Method and system for identifying power imbalance state of power system under climate and weather change

Abstract

The invention discloses a method and a system for identifying a power imbalance state of a power system under climate and weather changes. The method comprises the following steps: determining climate and weather environment data of a target region; determining a current climate and meteorological environment event of the target region; if the climate and meteorological environment event meets the preset type, triggering an electric power imbalance state identification process of a target electric power system in the target region; constructing a source side output model matched with the power supply side, and constructing a load model matched with the load side; constructing a source-load coupling prediction model of the target power system; inputting real-time power operation data of the target power system into the source-load coupling prediction model to obtain power gap data; obtaining an identification result of the current state of the target power system; and generating an imbalance state identification log of the target power system. According to the method and the system for identifying the electric power unbalance state of the electric power system under the climate and weather change, the electric power unbalance state is comprehensively evaluated from the source side and the load side.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of power system reliability evaluation and environmental intelligent monitoring, in particular to a power imbalance state identification method and system for a power system under climate and meteorological changes. BACKGROUND

[0002] With the intensification of global climate change, high-intensity and large-scale compound high-temperature drought events occur frequently. Such "compound" disasters caused by high temperature and drought have a very serious impact on the power system. On the power supply side, high temperature and drought can significantly reduce the power generation of renewable energy sources such as hydropower, wind power, and photovoltaic power. On the load side, sustained high temperature leads to a substantial increase in cooling load. In this case, it is easy to cause a "shortage of supply" in the regional power system, resulting in "power gap". If it continues to spread in time and space, it will eventually form a "drought" phenomenon similar to the meteorological and hydrological fields.

[0003] The existing technology for identifying the imbalance state of power supply and demand only focuses on analyzing the state of one side, such as only analyzing the power supply side or only analyzing the load side. However, this method cannot take into account the interactive influence between the power supply side and the load side, leading to inaccurate identification of the imbalance state of the power system, which in turn affects the subsequent stability control of the power system and is not conducive to the precise control of the power system. SUMMARY

[0004] The present application provides a power imbalance state identification method and system for a power system under climate and meteorological changes to solve the technical problem that the existing technology cannot identify the imbalance state of the power system from both the power supply side and the load side simultaneously while taking into account climate and meteorological environmental events, thereby achieving the effect of accurately identifying the imbalance state of the power system by correlating climate and meteorological environmental events with the characteristics of the power supply and load of the power system.

[0005] To solve the above technical problems, the present application provides a power imbalance state identification method for a power system under climate and meteorological changes, comprising:

[0006] In response to an identification request, determine the climate and meteorological environmental data of the target area;

[0007] Based on the climate and meteorological environmental data, determine the current climate and meteorological environmental event in the target area;

[0008] If the climate and meteorological environmental event meets the preset type, trigger the power imbalance state identification process of the target power system in the target area, wherein the power imbalance state identification process comprises:

[0009] Construct a source-side output model matched with the power supply side of the target power system, and construct a load model matched with the load side of the target power system;

[0010] constructing a source-load coupling prediction model of the target power system based on the source-side output model and the load model;

[0011] inputting real-time power operation data of the target power system into the source-load coupling prediction model to obtain power gap data of the target power system;

[0012] analyzing the power gap data based on a spatiotemporal clustering method to obtain an identification result of a current state of the target power system;

[0013] generating an imbalance state identification log of the target power system based on the identification result.

[0014] As one of the preferred schemes, after the imbalance state identification log of the target power system is generated based on the identification result, the method further includes:

[0015] inputting the identification result into a constructed spatiotemporal migration model to obtain a spatiotemporal migration analysis result of the current state;

[0016] determining a change result of the current state based on the spatiotemporal migration analysis result;

[0017] integrating the change result into the imbalance state identification log.

[0018] As one of the preferred schemes, the climate and meteorological environment data at least includes precipitation data and temperature data;

[0019] The method for determining the current climate and meteorological environment event of the target region based on the climate and meteorological environment data includes:

[0020] performing feature analysis on the precipitation data to obtain a standardized precipitation index of the target region; and

[0021] performing feature analysis on the temperature data to obtain a standardized temperature index of the target region.

[0022] integrating the standardized precipitation index and the standardized temperature index to obtain a comprehensive environment evaluation index;

[0023] performing comparative analysis on the comprehensive environment evaluation index to determine the corresponding current climate and meteorological environment event of the target region.

[0024] As one of the preferred schemes, the source-side output model at least includes a hydropower generation model, a wind power generation model, and a photovoltaic power generation model; and the load model at least includes a refrigeration load prediction model.

[0025] The source-load coupling prediction model of the target power system is constructed based on the source-side output model and the load model, and the source-load coupling prediction model of the target power system is constructed based on the source-side output model and the load model.

[0026] Water power prediction data obtained by the water power generation model, wind power prediction data obtained by the wind power generation model, and photovoltaic prediction data obtained by the photovoltaic generation model are respectively acquired.

[0027] The refrigeration load prediction data obtained by the load model is acquired.

[0028] The source-load coupling prediction model of the target power system is constructed based on the improved random forest regression algorithm and at least the water power prediction data, the wind power prediction data, the photovoltaic prediction data, the refrigeration load prediction data, and the climate and meteorological environment data.

