A pre-scheduling method in turning weather, electronic equipment and storage medium

Abstract

The application discloses a pre-dispatching method under turning weather, an electronic device and a storage medium. The method comprises the following steps: predicting a temperature-sensitive load prediction result corresponding to a dispatch date through a target load prediction model according to historical load data and historical meteorological data before the dispatch date; obtaining a wind power output prediction result and a photovoltaic output prediction result corresponding to the dispatch date; generating a source-load joint operation typical working condition set based on the temperature-sensitive load prediction result, the wind power output prediction result and the photovoltaic output prediction result, and constructing a target dispatching model based on the source-load joint operation typical working condition set; solving the target dispatching model to obtain a target dispatching scheme of the dispatch date; wherein the target dispatching model comprises a target function, a pre-dispatching constraint and a re-dispatching constraint. The technical scheme can realize active defense of supply-demand imbalance under extreme weather, thereby effectively improving the resilience and power supply reliability of the power grid under extreme scenarios.

Description

Technical Field

[0001] This invention relates to the field of smart grid technology, and in particular to a pre-dispatch method, electronic device and storage medium for changing weather conditions. Background Technology

[0002] In recent years, frequent extreme weather events such as cold waves and extreme heat waves have led to a decline in the output of new energy sources such as wind power and photovoltaic power, and a sharp increase in temperature-sensitive loads, posing a serious threat to the power system's supply and demand balance. Examples include large-scale power outages caused by cold waves and rotating power rationing implemented by the grid due to sustained high temperatures. At the same time, with the accelerated electrification of end-use energy, the proportion of temperature-sensitive loads such as electric heating and air conditioning in the total load continues to increase, making the power system more vulnerable to extreme weather. Against this backdrop, achieving accurate forecasting of temperature-sensitive loads during extreme weather transitions and conducting proactive power system dispatch based on these forecasts has become a crucial foundation for coping with extreme weather and ensuring a balance between power supply and demand.

[0003] However, historical data on extreme weather events such as cold waves and high temperatures are relatively scarce and exhibit complex patterns, displaying typical small sample characteristics. This makes traditional load forecasting methods, such as time series models relying on large amounts of stable historical data and single machine learning models, difficult to apply, generally resulting in a sharp drop in forecast accuracy and insufficient adaptability. Inaccurate load forecasting will directly lead to blind dispatching decisions, leaving the power system lacking the ability to proactively allocate resources when facing extreme load shocks, thus falling into a passive response, and may even trigger supply-demand imbalances and the risk of localized power curtailment.

[0004] Therefore, conducting precise forecasting research on temperature-sensitive loads in response to the increasingly frequent extreme weather events such as cold waves and high temperatures, and formulating forward-looking dispatching strategies accordingly, has important theoretical value and practical significance for ensuring the safety of the new power system, improving energy utilization efficiency, and supporting the stable operation of the social economy. Summary of the Invention

[0005] This invention provides a pre-scheduling method, electronic device, and storage medium for transitional weather conditions, enabling proactive defense against supply and demand imbalances under extreme weather conditions, thereby effectively improving the resilience and power supply reliability of the power grid in extreme scenarios.

[0006] According to one aspect of the present invention, a pre-scheduling method for transitional weather conditions is provided, comprising: Based on historical load data and historical meteorological data prior to the scheduling date, the temperature-sensitive load forecast result corresponding to the scheduling date is predicted using the target load forecasting model. Obtain the wind power output prediction results and photovoltaic power output prediction results corresponding to the scheduling date; Based on the predicted results of temperature-sensitive loads, wind power output, and photovoltaic power output, a typical set of operating conditions for source-load joint operation is generated, and a target scheduling model is constructed based on the typical set of operating conditions for source-load joint operation. Solving the target scheduling model yields the target scheduling scheme for the specified scheduling date; wherein the target scheduling model includes an objective function, pre-scheduling constraints, and rescheduling constraints.

[0007] In some possible implementations, the step of predicting the temperature-sensitive load forecast result corresponding to the scheduling date using a target load forecasting model based on historical load data and historical meteorological data prior to the scheduling date includes: Historical load data and historical meteorological data prior to the scheduling date are obtained, and the historical load data and historical meteorological data are preprocessed to obtain meteorological and load time series data; Based on the meteorological and load time series data, temperature-sensitive loads were identified by screening through correlation coefficients. The initial load prediction model is trained based on the temperature-sensitive load and the corresponding meteorological data to obtain the target load prediction model; The target load prediction model is used to predict the temperature-sensitive load for the scheduled date.

[0008] In some possible implementations, the step of training an initial load prediction model based on the temperature-sensitive load and the corresponding meteorological data to obtain the target load prediction model includes: The temperature-sensitive load and the corresponding meteorological data are augmented using a sample augmentation algorithm to obtain augmented sample data. The initial load prediction model is trained based on the expanded sample data to obtain the target load prediction model.

[0009] In some possible implementations, the generation of a typical set of source-load joint operation conditions based on the temperature-sensitive load prediction results, the wind power output prediction results, and the photovoltaic power output prediction results includes: Based on the predicted results of temperature-sensitive loads, wind power output, and photovoltaic power output, multiple initial source-load joint operation conditions are generated through random simulation. Clustering algorithms are used to cluster multiple initial source-load joint operation conditions to obtain a typical set of source-load joint operation conditions.

[0010] In some possible implementations, constructing the target scheduling model based on the typical operating condition set of source-load joint operation includes: Construct the objective function, the pre-scheduling constraints corresponding to the scheduling date, and the rescheduling constraints under each typical condition in the typical operating condition set of source-load joint operation. Based on the objective function, the pre-scheduling constraints, and the rescheduling constraints, the objective scheduling model is constructed. The objective function is used to minimize the total expected operating cost of the power system before and after the scheduling date. The total expected operating cost includes generator operating cost, wind and solar curtailment cost, and load shedding penalty cost.

[0011] In some possible implementations, the pre-scheduling constraints include: Operating constraints of distributed generator sets, operating constraints of battery energy storage systems, operating constraints of hydrogen storage systems, power constraints of wind and photovoltaic power generation, load shedding constraints, and system power balance constraints.

