplus power are initially paired and combined. Then, with the objectives of minimizing network losses, maximizing voltage stability, and maximizing safety margin, and taking into account safety constraints such as line capacity and voltage fluctuations, the optimal power transmission command between each pair of complementary regions is calculated.

[0103] The direct goal of this optimization is to find an optimal set of power transmission commands, namely, the active power vector P1 transmitted from the power surplus area to the load-scarce area.

[0104] The weighted summation method is used to integrate the minimum network loss, the most stable voltage, and the highest safety margin into a unified and quantifiable comprehensive objective function. The smaller the value of this function, the better the overall system performance.

[0105] Safety requirements such as line capacity and voltage fluctuations are incorporated into the model in the form of mathematical inequality constraints, and the solved P1 must satisfy these constraints.

[0106] Under the condition of satisfying all safety constraints, the genetic algorithm in the optimization algorithm obtains the set of P1 with the smallest comprehensive objective function value. This set of P1 is the optimal solution obtained after the algorithm converges, which is the optimal power transmission power command sought.

[0107] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for predicting electricity load under extreme weather conditions, characterized in that, The method includes the following steps: S1. Identify the sources of power transmission within the impact range of extreme weather, analyze whether they include renewable energy generation, and assess whether extreme weather will affect these power generation facilities in order to predict the total power generation capacity; S2. Based on the power grid connection, the area affected by extreme weather is divided into several different regions; S21. Based on the geographic information of the power grid, extract all key nodes within the range affected by extreme weather, including power plants, substations, distribution centers and transmission lines connecting them, and construct a complete physical connection map of the power grid. S22. Identify key hub nodes that undertake major power transmission tasks and connect multiple regions, as well as vulnerable lines affected by extreme weather, through network theory. S23. Taking the core hub node as the center, the substations and load areas that are directly electrically connected are divided into an initial unit. At the same time, the areas of weak lines affected by extreme weather are pre-divided to form multiple preliminary areas that are relatively independent electrically but closely connected internally. S24. Verify the initially divided areas to ensure that each area has the ability to operate independently under extreme weather conditions, and adjust the boundaries of the zones to balance the power supply and demand in each area. S3. Predict whether extreme weather will cause physical damage to the area. If damage occurs, calculate the additional electricity load demand during reconstruction and combine this demand with the regular electricity load predicted based on population distribution and building type to obtain the total electricity load demand of the area under extreme weather and in the post-disaster phase. S4. Based on the prediction results, identify areas with high load pressure and sufficient load, generate dynamic diagrams of power grid risk and future power flow of each line, and calculate the optimal power flow to transfer electricity from the surplus area to the strained area, with geographical distance and convenient power grid connection as the primary constraints.

2. The method for predicting electrical load under extreme weather conditions according to claim 1, characterized in that, The process of determining the power transmission sources within the impact range of extreme weather, analyzing whether they include renewable energy generation, and assessing whether extreme weather will affect these power generation facilities in order to predict total power generation includes the following steps: S11. Based on the forecast path, intensity and impact range of extreme weather, accurately delineate the affected area on the geographic information system and determine the power transmission sources in this area, including traditional thermal power plants, hydropower stations, as well as wind farms and photovoltaic power stations. S12. Based on the generated power supply list, the extent to which renewable energy power generation facilities are affected by extreme weather is analyzed, and this is combined with the basic power generation data of each facility. Through weighted calculation, the power generation during the entire extreme weather event is comprehensively predicted, and finally the total power generation is calculated.

3. The method for predicting electrical load under extreme weather conditions according to claim 2, characterized in that, The process involves analyzing the extent to which renewable energy power generation facilities are affected by extreme weather based on the generated power supply list, combining this with the basic power generation data of each facility, and using weighted calculations to comprehensively predict the power generation during the entire extreme weather event. Finally, the total power generation is calculated, including the following steps: S121. Based on the generated power supply list, analyze the differentiated impact of extreme weather on different power generation facilities, and set a quantitative output reduction factor for each renewable energy source. S122. Combine the base power generation of each power generation facility under normal weather conditions with the output reduction factor to obtain the expected power generation of each facility during the duration of an extreme weather event. S123. Calculate the predicted power generation of each facility based on its design output and reduction factor. Then, sum the predicted power generation of all facilities. Based on the intensity of extreme weather changes over time and the calculation of key time points, generate a regional total power generation prediction curve that dynamically changes over time. The calculation formula is as follows: ； in, The total number of power generation facilities, For the first The design and construction of each facility For the first Each facility at any time The reduction factor, 0 ≤ ≤ 1, This represents the total power generation capacity.

4. The method for predicting electrical load under extreme weather conditions according to claim 1, characterized in that, The specific steps for predicting whether extreme weather will cause physical damage to the area, calculating the additional electricity load demand during reconstruction if damage occurs, and combining this demand with the predicted conventional electricity load based on population distribution and building type to derive the total electricity load demand of the area under extreme weather conditions and in the post-disaster phase are as follows: S31. By comprehensively utilizing meteorological forecast data, geological and hydrological information, and historical disaster data, determine whether the current extreme weather will exceed the defense standards of buildings and infrastructure in the area, and construct a predictive model to calculate the damage value. S32. Based on the predicted physical damage values, the post-disaster process is divided into three stages: emergency rescue, temporary resettlement, and full recovery. At the same time, for each stage, the power consumption patterns of its key activities are analyzed to form a phased reconstruction load template. S33. Based on the inherent attributes of the region, such as population density and building function, predict the basic conventional electricity load curve under normal conditions, and immediately correct it according to real-time extreme weather to obtain the conventional load directly affected by the current weather. At the same time, dynamically superimpose the corrected conventional load with the predicted post-disaster reconstruction load. When the current forecast disaster intensity has exceeded the defense standard, the conventional load affected by the weather is the main factor, and the reconstruction load of the corresponding stage is added to the gradually recovering conventional load to obtain a total electricity load demand curve from the occurrence of the disaster to the post-disaster recovery.