[0029] As one of the preferred solutions, the real-time power operation data of the target power system is input into the source-load coupling prediction model to obtain power gap data of the target power system, and the real-time power operation data of the target power system is input into the source-load coupling prediction model to obtain power gap data of the target power system.

[0030] The power generation prediction result and the load prediction result of the target region are obtained based on the source-load coupling prediction model.

[0031] The power gap value is calculated according to the power generation prediction result and the load prediction result.

[0032] The power gap value is normalized to obtain the power gap data.

[0033] Another embodiment of the application provides a power system power imbalance state recognition system under climate and meteorological changes, comprising:

[0034] A data acquisition module is configured to determine climate and meteorological environment data of a target region in response to a recognition request.

[0035] An event determination module is configured to determine a current climate and meteorological environment event of the target region based on the climate and meteorological environment data.

[0036] A type verification module is configured to trigger a power imbalance state recognition process of a target power system in the target region if the climate and meteorological environment event meets a preset type, wherein the power imbalance state recognition process comprises:

[0037] A source-load modeling module is configured to construct a source-side output model matched with a power source side of the target power system, and construct a load model matched with a load side of the target power system.

[0038] An integration module is configured to construct a source-load coupling prediction model of the target power system based on the source-side output model and the load model.

[0039] a gap prediction module, configured to input real-time power operation data of the target power system into the source-load coupling prediction model to obtain power gap data of the target power system;

[0040] an identification module, configured to analyze the power gap data based on a spatiotemporal clustering method to obtain an identification result of a current state of the target power system;

[0041] a log generation module, configured to generate an imbalance state identification log of the target power system based on the identification result.

[0042] As one of the preferred solutions, after the log generation module, a spatiotemporal migration analysis module is further included, configured to:

[0043] input the identification result into a constructed spatiotemporal migration model to obtain a spatiotemporal migration analysis result of the current state;

[0044] determine a change result of the current state based on the spatiotemporal migration analysis result;

[0045] integrate the change result into the imbalance state identification log.

[0046] As one of the preferred solutions, the climate and meteorological environment data at least includes precipitation data and temperature data;

[0047] The event determination module is further configured to:

[0048] perform feature analysis on the precipitation data to obtain a standardized precipitation index of the target region; and

[0049] perform feature analysis on the temperature data to obtain a standardized temperature index of the target region.

[0050] integrate the standardized precipitation index and the standardized temperature index to obtain a comprehensive environment evaluation index;

[0051] perform comparative analysis on the comprehensive environment evaluation index to determine the corresponding climate and meteorological environment event of the target region at present.

[0052] As one of the preferred solutions, the source-side output model at least includes a hydropower generation model, a wind power generation model and a photovoltaic power generation model; and the load model at least includes a refrigeration load prediction model;

[0053] The integration module is further configured to:

[0054] obtain hydropower prediction data obtained by the hydropower generation model, wind power prediction data obtained by the wind power generation model and photovoltaic prediction data obtained by the photovoltaic power generation model, respectively.

[0055] obtaining refrigeration load prediction data obtained by the load model;

[0056] processing at least the water power prediction data, the wind power prediction data, the photovoltaic prediction data, the refrigeration load prediction data and the climate meteorological environment data based on the improved random forest regression algorithm to construct a source-load coupling prediction model of the target power system.

[0057] As one of the preferred solutions, the gap prediction module is further configured to:

[0058] obtaining power generation prediction results and load prediction results of the target region based on the source-load coupling prediction model;

[0059] calculating a power gap value according to the power generation prediction results and the load prediction results;

[0060] normalizing the power gap value to obtain the power gap data.

[0061] Compared with the prior art, the beneficial effects of the embodiments of the present application are at least one of the following:

[0062] (1) The present application determines the current climate meteorological environment event based on the climate meteorological environment data of the target region, and when the climate meteorological environment event of the region meets the preset type, triggers the subsequent identification action, further constructs the source-side output model of the power supply side and the load model of the load side, and further constructs the source-load coupling prediction model to obtain the power gap data of the target region, and finally obtains the identification result by analyzing it. Solving the problem that single-side analysis is difficult to reflect the comprehensive risk of the power system, realizing the early and accurate identification of the power imbalance state.

[0063] (2) The continuity of the power imbalance state is identified by the spatio-temporal clustering method, and the features of the power imbalance state are output in combination with the spatio-temporal migration model, providing fine spatio-temporal distribution basis for power system dispatching and emergency resource allocation.

[0064] (3) By quantifying the dynamic evolution law of the power imbalance state, the risk diffusion path and the influence range are predicted in advance to help the power grid operator optimize the power allocation and load management strategy, and reduce the influence of extreme climate on the power system. BRIEF DESCRIPTION OF DRAWINGS

[0065] Figure 1 is a flowchart of a power imbalance state identification method of a power system under climate meteorological changes in one of the embodiments of the present application;

[0066] Figure 2This is a schematic diagram of a power system power imbalance identification system under climate and meteorological changes in one embodiment of the present invention. Detailed Implementation

[0067] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. The purpose of providing these embodiments is to make the disclosure of the present invention more thorough and comprehensive. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0068] In the description of this application, the terms "first," "second," "third," etc., are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Therefore, a feature defined with "first," "second," "third," etc., may explicitly or implicitly include one or more of that feature. In the description of this application, unless otherwise stated, "a plurality of" means two or more.