[0012] In some possible implementations, the rescheduling constraints include: Operating constraints of distributed generator sets, operating constraints of battery energy storage systems, operating constraints of hydrogen storage systems, power constraints of wind and photovoltaic power generation, load shedding constraints, and system power balance constraints.

[0013] In some possible implementations, the step of solving the target scheduling model to obtain the target scheduling scheme for the scheduling date includes: The target scheduling model is solved using a commercial solver; The running parameters from the solution results of the commercial solver are obtained to form the target scheduling scheme for the scheduling date.

[0014] According to another aspect of the present invention, a pre-scheduling device for changing weather conditions is provided, comprising: The load forecasting module is used to predict the temperature-sensitive load forecasting result corresponding to the scheduling date based on historical load data and historical meteorological data prior to the scheduling date, using a target load forecasting model. The power output acquisition module is used to acquire the wind power output prediction results and photovoltaic power output prediction results corresponding to the scheduling date; The operating condition generation and modeling module is used to generate a typical operating condition set for source-load joint operation based on the temperature-sensitive load prediction results, the wind power output prediction results, and the photovoltaic output prediction results, and to construct a target scheduling model based on the typical operating condition set for source-load joint operation. The model solving module is used to solve the target scheduling model to obtain the target scheduling scheme for the scheduling date; wherein, the target scheduling model includes an objective function, pre-scheduling constraints, and rescheduling constraints.

[0015] According to another aspect of the present invention, an electronic device is provided, the electronic device comprising: At least one processor; and a memory communicatively connected to the at least one processor; The memory stores a computer program that can be executed by the at least one processor, which enables the at least one processor to perform the pre-scheduling method for changing weather conditions as described in any embodiment of the present invention.

[0016] According to another aspect of the present invention, a computer-readable storage medium is provided, the computer-readable storage medium storing computer instructions for causing a processor to execute and implement the pre-scheduling method under changing weather conditions as described in any embodiment of the present invention.

[0017] The technical solution of this invention adopts the method of SMOTE (Synthetic Minority Oversampling) data augmentation and bidirectional long short-term memory fusion to solve the load forecasting problem caused by the small sample characteristics of loads during extreme weather transitions. On this basis, the load forecasting results are coupled with the pre-scheduling decision of electric-hydrogen hybrid energy storage. By utilizing the unique long-cycle energy transfer capability of hydrogen energy storage, a forward-looking defense system for the power system to cope with extreme weather such as cold waves and high temperatures is constructed through pre-scheduling before extreme weather and re-scheduling during extreme weather.

[0018] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of the present invention, nor is it intended to limit the scope of the invention. Other features of the invention will become readily apparent from the following description. Attached Figure Description

[0019] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying 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.

[0020] Figure 1 A flowchart illustrating a pre-scheduling method under changing weather conditions provided in an embodiment of the present invention; Figure 2 A flowchart illustrating another pre-scheduling method under changing weather conditions provided in this embodiment of the invention; Figure 3 A flowchart for predicting temperature-sensitive loads under extreme weather conditions based on the SMOTE algorithm and bidirectional long short-term memory algorithm provided in this embodiment of the invention; Figure 4This is a schematic diagram showing the load prediction results before and after the cold wave transition weather corresponding to different prediction methods provided in the embodiments of the present invention. Figure 5 A line graph of the source load output scenario considering temperature-sensitive load prediction in the system is presented; Figure 6 A schematic diagram showing the power system supply and demand balance before and under extreme weather conditions, taking into account pre-scheduled scheduling. Figure 7 A schematic diagram showing the power system supply and demand balance before and under extreme weather conditions, without considering pre-scheduling. Figure 8 This is a schematic diagram of a pre-scheduling device for changing weather conditions provided in an embodiment of the present invention; Figure 9 A schematic diagram of the structure of an electronic device for implementing the pre-scheduling method under changing weather conditions according to an embodiment of the present invention. Detailed Implementation

[0021] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0022] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0023] Figure 1 This is a flowchart illustrating a pre-dispatch method under transitional weather conditions provided by an embodiment of the present invention. This embodiment is applicable to situations involving load forecasting under extreme transitional weather conditions and conducting forward-looking dispatching of the power system based on the forecast results. This method can be executed by a pre-dispatch device under transitional weather conditions, which can be implemented in hardware and / or software and can be configured in electronic equipment. Figure 1 As shown, the method specifically includes the following steps: S110. Based on historical load data and historical meteorological data prior to the scheduling date, predict the temperature-sensitive load forecast result corresponding to the scheduling date using the target load forecasting model.

[0024] The scheduling date can be understood as the date on which power system scheduling is required. The scheduling date can be a day in the future when there may be extreme weather changes. The load data on the date can be predicted by the technical solution of the present invention, and a corresponding scheduling plan can be generated. Based on the scheduling plan, forward scheduling can be carried out to cope with extreme weather and ensure the balance of power supply and demand.

[0025] Historical load data refers to the load data corresponding to each load within a historical time period prior to the scheduling date; historical meteorological data refers to the meteorological data within a historical time period prior to the scheduling date, such as meteorological elements including but not limited to temperature, humidity, and irradiance.

[0026] The target load forecasting model refers to a pre-trained deep learning model used to predict load data for a scheduled date. The temperature-sensitive load forecasting result refers to the specific load data of temperature-sensitive loads predicted by the target load forecasting model, such as air conditioning load data.

[0027] Specifically, historical load data and historical meteorological data prior to the scheduling date can be obtained. Based on the historical load data and historical meteorological data, the load data of temperature-sensitive loads corresponding to the scheduling date can be predicted using the target load prediction model, thereby obtaining the temperature-sensitive load prediction results.

[0028] In some possible implementations, the step of predicting the temperature-sensitive load forecast result corresponding to the scheduling date using a target load forecasting model based on historical load data and historical meteorological data prior to the scheduling date includes: acquiring historical load data and historical meteorological data prior to the scheduling date; preprocessing the historical load data and historical meteorological data to obtain meteorological and load time series data; filtering for temperature-sensitive loads based on the meteorological and load time series data using correlation coefficients; training an initial load forecasting model based on the load data of the temperature-sensitive loads and the corresponding meteorological data to obtain the target load forecasting model; and predicting the temperature-sensitive load forecast result corresponding to the scheduling date using the target load forecasting model.