5. The method for predicting electrical load under extreme weather conditions according to claim 4, characterized in that, The specific steps for constructing a predictive model by comprehensively utilizing meteorological forecast data, geological and hydrological information, and historical disaster data to determine whether current extreme weather will exceed the fortification standards of buildings and infrastructure in the area, and to calculate the damage value using the predictive model, are as follows: S311. Obtain current extreme weather forecast data from meteorological and hydrological departments and quantify it into a unified intensity of disaster-causing factors. Simultaneously, the statutory defense standards corresponding to this area are extracted and quantified into comparable defense standard values. ; S312, Intensity of disaster-causing factors Compared with the defense standard value In comparison, when This indicates that the intensity of the currently predicted disaster has exceeded the defense standard; S313. Construct a prediction model and use the formula of the prediction model to calculate the physical damage value, where the prediction formula is: ； in, This represents the physical damage value, ranging from 0 to 1. The intensity of disaster-causing factors is a vector, composed of quantitative indicators such as meteorological forecast data and geological and hydrological information. It represents the defense standard and is also a vector, indicating the design resistance of buildings and infrastructure in the area to various extreme weather conditions; These are all vulnerability parameters, obtained by analyzing historical disaster data. The parameters that control the overall rate of risk growth This reflects the nonlinear relationship between risk and the degree of exceeding the limit.

6. The method for predicting electrical load under extreme weather conditions according to claim 4, characterized in that, Based on the predicted physical damage values, the post-disaster process is divided into three stages: emergency rescue, temporary resettlement, and full recovery. For each stage, the power consumption patterns of key activities are analyzed to form a phased reconstruction load template, including the following steps: S321. Based on the predicted physical damage values, dynamically set the equipment, duration, and conversion nodes required for each stage; S322. During the emergency rescue phase, analyze its core activities, including search and rescue, emergency lighting, medical emergency care, operation of drainage pumping stations, and operation of the emergency command center. Combine the typical equipment power and expected operating time of these activities to form a load template with safety as the core. During the temporary resettlement phase, analyze its core activities, including basic living electricity at centralized resettlement sites, continuous operation of medical points, and power supply for water supply facilities and temporary communication facilities. Based on this, form a load template with a long duration and stable load level. S323. Connect the load templates of these three stages sequentially on the time axis, and scale the total load in each template proportionally according to the predicted physical damage values ​​to form a dynamic reconstruction load curve that starts from the post-disaster period and runs through the entire recovery cycle.

7. The method for predicting electrical load under extreme weather conditions according to claim 4, characterized in that, Based on the inherent attributes of the region, such as population density and building function, the method predicts the basic conventional electricity load curve under normal conditions, and immediately corrects it according to real-time extreme weather to obtain the conventional load directly affected by the current weather. Simultaneously, the corrected conventional load is dynamically superimposed with the predicted post-disaster reconstruction load. When the currently predicted disaster intensity exceeds the design standard, the conventional load affected by the weather takes precedence, and the reconstruction load at the corresponding stage is added to the gradually recovering conventional load to obtain a total electricity load demand curve from the occurrence of the disaster to post-disaster recovery. This includes the following steps: S331. Based on the inherent attributes of the region, such as population density, building ratio, and historical electricity consumption data, machine learning is used to predict the daily basic conventional electricity load curve under normal weather conditions. S332. Integrate real-time extreme weather data and quantify the correction amount of weather factors to the baseline load through correlation analysis; S333. Set the time point when the predicted disaster intensity exceeds the defense standard as the disaster critical point, and dynamically overlay the reconstruction load templates corresponding to the emergency rescue, temporary resettlement and full recovery stages with the corrected conventional load that is in the recovery and rise channel point by point to generate the total power load demand curve from the beginning of the disaster through the post-disaster emergency response.

8. The method for predicting electrical load under extreme weather conditions according to claim 1, characterized in that, The process of identifying areas with high load pressure and sufficient load based on prediction results, generating dynamic maps of grid risk and future power flow for each line, and calculating the optimal power flow to transfer electricity from surplus areas to strained areas, with geographical distance and grid connection convenience as primary constraints, includes the following steps: S41. Compare the total power generation and total load demand of each region in real time, identify the load-short areas where the power generation is consistently less than the load demand and there is a power shortage, and the load-surplus areas where the power generation is greater than the load demand and there is surplus power, and quantify the amount and time distribution of their power surplus and deficit. S42. Map the regional power surplus and deficit results onto the power grid geographic connection map to visually display the risk distribution; at the same time, based on the power grid topology and predicted power generation and load data, generate the expected power flow direction and magnitude of each line in the future through power flow calculation, and generate a dynamic power flow distribution map to visually display the bottlenecks and critical paths of power transmission. S43. Taking the actual connection relationship of the power grid and the geographical electrical distance as the primary constraints, the regions with tight loads and regions with surplus power are initially paired and combined. Then, with the objectives of minimizing network losses, maximizing voltage stability, and maximizing safety margin, and taking into account safety constraints such as line capacity and voltage fluctuations, the optimal power transmission command between each pair of complementary regions is calculated.

Images