[0069] In the description of this application, it should be noted that, unless otherwise expressly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to fixed connections, detachable connections, or integral connections; they can refer to mechanical connections or electrical connections; they can refer to direct connections or indirect connections through an intermediate medium; and they can refer to the internal communication between two components. The terms "vertical," "horizontal," "left," "right," "upper," "lower," and similar expressions used herein are for illustrative purposes only and do not indicate or imply that the device or component referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as limiting the invention. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items. Those skilled in the art can understand the specific meaning of the above terms in this application based on the specific circumstances.

[0070] In the description of this application, it should be noted that, unless otherwise defined, all technical and scientific terms used in this invention have the same meaning as commonly understood by one of ordinary skill in the art. The terminology used in this specification is for the purpose of describing specific embodiments only and is not intended to limit the invention. Those skilled in the art can understand the specific meaning of the above terms in this application based on the specific circumstances.

[0071] One embodiment of the present invention provides a method for identifying power imbalance in a power system under climate and meteorological changes. For details, please refer to [link to relevant documentation]. Figure 1 ,Figure 1 The diagram shown is a flowchart illustrating a method for identifying power system imbalance under climate and meteorological changes in one embodiment of the present invention, which includes steps S1-S8:

[0072] S1. In response to the identification request, determine the climate and meteorological environment data of the target area;

[0073] Climate and meteorological environmental data typically include meteorological data such as precipitation, temperature, wind speed, and solar radiation intensity, as well as other environmental factors that may affect power supply and demand. By connecting to data collection channels such as meteorological monitoring equipment and environmental sensors, environmental parameters related to the operation of the power system in the target area can be obtained, providing basic data support for subsequent analysis of the impact of climate and meteorological environmental events on the power system.

[0074] S2. Based on climate and meteorological environmental data, determine the current climate and meteorological environmental events in the target area;

[0075] Among them, known climate and meteorological environmental data can be used to calculate standardized precipitation index based on precipitation data, standardized temperature index based on temperature data, and composite high temperature and drought index calculated by the cumulative distribution function of the joint probability.

[0076] The summarizing precipitation data is fitted with a probability distribution g(x), which is usually a Gamma distribution. The fitted probability density function is then converted into a cumulative distribution function G(x), and after removing the zero values, it is corrected to H(x):

[0077] H(x) = q + (1-q)G(x)

[0078] Here, q represents the probability of zero precipitation.

[0079] Finally, the corrected cumulative probability H(x) is transformed into the standard normal distribution quantile of the corresponding cumulative probability, i.e., the standardized precipitation index:

[0080]

[0081] in, It follows a standard normal distribution. The positive and negative values ​​of the standardized precipitation index represent drought and flood conditions within a specific time period, respectively.

[0082] Similarly, a standardized temperature index is selected to characterize high-temperature events. The standardized temperature index is often used to measure temperature anomalies. It usually does not require fitting a Gamma distribution and is calculated directly from the temperature series after removing the trend.

[0083] Based on the calculation results of the aforementioned two indicators, a comprehensive environmental assessment index (SCEI) for evaluating the intensity of combined high temperature and drought is constructed by combining the standardized precipitation index and the standardized temperature index through a joint cumulative probability distribution function:

[0084] SCEI=Φ -1 (F(P(X1≤x1,X2>x2)))

[0085] Where P(X1≤x1,X2>x2) is the joint probability distribution, and X1 and X2 represent the standardized precipitation index and the standardized temperature index, respectively, and F is the cumulative distribution function of the joint probability.

[0086] Preferably, in one embodiment of the present invention, the climate and meteorological environmental data includes at least precipitation data and temperature data;

[0087] Based on the aforementioned climate and meteorological environmental data, determine the current climate and meteorological environmental events in the target area, including:

[0088] Characteristic analysis was performed on precipitation data to obtain the standardized precipitation index for the target area; and,

[0089] Characteristic analysis of temperature data yields the standardized temperature index for the target region.

[0090] By integrating the standardized precipitation index and the standardized temperature index, a comprehensive environmental assessment index is obtained;

[0091] Comparative analysis of comprehensive environmental assessment indicators is conducted to determine the current climate, meteorological, and environmental events in the corresponding target areas.

[0092] S3. If the climate and meteorological environmental event meets the preset type, the power imbalance state identification process of the target power system in the target area is triggered, wherein the power imbalance state identification process includes:

[0093] The preset types refer to climate and meteorological environmental events (such as compound high-temperature drought, extreme high temperatures, and persistent drought) that may significantly impact the power system's source-load balance. When the current climate and meteorological environmental event is determined to belong to this preset type, it means that the power system's power output (such as hydropower, photovoltaic, and wind power) or load demand (such as a surge in cooling load) may experience abnormal fluctuations due to environmental factors, thereby triggering the risk of power grid supply-demand imbalance. Therefore, a targeted power grid imbalance identification process needs to be initiated. Through steps such as constructing source-load models, predicting power gaps, and analyzing spatiotemporal characteristics, the current imbalance state of the power grid can be accurately identified, providing a basis for subsequent control and management decisions.