[0029] The correlation coefficient can be the Pearson correlation coefficient; the initial load prediction model can be a deep learning model that has not yet been trained on the training samples of this invention, such as a Bi-LSTM (Bidirectional Long Short-Term Memory) deep learning model.

[0030] Specifically, historical load data and meteorological data prior to the scheduling date are acquired and preprocessed to form continuous and unified meteorological and load time series data. Since the historical load data includes both temperature-sensitive and non-temperature-sensitive loads, temperature-sensitive loads significantly affected by temperature can be identified based on the correlation coefficient of the meteorological and load time series data. Then, the temperature-sensitive loads and their corresponding meteorological data are used as training samples to train an initial load prediction model, resulting in a target load prediction model with satisfactory accuracy. Finally, the target load prediction model is used to predict the temperature-sensitive load corresponding to the scheduling date.

[0031] In a preferred implementation, firstly, the historical load-meteorological data for one year is cleaned. The 3σ rule is used to detect outliers in the historical load data to identify abnormal loads. For any missing values ​​or outliers detected, Lagrange interpolation is used to fill in the data, ensuring the completeness and accuracy of the input data. After this processing step, continuous and reliable meteorological and load time series data are obtained.

[0032] Then, using temperature as a feature, the Pearson correlation coefficient between temperature and various types of loads is calculated. Temperature-sensitive loads with |r|≥0.6 are selected as model inputs, and types with weak correlation are removed.

[0033] (1) in, For the i-th load output data, The j-th load data corresponds to the outside temperature. and These represent the average values ​​of load and temperature, respectively.

[0034] Because cold waves and high temperatures occur infrequently, the corresponding historical sample size is limited, making direct training of the prediction model prone to insufficient accuracy. Therefore, training an initial load prediction model based on the load data of the temperature-sensitive load and the corresponding meteorological data to obtain the target load prediction model can include: using a sample augmentation algorithm to augment the load data of the temperature-sensitive load and the corresponding meteorological data to obtain augmented sample data; training the initial load prediction model based on the augmented sample data to obtain the target load prediction model; and using the target load prediction model to predict the temperature-sensitive load forecast result corresponding to the scheduling date.

[0035] Understandably, by using sample augmentation algorithms to enhance the load data and corresponding meteorological data of temperature-sensitive loads, and expanding the sample distribution and data volume, the model's learning and generalization capabilities for extreme transitional weather can be improved, thereby obtaining a target load prediction model with higher prediction stability and adaptability to transitional weather scenarios.

[0036] Specifically, the training sample set (load data of temperature-sensitive loads and corresponding meteorological data) is divided into training and testing sets in an 8:2 ratio. The training set is then augmented using SMOTE and input into a load forecasting model based on a Bi-LSTM deep learning model to improve its load forecasting performance under extreme weather conditions. After obtaining the target load forecasting model, the load data, meteorological data, and time information corresponding to temperature-sensitive loads over a past period are input into the target load forecasting model. Based on the learned meteorological, temporal, and load variation patterns, the target load forecasting model can make predictions and output hourly power load data for the day on the scheduling date, serving as the temperature-sensitive load forecast result.

[0037] S120. Obtain the wind power output prediction results and photovoltaic power output prediction results corresponding to the scheduling date.

[0038] Among them, the wind power output prediction result can be the predicted active power output of the wind turbine generator set, and the photovoltaic power output prediction result can be the predicted active power output of the photovoltaic generator set.

[0039] Specifically, wind power output prediction results can be obtained based on the active power of wind turbine generators in historical time periods, corresponding meteorological wind speed, wind direction, etc.; photovoltaic power output prediction results can be obtained based on the active power of photovoltaic generators in historical time periods, corresponding solar irradiance, ambient temperature, etc.

[0040] In this embodiment of the invention, there are no restrictions on the method of obtaining the wind power output prediction results and the photovoltaic power output prediction results. For example, the wind power output prediction results and the photovoltaic power output prediction results can also be obtained through some publicly available datasets.

[0041] S130. Based on the temperature-sensitive load prediction results, the wind power output prediction results, and the photovoltaic power output prediction results, a typical operating condition set for source-load joint operation is generated, and a target scheduling model is constructed based on the typical operating condition set for source-load joint operation.

[0042] The typical operating condition set for combined source-load operation refers to multiple sets of typical operating conditions (scenarios) formed by combining the hourly load output, wind power output, and photovoltaic output within the scheduling date. For example, a set of operating conditions can be represented as: load output, wind power output, and photovoltaic processing during time period t on the scheduling date. Multiple representative sets of operating conditions constitute the typical operating condition set. The target scheduling model refers to the constructed mathematical model. Based on the target scheduling model, the optimal power dispatch strategy, i.e., the target dispatch scheme, can be generated. The target scheduling model includes the objective function, pre-schedule constraints, and reschedule constraints.

[0043] Specifically, based on the forecast results of temperature-sensitive loads, wind power output, and photovoltaic power output, multiple possible source-load operation scenarios are generated through random simulation. A representative set of typical operating conditions for source-load joint operation is then generated based on these scenarios. Finally, a target scheduling model, including an objective function, pre-scheduling constraints, and rescheduling constraints, is constructed based on this set of typical operating conditions. This approach comprehensively reflects the uncertainty distribution of source loads under extreme weather conditions, making the target scheduling model more closely aligned with actual operating patterns and improving the reliability and adaptability of the scheduling scheme.

[0044] It should also be noted that the target scheduling model established based on the typical operating conditions of source-load joint operation is a two-stage pre-scheduling model of the electric-hydrogen hybrid energy storage system based on two-stage stochastic programming. It includes the objective function, the pre-scheduling constraints of the electric-hydrogen energy storage system before extreme weather, and the rescheduling constraints of the electric-hydrogen energy storage system during extreme weather under various typical scenarios.

[0045] In some possible implementations, generating a typical set of source-load joint operation conditions based on the temperature-sensitive load prediction results, the wind power output prediction results, and the photovoltaic output prediction results includes: generating multiple initial source-load joint operation conditions through random simulation based on the temperature-sensitive load prediction results, the wind power output prediction results, and the photovoltaic output prediction results; and using a clustering algorithm to cluster the multiple initial source-load joint operation conditions to obtain the typical set of source-load joint operation conditions.

[0046] Among them, the initial source-load joint operation condition refers to multiple sets of source-load operation scenarios generated by a random simulation algorithm based on the prediction results of temperature-sensitive load, wind power output, and photovoltaic power output, which are used to comprehensively cover various uncertainties in source-load fluctuations under changing weather conditions.

[0047] Specifically, based on the forecast results of temperature-sensitive loads, wind power output, and photovoltaic power output, multiple sets of different initial source-load joint operation conditions are generated through random simulation algorithms, such as Monte Carlo simulation, to ensure coverage of various possible operating scenarios of the source-load.

[0048] Furthermore, by using clustering algorithms, such as k-means clustering, all initial source-load joint operation conditions are aggregated, redundant conditions with extremely high similarity are eliminated, and several sets of typical operation conditions that can comprehensively characterize the uncertainty distribution of source-load and cover the main fluctuation characteristics are extracted to form a set of typical operation conditions for source-load joint operation.

[0049] The beneficial effect of this is that by generating sufficient initial operating conditions through random simulation, the uncertainty of the source load can be fully covered. Combined with clustering to remove redundancy, the representativeness of the operating conditions is ensured, while the amount of data is reduced.

[0050] In a preferred implementation, a typical set of operating conditions for joint source-load operation is generated through the following steps: (1) Based on the prediction results of temperature-sensitive loads And combined with wind power output forecast results And photovoltaic power output forecast results Monte Carlo simulation was used to randomly generate... A typical operating scenario (i.e., the initial source-load joint operation condition).

[0051] (2) in, , and Representing the generated first In the scenario, wind power, photovoltaics, and load are the first T Output value for a given time period. For one OK, T A matrix of columns.

[0052] (2) The k-means clustering algorithm is used to analyze the generated source load scene. Clustering is performed to identify typical operational scenarios that represent the characteristics of source loads under extreme weather conditions. (i.e., typical operating conditions for combined source and load operation): (3) in The first cluster obtained by clustering S A typical Japanese-Chinese day T The source load output during the time period, the first s The probability of occurrence for each typical scenario is denoted as . .Will After conversion, it is finally converted into 3 lines. The column is a source-load output matrix, where the three rows are wind power, photovoltaic power and load output respectively.

[0053] S140. Solve the target scheduling model to obtain the target scheduling scheme for the scheduling date.

[0054] The target scheduling scheme can be a set of optimal power scheduling strategies for the electric-hydrogen energy storage system before and on the scheduling date, such as including unit output, energy storage charging and discharging, hydrogen production and power generation, and load control instructions for each time period.

[0055] In some possible implementations, the step of solving the target scheduling model to obtain the target scheduling scheme for the scheduling date includes: solving the target scheduling model using a commercial solver; obtaining the running parameters from the solution results of the commercial solver to form the target scheduling scheme for the scheduling date.

[0056] It should be noted that the target scheduling model is a mixed-integer linear programming problem, which can be solved directly using commercial solvers such as Gurobi and CPLEX. The completed target scheduling model can be imported into a commercial solver, and solution parameters such as solution accuracy and iteration limit can be set. Then, the solver is started to optimize the target scheduling model, obtaining a solution that satisfies all constraints and optimizes the objective function (minimizing the total expected operating cost). Finally, operating parameters such as generator operating power, energy storage system charging and discharging power, wind and solar curtailment, and load shedding for each time period within the scheduling date are extracted from the commercial solver's results and used as the target scheduling scheme for the scheduling date.

[0057] The technical solution of this invention employs SMOTE data augmentation and the Bi-LSTM algorithm to conduct day-ahead forecasting of temperature-sensitive loads under extreme transitional weather conditions such as cold waves and high temperatures. Based on the load forecasting results under transitional weather conditions, typical operating scenarios of the power system that characterize load uncertainty are generated. Based on the load forecasting and its typical uncertainty scenarios, a two-stage pre-schedule model of the electric-hydrogen hybrid energy storage system is established. The established electric-hydrogen energy storage pre-schedule model is solved to obtain an optimized scheduling scheme for the electric-hydrogen energy storage system to cope with changes in the demand of temperature-sensitive loads under extreme transitional weather conditions. This solution can address the challenge of small sample data under extreme transitional weather conditions such as cold waves and high temperatures, and by combining the forecasting results with the pre-schedule of the electric-hydrogen hybrid energy storage system, it achieves proactive defense against supply and demand imbalances under extreme weather conditions by pre-charging before the arrival of extreme weather and accurately discharging during load surges, thereby effectively improving the resilience and power supply reliability of the power grid under extreme scenarios.

[0058] Figure 2 This is a flowchart illustrating another pre-scheduling method under changing weather conditions provided by an embodiment of the present invention. Based on the above embodiments, this embodiment optimizes the construction process of the target scheduling model. Specific implementation details can be found in the technical solution of this embodiment. Technical terms that are the same as or corresponding to those in the above embodiments will not be repeated here. Figure 2 As shown, the method specifically includes the following steps: S210. Based on historical load data and historical meteorological data prior to the scheduling date, predict the temperature-sensitive load forecast result corresponding to the scheduling date using the target load forecasting model.

[0059] S220. Obtain the wind power output prediction results and photovoltaic power output prediction results corresponding to the scheduling date.

[0060] S230. Based on the predicted results of the temperature-sensitive load, the predicted results of the wind power output, and the predicted results of the photovoltaic power output, a typical set of operating conditions for source-load joint operation is generated.

[0061] S240. Construct the objective function, the pre-scheduling constraints corresponding to the scheduling date, and the rescheduling constraints under each typical condition in the typical operating condition set of source-load joint operation.

[0062] S250. Based on the objective function, the pre-scheduling constraints, and the rescheduling constraints, the objective scheduling model is constructed.

[0063] The objective function is used to minimize the total expected operating cost of the power system before and after the scheduling date. The total expected operating cost includes generator operating cost, wind and solar curtailment cost, and load shedding penalty cost.

[0064] Specifically, the objective function aims to minimize the total expected operating cost of the system before and under extreme weather conditions, including generator operating costs, wind and solar curtailment costs, and load shedding penalty costs.