[0094] This invention uses a complex high-temperature drought event as a preset type. To achieve automatic identification of complex high-temperature drought events, a threshold method is further set to classify the SCEI results. Specifically, a critical threshold (e.g., SCEI ≤ -1.0) is set as the identification standard. When the SCEI value in a continuous period or spatial region remains below this threshold, it can be identified as a complex high-temperature drought event. Based on this, the duration, coverage area, center location (centroid), and migration trajectory of the event can be further statistically analyzed to achieve dynamic monitoring and identification of complex events.

[0095] S4. Construct a source-side output model that matches the power supply side of the target power system, and construct a load model that matches the load side of the target power system.

[0096] Preferably, in one embodiment of the present invention, the source-side power output model includes at least a hydropower generation model, a wind power generation model, and a photovoltaic power generation model; the load model includes at least a cooling load prediction model.

[0097] After identifying the regions and time periods of combined high-temperature drought, source-side power output and load models were constructed to clarify the nonlinear impact mechanism of key meteorological factors on power generation capacity and electricity demand. Hydropower, photovoltaic, and wind power models, as well as a load model, were constructed, using key meteorological variables such as temperature, precipitation, solar radiation, and wind speed as driving factors. This systematically characterizes the impact mechanism of the synergistic effect of high-temperature drought on electricity supply and demand, and is particularly suitable for addressing the uncertainty of new energy sources and the risk of load surges under extreme climate conditions. The calculation formulas are as follows:

[0098] To address the issue of sharp decreases in water flow under drought scenarios in hydropower models, this model introduces a rainfall transformation term C. r ·A·pre t This represents the amount of precipitation that can actually be converted into runoff per unit time. Additionally, the evaporation term k·A is introduced. res ·(T t -T0) reflects the evaporation behavior of reservoirs under high-temperature conditions. As the ambient temperature rises, evaporation losses increase significantly, directly reducing the effective water volume available for power generation. Considering the feedback process in the hydrological cycle—the water vapor formed by reservoir evaporation needs to undergo atmospheric transport and condensation processes, and its return to the region to form effective precipitation has significant time lags and uncertainties—it can be considered an unrecoverable loss within the prediction timescale. With the above improvements, this model can more accurately reflect the weakening effect of combined high-temperature and drought conditions on hydropower output.

[0099] P t h =η·ρ w ·g·(C r ·A·pre t -k·Ares ·(T t -T0))·H t

[0100] Among them, P t h η represents the hydropower output (MW) at time t, that is, the power output of the hydropower generation system per unit time; η represents the turbine efficiency, reflecting the efficiency of the turbine in converting water energy into electrical energy; ρ w The density of water (kg / m³) 3 The value is generally taken as 1000 kg / m³. 3 g is the acceleration due to gravity (m / s²) 2 The commonly used value is 9.81 m / s. 2 C r The rainfall conversion factor represents the proportion of precipitation converted into flow; A represents the catchment area (m²). 2 Precipitation area refers to the area of ​​the land surface where precipitation collects. t The value represents the precipitation (mm) at time t; k is the evaporation coefficient (m / ℃), indicating the degree of influence of temperature on evaporation loss; A res Represents the area of ​​the reservoir (m²) 2 );T t The ambient temperature (°C) at time t is represented by T; T0 represents the reference temperature (°C); H t The water head height (m) represents the water head height at time t.

[0101] The photovoltaic power generation model incorporates a temperature correction coefficient and a wind speed influence coefficient. The temperature correction term α... p (T t -T0) describes the efficiency decrease when the component surface temperature deviates from the standard operating condition (25℃), reflecting the output attenuation mechanism caused by decreased carrier mobility and increased resistance in semiconductor devices under high-temperature conditions; the wind speed effect term β w (η w ·W 10 The model incorporates an assessment of the effect of ambient wind speed on module cooling. Higher wind speeds result in less heat accumulation on the module surface and a slower temperature rise, helping to mitigate efficiency losses caused by high temperatures. Conversely, in low-wind or no-wind conditions, heat is difficult to dissipate quickly, causing module temperatures to rise more easily and exacerbating the decline in output capacity. Since high temperatures and low winds are common during periods of combined high-temperature and drought conditions, the above modifications effectively improve the model's ability to reflect module efficiency decline under such circumstances.

[0102]

[0103] Among them, P t pv Photovoltaic output at time t; Y pvIt is the rated photovoltaic power (MW); ssrd t The actual solar radiation intensity at time t (W / m²) 2 ); ssrd STC This is the solar radiation intensity under standard test conditions, 1000 W / m². 2 ;α p This is the temperature correction coefficient. For the complex high-temperature and drought scenario addressed in this design, it is taken as a negative value to reflect the adverse effects of high temperature on photovoltaic output; T t The component surface temperature at time t (°C), where T0 is 25°C; β w This is the wind speed influence coefficient. Considering that under arid and high-temperature conditions, moderate wind speeds can effectively reduce the surface temperature of the modules, thereby mitigating thermally induced efficiency losses, it is set to a positive value to reflect the beneficial cooling effect of wind speed on the temperature of photovoltaic modules; η w W is the wind regulation coefficient. 10 It is the wind speed (m / s) at a height of 10 meters.