[0065] (4) (5) (6) In the formula, This refers to the set of nodes for conventional thermal power units in the system. This refers to the set of nodes of the wind farm in the system. This refers to the set of nodes in the photovoltaic power station within the system. This refers to the set of nodes that handle the load in the system. For the first i The unit power generation cost of a thermal power unit, $ / kWh; For the first i The conventional thermal power unit at the s In the first scenario t The power generation value at any given time, in kW; and The first s In the first scenario t Time of the first i Wind and solar power curtailment, in kW; and The costs for wind and solar power curtailment are $ / kWh, respectively. For the first s In the first scenario t Time of the first i The load shedding power of each load, in kW; The load shedding penalty cost per unit power, $ / kWh; and These are sets of runtime segments before and during extreme weather, with each segment lasting for a duration of [duration value missing]. .

[0066] In some possible implementations, the pre-scheduling constraints include: distributed generator set operation constraints, battery energy storage system operation constraints, hydrogen storage system operation constraints, wind power and photovoltaic power generation constraints, load shedding constraints, and system power balance constraints.

[0067] Specifically, the pre-scheduling constraints (pre-scheduling constraints) for the electric-hydrogen energy storage system before extreme weather events are as follows: a) Operational constraints of distributed generator sets: (7) (8) in, and The first i The lower and upper limits of the generator unit's output, in kW. and The first i The upward and downward ramp rates of the generator set, kW / h. Equation (7) indicates that the output of the generator set should be within the upper and lower limits of the generator set output, and Equation (8) indicates the ramp rate of the generator set.

[0068] b) Operating constraints of battery energy storage systems: (9) (10) (11) (12) in, For a set of nodes in a battery energy storage system; and The first i The battery energy storage in the first t Discharge and charge power at any given time, in kW; For a representation of the first i The battery energy storage in the first t The 0-1 variable representing the charging and discharging state at any given time. This indicates that the energy storage is in a discharging state, and vice versa. This indicates that the energy storage is in a charging state; For the first i The rated power of each battery energy storage system is kW; For the first i The battery energy storage in the first t Storage capacity at any given time, in kWh; and They represent the first i The charging and discharging power of a battery energy storage system; and They represent the first i The lower and upper limits of the storage capacity of a battery energy storage system. Equation (9) indicates that the energy storage charging and discharging power should be within its rated power range and should not be charged and discharged simultaneously; Equations (10)-(11) limit the energy storage capacity at any time to be within the upper and lower limits of its energy storage capacity; Equation (12) indicates that the battery energy storage capacity should be equal at the beginning and end of the operating day before an extreme event.

[0069] c) Operational constraints of hydrogen storage systems: (13) (14) in, A set of nodes for a battery energy storage system; Electrolytic cell efficiency, kg / kW and The first i The hydrogen storage system in the first t The power consumption and hydrogen production of the electrolyzer at each moment are expressed in kW and kg, respectively. For the first i The capacity of the electrolyzer in the hydrogen storage system is kW. Equation (13) gives the relationship between the power consumption of the electrolyzer and the hydrogen production; Equation (14) represents the upper and lower limits of the power consumption of the electrolyzer.

[0070] (15) (16) in, and The first i The hydrogen storage system in the first t Hydrogen consumption and discharge power at any given time, in kg and kW; The discharge efficiency of hydrogen fuel cells in a hydrogen storage system; For the first i The rated power of the hydrogen fuel cell in the hydrogen storage system is kW. Equation (15) gives the relationship between the discharge power of the hydrogen fuel cell and the hydrogen consumption; Equation (16) represents the upper and lower limits of the discharge power of the hydrogen fuel cell.

[0071] (17) (18) (19) in, For a 0-1 variable, This indicates that the hydrogen storage system is in a charging state. This indicates that the hydrogen storage system is in a discharge state; Indicates the first i The hydrogen storage system in the first t Hydrogen storage capacity at any given time and These represent the lower and upper limits of the allowable hydrogen storage capacity of the hydrogen storage container, respectively. Equation (17) indicates that the hydrogen storage system cannot be charged and discharged simultaneously at any given time; Equation (18) is used to calculate the first hydrogen storage capacity within each typical day. t The amount of hydrogen stored in the hydrogen storage system during a given time period; Equation (19) represents the upper and lower limits of the amount of hydrogen stored in the hydrogen storage system.

[0072] d) Wind and solar power generation constraints: (20) (twenty one) Equations (20)-(21) respectively restrict the first... t The amount of wind and solar power curtailed at any given time should be less than the current output of wind farms and photovoltaic power plants.

[0073] e) Load shedding constraint: (twenty two) Equation (22) restricts the first... t The load shedding capacity should be less than the current load demand.

[0074] f) System power balance constraints: (twenty three) In some possible implementations, the rescheduling constraints include: distributed generator set operation constraints, battery energy storage system operation constraints, hydrogen storage system operation constraints, wind power and photovoltaic power generation constraints, load shedding constraints, and system power balance constraints.

[0075] Specifically, the rescheduling constraints (rescheduling constraints) for the electric-hydrogen energy storage system during extreme weather events are as follows: a) Operational constraints of distributed generator sets: (twenty four) (25) (26) Equation (24) indicates that the output of the generator set should be within the upper and lower limits of the generator set output. Equation (25) indicates the ramp rate of the generator set. Equation (26) indicates the ramp relationship between the generator output at the beginning of the extreme weather and the period before the extreme weather.

[0076] b) Operating constraints of battery energy storage systems: (27) (28) (29) (30) in, and The first i The battery energy storage in the first s In the scenario, the first t Discharge and charge power at any given time, in kW; For a representation of the first i The battery energy storage in the first s In the scenario, the first t The 0-1 variable representing the charging and discharging state at any given time. This indicates that the energy storage is in a discharging state, and vice versa. This indicates that the energy storage is in a charging state; For the first i The rated power of each battery energy storage system is kW; For the first i The battery energy storage in the first s In the scenario, the first t The stored energy at any given time is kWh. Equation (27) indicates that the energy storage charging and discharging power should be within its rated power range and should not be charged and discharged simultaneously; Equations (28)-(29) limit the stored energy at any given time to be within the upper and lower limits of its energy storage capacity; Equation (30) indicates that the stored energy of the battery should be equal at the beginning and end of the operating day before an extreme event.

[0077] c) Operational constraints of hydrogen storage systems: (31) (32) in, and The first i The hydrogen storage system in the first s In the scenario, the first t The power consumption and hydrogen production of the electrolyzer at each time point are given in kW and kg, respectively. Equation (31) gives the relationship between the power consumption and hydrogen production of the electrolyzer; Equation (32) indicates the upper and lower limits of the power consumption of the electrolyzer.

[0078] (33) (34) in, and The first i The hydrogen storage system in the first s In the scenario, the first t The hydrogen consumption and discharge power at any given time are in kg and kW. Equation (33) gives the relationship between the discharge power of the hydrogen fuel cell and the hydrogen consumption; Equation (34) represents the upper and lower limits of the discharge power of the hydrogen fuel cell.

[0079] (35) (36) (37) (38) in, For a 0-1 variable, This indicates that the hydrogen storage system is in a charging state. This indicates that the hydrogen storage system is in a discharge state; Indicates the first i The hydrogen storage system in the first s In the scenario, the first t The amount of hydrogen stored at any given time. Equation (35) indicates that the hydrogen storage system cannot be charged and discharged simultaneously at any given time; Equation (36) is used to calculate the amount of hydrogen stored at any given time within each typical day. t The amount of hydrogen stored in the hydrogen storage system during a given period; Equation (37) is used to calculate the amount of hydrogen stored in the hydrogen storage system during the period when extreme weather begins; Equation (38) represents the upper and lower limits of the amount of hydrogen stored in the hydrogen storage system.

[0080] d) Wind and solar power generation constraints: (39) (40) Equations (39)-(40) respectively restrict the first... s In the scenario, the first t The amount of wind and solar power curtailed at any given time should be less than the current output of wind farms and photovoltaic power plants.

[0081] e) Load shedding constraint: (41) Equation (41) restricts the first... s In the scenario, the first t The load shedding capacity should be less than the current load demand.

[0082] f) System power balance constraints: (42) S260. Solve the target scheduling model to obtain the target scheduling scheme for the scheduling date.

[0083] In a preferred implementation, pre-scheduling under changing weather conditions is performed through the following process: In this embodiment of the invention, an improved IEEE 24-bus system (a common standard power system test case system) is used to conduct case analysis, and the data is based on load and meteorological data from publicly available datasets.

[0084] First, the 3σ rule was used to detect outliers in the historical load data, thereby identifying anomalies. For detected outliers or missing values, Lagrange interpolation was used to impute them, ensuring the integrity and accuracy of the input data. After data cleaning and outlier processing, continuous and reliable meteorological and load time series data were obtained.

[0085] Figure 3 A flowchart illustrating the prediction process for temperature-sensitive loads under extreme weather conditions based on the SMOTE algorithm and bidirectional long short-term memory algorithm, provided for embodiments of the present invention. Figure 3 As shown, in the feature selection stage, temperature was used as the primary feature. Pearson correlation coefficients between temperature and various load types were calculated, and load types with |r|≥0.6 were selected as having strong temperature sensitivity and used as model input variables, while load types with weak correlations were removed. Based on the cold wave and high temperature standards of meteorological monitoring agencies, the annual weather data were classified to identify cold wave and high temperature days. Load data under these extreme weather conditions were combined with meteorological elements (such as temperature and humidity) to analyze their response characteristics under extreme weather conditions. Then, the SMOTE technique was used to augment 80% of the training set to increase the number of extreme weather samples and improve the model's predictive ability under such extreme weather scenarios. Finally, the augmented dataset was input into the Bi-LSTM prediction model for load prediction. Figure 4 The diagram shows the load prediction results before and after the cold wave transition weather corresponding to different prediction methods provided in the embodiments of the present invention. It includes the prediction results of the prediction method with SMOTE data augmentation combined with Bi-LSTM and the prediction method without considering data augmentation. Table 1 shows the error analysis of different prediction methods. It can be seen that the SMOTE+BiLSTM model shows better performance in the overall load prediction task and extreme transition weather scenarios.

[0086] Table 1 Figure 5A line graph showing the source-load output scenario considering temperature-sensitive load forecasting in the system is presented. Based on a typical output scenario, the pre-scheduling strategy for the electric-hydrogen energy storage system, which combines pre-scheduling before extreme weather and rescheduling under extreme weather conditions, provided by this invention, is used to obtain the system supply-demand balance results for a typical scenario before and under extreme weather conditions. Figure 6 This diagram illustrates the power system supply and demand balance before and under extreme weather conditions, taking into account pre-scheduling. Without considering pre-scheduling of the electricity-hydrogen system, the system supply and demand balance results for a typical scenario before and under extreme weather conditions are shown below. Figure 7 As shown, Figure 7 Table 2 shows the power system supply and demand balance results before and under extreme weather conditions, without considering pre-scheduling. It also presents the system operating costs corresponding to different strategies.

[0087] Table 2 It can be seen that the method provided by the present invention can enable the electric-hydrogen energy storage system to strategically pre-store energy before extreme weather and accurately discharge energy during extreme weather, thereby coping with the surge in temperature-sensitive loads and potential supply-demand mismatch risks caused by extreme weather.

[0088] The technical solution of this invention employs SMOTE data augmentation and the Bi-LSTM algorithm to conduct day-ahead forecasting of temperature-sensitive loads under extreme transitional weather conditions such as cold waves and high temperatures. Based on the load forecasting results under transitional weather conditions, typical operating scenarios of the power system that characterize load uncertainty are generated. Based on the load forecasting and its typical uncertainty scenarios, a two-stage pre-schedule model of the electric-hydrogen hybrid energy storage system is established. The established electric-hydrogen energy storage pre-schedule model is solved to obtain an optimized scheduling scheme for the electric-hydrogen energy storage system to cope with changes in the demand of temperature-sensitive loads under extreme transitional weather conditions. This solution can address the challenge of small sample data under extreme transitional weather conditions such as cold waves and high temperatures, and by combining the forecasting results with the pre-schedule of the electric-hydrogen hybrid energy storage system, it achieves proactive defense against supply and demand imbalances under extreme weather conditions by pre-charging before the arrival of extreme weather and accurately discharging during load surges, thereby effectively improving the resilience and power supply reliability of the power grid under extreme scenarios.