[0104] The wind power generation model incorporates a correction term for the change in air density with temperature. This study achieves a dynamic response to the impact of temperature on wind power output across different wind speed ranges. Wind power output is directly proportional to air density; in high-temperature environments, air density decreases, reducing the kinetic energy carried per unit wind speed, thus lowering wind power output. Without density correction, the model would systematically overestimate wind power output under high-temperature conditions. Density correction allows for a more accurate simulation of the power output reduction caused by decreased wind energy density under high-temperature conditions, particularly at medium wind speeds or critical operating conditions. This improves the model's accuracy in depicting wind power output trends and more realistically simulates the power output reduction phenomenon under combined high-temperature and drought scenarios.

[0105]

[0106] Among them, P t w P is the wind power output at time t; N This refers to the rated installed capacity of the wind turbine; W t W is the wind speed corrected for time t of the fan. in W out and W N These are the wind turbine's inlet wind speed, outlet wind speed, and full-load wind speed, respectively; ρ(T) t ) represents the air density at time t (kg / m³) 3 ), can be determined by temperature T t To correct:

[0107]

[0108] Where ρ0 is the standard air density, typically 1.225 kg / m³.3 T0 is the standard temperature, usually taken as 15℃; T t The temperature (°C) at time t is expressed as follows.

[0109] In terms of load-side modeling, to depict changes in residents' dependence on air conditioning and other cooling equipment during high-temperature and drought events, the load model is divided into two parts: base load P. l，b and cooling load P l，c Base load is electricity demand independent of meteorological factors, while cooling load is directly related to temperature and increases with rising temperatures, especially when exceeding a certain temperature threshold, where the use of cooling equipment such as air conditioners significantly increases electricity demand. The formula for cooling load is as follows:

[0110] P l，c =μ·(T) t -T0) γ +ξ·(T t -T0) 2

[0111] Where μ is the temperature sensitivity coefficient, used to describe the basic impact of temperature difference on cooling load; ν is the exponential coefficient, used to reflect the nonlinear effect of temperature difference on load; and ξ is the temperature square term coefficient, used to simulate the secondary impact of temperature difference on cooling demand.

[0112] Furthermore, it should be noted that, to ensure the accuracy and robustness of subsequent predictive model training, this invention utilizes pre-constructed models of hydropower, photovoltaic, wind power, and cooling loads to supplement or correct missing generation and load time-series data based on known historical meteorological data, thereby completing the dataset and improving sample validity. On the other hand, the outputs of each physical model themselves contain rich meteorological response characteristics. Therefore, this invention uses these as high-quality derived features, along with original meteorological features and composite high-temperature and drought indicators, to constitute the input variables of the training set, further enhancing the ability of subsequent machine learning models to characterize source-load response characteristics.

[0113] S5. Based on the source-side output model and the load model, a source-load coupling prediction model for the target power system is constructed.

[0114] This model, based on the previously established hydropower, wind power, photovoltaic, and load models, integrates composite high-temperature and drought indicators, typical meteorological factors, and source-load output data as training features. It introduces a random forest ensemble regression framework to establish a nonlinear mapping relationship between meteorological factors and the source-load side response of the power system. By randomly sampling the training set multiple times, multiple decision trees are constructed and trained separately. During the testing phase, the results of each predictor are weighted and integrated to output a comprehensive prediction result for hydropower, wind power, photovoltaic output, and cooling load. This invention improves the adaptability of the traditional random forest regression algorithm, making it more suitable for power system supply and demand forecasting under composite high-temperature and drought conditions. The main improvements include:

[0115] (1) Introduce meteorological features (temperature, precipitation, etc.) related to high temperature and drought, as well as the output results of the constructed hydropower, wind power, photovoltaic and load models, to improve historical data and use them as model inputs to enhance feature expression capabilities;

[0116] (2) Weighting is applied to typical high-temperature and drought periods in historical samples to enhance the model’s ability to learn from extreme events;

[0117] (3) Optimize the number of decision trees and the feature selection mechanism to improve the stability and prediction accuracy of the model under high-dimensional input. This will provide high-precision input data support for subsequent power shortage identification and power drought assessment.

[0118] Preferably, in one embodiment of the present invention, a source-load coupling prediction model of the target power system is constructed based on the source-side output model and the load model, including:

[0119] We obtained hydropower prediction data from the hydropower generation model, wind power prediction data from the wind power generation model, and photovoltaic prediction data from the photovoltaic power generation model, respectively.

[0120] Obtain the cooling load forecast data obtained from the load model;

[0121] Based on the improved random forest regression algorithm, at least hydropower forecast data, wind power forecast data, photovoltaic forecast data, cooling load forecast data, and climate and meteorological environmental data are processed to construct a source-load coupling forecast model for the target power system.

[0122] S6. Input the real-time power operation data of the target power system into the source-load coupling prediction model to obtain the power gap data of the target power system;

[0123] The power shortage value P is calculated based on the generator output and load demand of each node on day k. gap (i, j, k), then normalize the power gap value to [-1, 1] to obtain the corresponding power gap data PDI(i, j, k):

[0124]

[0125] Among them, P gap (i, j, k) represents the power shortage value at longitude i and latitude j on the kth day; Ω represents the power generation contribution of generator S to node (i, j) over time k days; g It is a collection of generator sets: including hydropower, photovoltaic power, wind power, thermal power, nuclear power, and others; T is the entire analysis period; P L (i, j, k) represents the load demand of node (i, j) within time k days.

[0126]

[0127] u gap (i, j, k) = sgn[P] gap (i, j, k)]

[0128] Wherein, PDI(i,j,k) is the normalized power shortage data at node (i,j) within time k days; and These represent the minimum and maximum values ​​of the power shortage in the study area over time k days; u gap (i, j, k) represents the power shortage status index; sgn(·) is the sign function, where 1, 0, and -1 correspond to the aforementioned power shortage value P, respectively. gap Positive, 0, and negative values ​​of (i, j, k).

[0129] Based on the prediction results of the power shortage, a set of future power imbalance events is identified, and the following evaluation indicators are further defined: power imbalance intensity (PDS), power imbalance duration (PDD), and power imbalance area (PDA).

[0130]

[0131] Among them, PDS n S represents the total intensity of the nth power imbalance event. n (i, j, k) represents the intensity of the nth power imbalance event at longitude i and latitude j on the kth day, specifically defined as:

[0132] S n (i,j,k)=|PDI(i,j,k)|×area(i,j)

[0133] Where area(i,j) is the area of ​​the latitude and longitude grid region, in km². 2 .

[0134] PDD n=t e -t s

[0135] Among them, PDD n Let t represent the duration of the nth power imbalance event. s Let t be the start time of this power imbalance, that is, the moment when the first power shortage occurs. e This is the end time of the power imbalance, that is, the moment when the last power shortage disappears.

[0136]

[0137] Among them, PDA n The total area of ​​the nth power imbalance event is represented in km². 2 area(i,j) is the area of ​​grid (i,j), in km². 2 I(i,j) is an indicator function for the power imbalance region. If there is a power imbalance at grid (i,j), then I(i,j) = 1; otherwise, I(i,j) = 0.

[0138] Preferably, in one embodiment of the present invention, real-time power operation data of the target power system is input into the source-load coupling prediction model to obtain power gap data of the target power system, including:

[0139] Based on the source-load coupling prediction model, the power generation prediction results and load prediction results for the target area are obtained;

[0140] The power shortage value is calculated based on the power generation forecast and load forecast results;

[0141] The power shortage value is normalized to obtain power shortage data.

[0142] S7. Analyze the power shortage data based on the spatiotemporal clustering method to obtain the identification result of the current state of the target power system;

[0143] The process includes: Power imbalance grid identification: The study area is divided into several grids with a phase source size of 0.1°×0.1° according to the spatial resolution of meteorological data. The power deficit data PDI(i,j,k) for each grid on day k is calculated. When the PDI(i,j,k) of a certain grid is <0, the grid is considered to be in a power imbalance state. At this time, the grid is marked as having experienced a power imbalance event. Spatial continuity identification of power imbalance events: On day k, if neighboring grids also have PDI(i,j,k) <0, these grids can be classified as belonging to the same power imbalance event PD. n If the total area of ​​a certain power imbalance event is PDA nIf the area of ​​the power imbalance is less than the preset minimum power imbalance area threshold A0, the event is considered invalid. Temporal continuity identification of power imbalance events: If the area of ​​the power gap grid in adjacent time periods is less than the preset threshold A0, or the grid is not continuous, the power imbalance event is considered to be interrupted in time.

[0144] S8. Generate an imbalance state identification log of the target power system based on the identification results.

[0145] The imbalance state identification log includes at least the intensity, range, duration, centroid location, centroid migration trajectory, and intensity changes of the power imbalance state in the target area.

[0146] In this study of power imbalance, the concept of the centroid is introduced. The centroid is a hypothetical point on a material system where mass is considered to be concentrated, used to analyze the spatiotemporal migration characteristics of power imbalance events. First, based on regional power supply gaps and spatiotemporal continuity, power imbalance events exceeding a time threshold T0 are considered. Second, using the centroid formula, the location of the centroid of the power imbalance event at different time periods is calculated by weighted averaging of spatial coordinates and the power gap.

[0147]

[0148] Where X represents the longitude of the centroid of the power imbalance event, and Y is the latitude, (X... i Y j ) represents the coordinates of the grid for the power imbalance event during this time period; n is the total number of grids contained in the power imbalance event area.

[0149] Secondly, the centroids of the power imbalance events are connected sequentially in time to obtain the dynamic characteristics of the power imbalance, such as its trajectory and direction. Finally, the intensity of the power imbalance is calculated: based on the physical meaning of the centroid, the power deficit at the centroid is denoted as S. nm S represents the intensity of power imbalance in month m of drought event n, and the intensity of imbalance in event n is denoted as S. n The formula is as follows:

[0150]

[0151] Wherein, P0 is the baseline value, which is based on the historical average power supply level and reflects the normal supply and demand balance under non-power imbalance conditions; w is the power grid number of power imbalance event n within k days; and u is the total number of months in which imbalance event n lasts.