[0089] Figure 8 This is a schematic diagram of a pre-scheduling device for changing weather conditions, provided as an embodiment of the present invention. Figure 8 As shown, the device includes: The load forecasting module 810 is used to predict the temperature-sensitive load forecasting result corresponding to the scheduling date based on historical load data and historical meteorological data prior to the scheduling date, using a target load forecasting model. The power output acquisition module 820 is used to acquire the wind power output prediction results and photovoltaic power output prediction results corresponding to the scheduling date; The operating condition generation and modeling module 830 is used to generate a typical operating condition set for source-load joint operation based on the temperature-sensitive load prediction results, the wind power output prediction results, and the photovoltaic output prediction results, and to construct a target scheduling model based on the typical operating condition set for source-load joint operation. The model solving module 840 is used to solve the target scheduling model to obtain the target scheduling scheme for the scheduling date; wherein, the target scheduling model includes an objective function, pre-scheduling constraints, and rescheduling constraints.

[0090] The technical solution of this invention employs SMOTE data augmentation and the Bidirectional Long Short-Term Memory (BiLSTM) algorithm to conduct day-ahead forecasting of temperature-sensitive loads under extreme transitional weather conditions such as cold waves and high temperatures. Based on the load forecasting results under transitional weather conditions, typical operating scenarios of the power system that characterize load uncertainty are generated. Based on the load forecasting and its typical uncertainty scenarios, a two-stage pre-schedule model of the electric-hydrogen hybrid energy storage system is established. The established electric-hydrogen energy storage pre-schedule model is solved to obtain an optimized scheduling scheme for the electric-hydrogen energy storage system to cope with changes in the demand of temperature-sensitive loads under extreme transitional weather conditions. This solution can address the challenge of small sample data under extreme transitional weather conditions such as cold waves and high temperatures, and by combining the forecasting results with the pre-schedule of the electric-hydrogen hybrid energy storage system, it achieves proactive defense against supply and demand imbalances under extreme weather conditions by pre-charging before the arrival of extreme weather and accurately discharging during load surges, thereby effectively improving the resilience and power supply reliability of the power grid under extreme scenarios.

[0091] In some possible implementations, the load forecasting module 810 includes: The data acquisition and preprocessing submodule is used to acquire historical load data and historical meteorological data prior to the scheduling date, and to preprocess the historical load data and historical meteorological data to obtain meteorological and load time series data. The sensitive load screening submodule is used to screen temperature-sensitive loads based on the meteorological and load time series data and through correlation coefficients. The prediction model training submodule is used to train an initial load prediction model based on the load data of the temperature-sensitive load and the corresponding meteorological data, so as to obtain the target load prediction model. The load forecasting submodule is used to predict the temperature-sensitive load forecasting result corresponding to the scheduling date using the target load forecasting model.

[0092] In some possible implementations, the prediction model training submodule includes: The sample expansion unit is used to expand the load data and corresponding meteorological data of the temperature-sensitive load using a sample expansion algorithm to obtain expanded sample data. The model training unit is used to train the initial load prediction model based on the expanded sample data to obtain the target scheduling model.

[0093] In some possible implementations, the working condition generation and modeling module 830 includes: The initial operating condition generation submodule is used to generate multiple initial source-load joint operating conditions through random simulation based on the temperature-sensitive load prediction results, the wind power output prediction results, and the photovoltaic output prediction results. The working condition clustering submodule is used to perform clustering processing on multiple initial source-load joint operation conditions using a clustering algorithm to obtain a typical set of source-load joint operation conditions; The scheduling model construction submodule is used to construct the target scheduling model based on the typical operating conditions set of source-load joint operation.

[0094] In some possible implementations, the scheduling model construction submodule includes: The objective function construction unit is used to construct the objective function, which is used to minimize the total expected operating cost of the power system before and after the scheduling date. The total expected operating cost includes generator unit operating cost, wind and solar curtailment cost, and load shedding penalty cost. The constraint construction unit is used to construct the pre-scheduling constraints corresponding to the scheduling date and the rescheduling constraints under each typical working condition in the typical working condition set of source-load joint operation. The model integration unit is used to construct the target scheduling model based on the objective function, the pre-scheduling constraints, and the rescheduling constraints.

[0095] In some possible implementations, the pre-scheduling constraints constructed by the constraint construction unit include distributed generator set operation constraints, battery energy storage system operation constraints, hydrogen storage system operation constraints, wind power and photovoltaic power generation constraints, load shedding constraints, and system power balance constraints.

[0096] In some possible implementations, the rescheduling constraints constructed by the constraint construction unit include distributed generator set operation constraints, battery energy storage system operation constraints, hydrogen storage system operation constraints, wind power and photovoltaic power generation constraints, load shedding constraints, and system power balance constraints.

[0097] In some possible implementations, the model solving module 840 includes: The model solving submodule is used to solve the target scheduling model using a commercial solver; The scheduling scheme generation submodule is used to obtain the running parameters from the solution results of the commercial solver and form the target scheduling scheme for the scheduling date.

[0098] The pre-scheduling device for transitional weather provided in the embodiments of the present invention can execute the pre-scheduling method for transitional weather provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0099] Figure 9 This is a schematic diagram of an electronic device for implementing the pre-scheduling method under changing weather conditions according to embodiments of the present invention. The electronic device is intended to represent various forms of digital computers, such as laptops, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframes, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital assistants, cellular phones, smartphones, wearable devices (e.g., helmets, glasses, watches, etc.), and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the invention described and / or claimed herein.

[0100] like Figure 9 As shown, the electronic device 10 includes at least one processor 11 and a memory, such as a read-only memory (ROM) 12 or a random access memory (RAM) 13, communicatively connected to the at least one processor 11. The memory stores computer programs executable by the at least one processor. The processor 11 can perform various appropriate actions and processes based on the computer program stored in the ROM 12 or loaded into the RAM 13 from storage unit 18. The RAM 13 can also store various programs and data required for the operation of the electronic device 10. The processor 11, ROM 12, and RAM 13 are interconnected via a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0101] Multiple components in electronic device 10 are connected to I / O interface 15, including: input unit 16, such as keyboard, mouse, etc.; output unit 17, such as various types of displays, speakers, etc.; storage unit 18, such as disk, optical disk, etc.; and communication unit 19, such as network card, modem, wireless transceiver, etc. Communication unit 19 allows electronic device 10 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0102] Processor 11 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, digital signal processors (DSPs), and any suitable processor, controller, microcontroller, etc. Processor 11 performs the various methods and processes described above, such as pre-scheduling methods under changing weather conditions.