[0152] Another embodiment of the present invention provides a power system imbalance identification system under climate and meteorological changes. For details, please refer to [link to relevant documentation]. Figure 2 , Figure 2The diagram shown is a schematic of a power system power imbalance identification system under climate and meteorological changes in one embodiment of the present invention, which includes:

[0153] Data acquisition module 11 is used to determine the climate and meteorological environment data of the target area in response to the identification request;

[0154] Event determination module 12 is used to determine the current climate and meteorological events in the target area based on the climate and meteorological environment data;

[0155] The type verification module 13 is used to trigger a power imbalance state identification process for the target power system in the target area if the climate and meteorological environmental event meets a preset type. The power imbalance state identification process includes:

[0156] The source-load modeling module 14 is used to construct a source-side output model that matches the power supply side of the target power system and a load model that matches the load side of the target power system.

[0157] Integration module 15 is used to construct a source-load coupling prediction model of the target power system based on the source-side output model and the load model;

[0158] The gap prediction module 16 is used to input the real-time power operation data of the target power system into the source-load coupling prediction model to obtain the power gap data of the target power system;

[0159] The identification module 17 is used to analyze the power shortage data based on the spatiotemporal clustering method to obtain the identification result of the current state of the target power system;

[0160] The log generation module 18 is used to generate an imbalance state identification log of the target power system based on the identification results.

[0161] Preferably, in one embodiment of the present invention, after the log generation module, a spatiotemporal migration analysis module 19 is further included, for:

[0162] The identification results are input into the constructed spatiotemporal migration model to obtain the spatiotemporal migration analysis results of the current state;

[0163] The change in the current state is determined based on the spatiotemporal migration analysis results;

[0164] The changes are integrated into the imbalance state identification log.

[0165] Preferably, in one embodiment of the present invention, the climate and meteorological environmental data includes at least precipitation data and temperature data;

[0166] The event determination module is also used for:

[0167] Feature analysis is performed on the precipitation data to obtain the standardized precipitation index for the target area; and,

[0168] The temperature data is subjected to feature analysis to obtain the standardized temperature index of the target region.

[0169] By integrating the standardized precipitation index and the standardized temperature index, a comprehensive environmental assessment index is obtained.

[0170] The comprehensive environmental assessment indicators are compared and analyzed to determine the current climate and meteorological environmental events in the target area.

[0171] Preferably, in one embodiment of the present invention, the source-side power output model includes at least a hydropower generation model, a wind power generation model, and a photovoltaic power generation model; the load model includes at least a cooling load prediction model;

[0172] The integrated module is also used for:

[0173] Obtain hydropower prediction data from the hydropower generation model, wind power prediction data from the wind power generation model, and photovoltaic prediction data from the photovoltaic power generation model, respectively.

[0174] Obtain the cooling load forecast data obtained from the load model;

[0175] Based on the improved random forest regression algorithm, at least the hydropower forecast data, the wind power forecast data, the photovoltaic forecast data, the cooling load forecast data, and the climate and meteorological environment data are processed to construct the source-load coupling forecast model of the target power system.

[0176] Preferably, in one embodiment of the present invention, the gap prediction module is further configured to:

[0177] Based on the source-load coupling prediction model, the power generation prediction results and load prediction results for the target area are obtained;

[0178] The power shortage value is calculated based on the power generation forecast results and the load forecast results;

[0179] The power shortage value is normalized to obtain the power shortage data.

[0180] Compared with the prior art, the beneficial effects of the embodiments of the present invention are at least one of the following:

[0181] (1) This invention determines current climate and meteorological events based on climate and meteorological environmental data of the target area. When the climate and meteorological environmental events in the area meet the preset type, subsequent identification actions are triggered. By constructing a source-side output model on the power supply side and a load model on the load side, a source-load coupling prediction model is further constructed to obtain the power shortage data of the target area. The analysis of this data ultimately yields the identification result. This solves the problem that it is difficult to reflect the comprehensive risk of the power system from a single-side analysis, and realizes early and accurate identification of power imbalance.

[0182] (2) The continuity of power imbalance state is identified by spatiotemporal clustering method, and the various characteristics of power imbalance state are output by spatiotemporal migration model, so as to provide a refined spatiotemporal distribution basis for power system scheduling and emergency resource allocation.

[0183] (3) By quantifying the dynamic evolution of power imbalance, we can help grid operators predict the risk diffusion path and impact range in advance, optimize power dispatch and load management strategies, and reduce the impact of extreme weather on the power system.

[0184] The embodiments described above are merely illustrative of several implementations of the present invention, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these modifications and improvements all fall within the scope of protection of the present invention. Therefore, the scope of protection of this patent should be determined by the appended claims.

Claims

1. A method for identifying power imbalance in a power system under climate and meteorological changes, characterized in that, include: In response to an identification request, determine the climate and meteorological environmental data for the target area; Based on the climate and meteorological environmental data, determine the current climate and meteorological environmental events in the target area; If the climate and meteorological environmental event meets the preset type, the power imbalance state identification process of the target power system in the target area is triggered, wherein the power imbalance state identification process includes: Construct a source-side output model that matches the power supply side of the target power system, and construct a load model that matches the load side of the target power system; Based on the source-side output model and the load model, a source-load coupling prediction model for the target power system is constructed. The real-time power operation data of the target power system is input into the source-load coupling prediction model to obtain the power gap data of the target power system; The power shortage data is analyzed based on the spatiotemporal clustering method to obtain the identification result of the current state of the target power system; An imbalance state identification log of the target power system is generated based on the identification results.