[0103] In some embodiments, the pre-scheduling method for transitional weather can be implemented as a computer program tangibly contained in a computer-readable storage medium, such as storage unit 18. In some embodiments, part or all of the computer program can be loaded and / or installed on electronic device 10 via ROM 12 and / or communication unit 19. When the computer program is loaded into RAM 13 and executed by processor 11, one or more steps of the pre-scheduling method for transitional weather described above can be performed. Alternatively, in other embodiments, processor 11 can be configured to perform the pre-scheduling method for transitional weather by any other suitable means (e.g., by means of firmware).

[0104] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations may include: implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0105] Computer programs used to implement the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when executed by the processor, the computer programs cause the functions / operations specified in the flowcharts and / or block diagrams to be performed. The computer programs may be executed entirely on a machine, partially on a machine, or as a standalone software package, partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0106] In the context of this invention, a computer-readable storage medium can be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution system, apparatus, or device. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination thereof. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0107] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the electronic device. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0108] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), or middleware components (e.g., application servers), or frontend components (e.g., user computers with graphical user interfaces or web browsers through which users can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., communication networks). Examples of communication networks include local area networks (LANs), wide area networks (WANs), blockchain networks, and the Internet.

[0109] A computing system can include clients and servers. Clients and servers are generally located far apart and typically interact through communication networks. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or cloud host, which is a hosting product within the cloud computing service system to address the shortcomings of traditional physical hosts and VPS services, such as high management difficulty and weak business scalability.

[0110] It should be understood that the various forms of processes shown above can be used, with steps reordered, added, or deleted. For example, the steps described in this invention can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution of this invention can be achieved, and this is not limited herein.

[0111] The specific embodiments described above do not constitute a limitation on the scope of protection of this invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this invention should be included within the scope of protection of this invention.

Claims

1. A pre-scheduling method under changing weather conditions, characterized in that, include: Based on historical load data and historical meteorological data prior to the scheduling date, the temperature-sensitive load forecast result corresponding to the scheduling date is predicted using the target load forecasting model. Obtain the wind power output prediction results and photovoltaic power output prediction results corresponding to the scheduling date; Based on the predicted results of temperature-sensitive loads, wind power output, and photovoltaic power output, a typical set of operating conditions for source-load joint operation is generated, and a target scheduling model is constructed based on the typical set of operating conditions for source-load joint operation. Solving the target scheduling model yields the target scheduling scheme for the specified scheduling date; wherein the target scheduling model includes an objective function, pre-scheduling constraints, and rescheduling constraints.

2. The method according to claim 1, characterized in that, The step of predicting the temperature-sensitive load forecast result corresponding to the scheduling date based on historical load data and historical meteorological data prior to the scheduling date using a target load forecasting model includes: Historical load data and historical meteorological data prior to the scheduling date are obtained, and the historical load data and historical meteorological data are preprocessed to obtain meteorological and load time series data; Based on the meteorological and load time series data, temperature-sensitive loads were identified by screening through correlation coefficients. The initial load prediction model is trained based on the load data of the temperature-sensitive load and the corresponding meteorological data to obtain the target load prediction model; The target load prediction model is used to predict the temperature-sensitive load for the scheduled date.

3. The method according to claim 2, characterized in that, The process of training an initial load prediction model based on the load data of the temperature-sensitive load and the corresponding meteorological data to obtain the target load prediction model includes: The load data and corresponding meteorological data of the temperature-sensitive load are augmented using a sample augmentation algorithm to obtain augmented sample data. The initial load prediction model is trained based on the expanded sample data to obtain the target load prediction model.

4. The method according to claim 1, characterized in that, Based on the temperature-sensitive load forecast results, the wind power output forecast results, and the photovoltaic power output forecast results, a typical set of source-load joint operation conditions is generated, including: Based on the predicted results of temperature-sensitive loads, wind power output, and photovoltaic power output, multiple initial source-load joint operation conditions are generated through random simulation. Clustering algorithms are used to cluster multiple initial source-load joint operation conditions to obtain a typical set of source-load joint operation conditions.

5. The method according to claim 1, characterized in that, The construction of the target scheduling model based on the typical operating condition set of source-load joint operation includes: Construct the objective function, the pre-scheduling constraints corresponding to the scheduling date, and the rescheduling constraints under each typical condition in the typical operating condition set of source-load joint operation. Based on the objective function, the pre-scheduling constraints, and the rescheduling constraints, the objective scheduling model is constructed. The objective function is used to minimize the total expected operating cost of the power system before and after the scheduling date. The total expected operating cost includes generator operating cost, wind and solar curtailment cost, and load shedding penalty cost.

6. The method according to claim 5, characterized in that, The pre-scheduling constraints include: Operating constraints of distributed generator sets, operating constraints of battery energy storage systems, operating constraints of hydrogen storage systems, power constraints of wind and photovoltaic power generation, load shedding constraints, and system power balance constraints.

7. The method according to claim 6, characterized in that, The rescheduling constraints include: Operating constraints of distributed generator sets, operating constraints of battery energy storage systems, operating constraints of hydrogen storage systems, power constraints of wind and photovoltaic power generation, load shedding constraints, and system power balance constraints.

8. The method according to claim 1, characterized in that, The process of solving the target scheduling model to obtain the target scheduling scheme for the scheduling date includes: The target scheduling model is solved using a commercial solver; The running parameters from the solution results of the commercial solver are obtained to form the target scheduling scheme for the scheduling date.

9. An electronic device, characterized in that, The electronic device includes: At least one processor; and a memory communicatively connected to the at least one processor; The memory stores a computer program that can be executed by the at least one processor, which enables the at least one processor to perform the pre-scheduling method for changing weather conditions as described in any one of claims 1-8.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions that cause a processor to execute the pre-scheduling method under changing weather conditions as described in any one of claims 1-8.

Images