2. The method for identifying power imbalance in a power system under climate and meteorological changes as described in claim 1, characterized in that, After generating the imbalance state identification log of the target power system based on the identification results, the method further includes: The identification results are input into the constructed spatiotemporal migration model to obtain the spatiotemporal migration analysis results of the current state; The change in the current state is determined based on the spatiotemporal migration analysis results; The changes are integrated into the imbalance state identification log.

3. The method for identifying power imbalance in a power system under climate and meteorological changes as described in claim 1, characterized in that, The climate and meteorological environmental data include at least precipitation data and temperature data; The determination of current climate and meteorological events in the target area based on the climate and meteorological environmental data includes: Feature analysis is performed on the precipitation data to obtain the standardized precipitation index for the target area; and, The temperature data is subjected to feature analysis to obtain the standardized temperature index of the target region; By integrating the standardized precipitation index and the standardized temperature index, a comprehensive environmental assessment index is obtained. The comprehensive environmental assessment indicators are compared and analyzed to determine the current climate and meteorological environmental events in the target area.

4. The method for identifying power system imbalance under climate and meteorological changes as described in claim 1, characterized in that, The source-side power output model includes at least a hydropower generation model, a wind power generation model, and a photovoltaic power generation model; the load model includes at least a cooling load prediction model. The source-load coupling prediction model of the target power system, constructed based on the source-side output model and the load model, includes: Obtain hydropower prediction data from the hydropower generation model, wind power prediction data from the wind power generation model, and photovoltaic prediction data from the photovoltaic power generation model, respectively. Obtain the cooling load forecast data obtained from the load model; Based on the improved random forest regression algorithm, at least the hydropower forecast data, the wind power forecast data, the photovoltaic forecast data, the cooling load forecast data, and the climate and meteorological environment data are processed to construct the source-load coupling forecast model of the target power system.

5. The method for identifying power imbalance in a power system under climate and meteorological changes as described in claim 1, characterized in that, The step of inputting real-time power operation data of the target power system into the source-load coupling prediction model to obtain power gap data of the target power system includes: Based on the source-load coupling prediction model, the power generation prediction results and load prediction results for the target area are obtained; The power shortage value is calculated based on the power generation forecast results and the load forecast results; The power shortage value is normalized to obtain the power shortage data.

6. A power system imbalance identification system under climate and meteorological changes, characterized in that, include: The data acquisition module is used to determine the climate and meteorological environmental data of the target area in response to the identification request; The event determination module is used to determine the current climate and meteorological events in the target area based on the climate and meteorological environmental data. A type verification module is used to trigger a power imbalance state identification process for the target power system in the target area if the climate and meteorological environmental event meets a preset type. The power imbalance state identification process includes: The source-load modeling module is used to construct a source-side output model that matches the power supply side of the target power system, and to construct a load model that matches the load side of the target power system. An integration module is used to construct a source-load coupling prediction model of the target power system based on the source-side output model and the load model. The gap prediction module is used to input the real-time power operation data of the target power system into the source-load coupling prediction model to obtain the power gap data of the target power system. The identification module is used to analyze the power shortage data based on the spatiotemporal clustering method to obtain the identification result of the current state of the target power system; The log generation module is used to generate an imbalance state identification log of the target power system based on the identification results.

7. A power system imbalance identification system under climate and meteorological changes as described in claim 6, characterized in that, Following the log generation module, a spatiotemporal migration analysis module is also included, used for: The identification results are input into the constructed spatiotemporal migration model to obtain the spatiotemporal migration analysis results of the current state; The change in the current state is determined based on the spatiotemporal migration analysis results; The changes are integrated into the imbalance state identification log.

8. A power system imbalance identification system under climate and meteorological changes as described in claim 6, characterized in that, The climate and meteorological environmental data include at least precipitation data and temperature data; The event determination module is also used for: Feature analysis is performed on the precipitation data to obtain the standardized precipitation index for the target area; and, The temperature data is subjected to feature analysis to obtain the standardized temperature index of the target region; By integrating the standardized precipitation index and the standardized temperature index, a comprehensive environmental assessment index is obtained. The comprehensive environmental assessment indicators are compared and analyzed to determine the current climate and meteorological environmental events in the target area.

9. A power system imbalance identification system under climate and meteorological changes as described in claim 6, characterized in that, The source-side power output model includes at least a hydropower generation model, a wind power generation model, and a photovoltaic power generation model; the load model includes at least a cooling load prediction model. The integrated module is also used for: Obtain hydropower prediction data from the hydropower generation model, wind power prediction data from the wind power generation model, and photovoltaic prediction data from the photovoltaic power generation model, respectively. Obtain the cooling load forecast data obtained from the load model; Based on the improved random forest regression algorithm, at least the hydropower forecast data, the wind power forecast data, the photovoltaic forecast data, the cooling load forecast data, and the climate and meteorological environment data are processed to construct the source-load coupling forecast model of the target power system.

10. A power system imbalance identification system under climate and meteorological changes as described in claim 6, characterized in that, The gap prediction module is also used for: Based on the source-load coupling prediction model, the power generation prediction results and load prediction results for the target area are obtained; The power shortage value is calculated based on the power generation forecast results and the load forecast results; The power shortage value is normalized to obtain the power shortage data.

Images