A multi-objective power grid dispatching decision-making method and system

Abstract

This application discloses a multi-objective power grid dispatching decision-making method and system, relating to the field of power grid dispatching technology. The method includes: constructing a normal dispatching objective based on minimum load fluctuation; constructing an abnormal weather dispatching objective based on minimum dispatching quantity and minimum dispatching loss; constructing a device abnormality dispatching objective based on minimum switching loss; constructing a multi-objective optimization function by combining the normal dispatching objective, the abnormal weather dispatching objective, and the device abnormality dispatching objective with initial priority weights; obtaining probabilistic prediction values ​​based on regional historical load data, regional real-time load data, regional historical weather data, regional real-time weather data, regional historical device maintenance data, and regional real-time device data; obtaining the dispatching weights for each dispatching objective based on the probabilistic prediction values; and updating the multi-objective optimization function with the dispatching weights to update the multi-objective power grid dispatching decision. The beneficial effects of this application are: balancing scenario switching flexibility and power dispatching accuracy when facing complex real-world scenarios.

Description

Technical Field

[0001] This application relates to the field of power grid dispatching technology, and in particular to a power grid dispatching decision-making method and system oriented towards multiple objectives. Background Technology

[0002] With the large-scale integration of distributed energy resources, energy storage systems, and flexible loads, the traditional power grid operation mode is developing towards a direction of high proportion of new energy, multi-source integration, and regional autonomy. The current power grid dispatching system is gradually evolving from static control to dynamic optimization, while placing greater emphasis on the intelligence and multi-objective orientation of dispatching decisions, such as simultaneously considering economy, safety, green and low-carbon development, and operational stability.

[0003] In related technologies, a single scheduling objective is usually used to output scheduling decisions. Under normal operating conditions, scheduling decisions are mostly based on load balancing as the core objective, focusing only on the static matching of power grid supply and demand. When abnormal weather or equipment failure occurs, other scheduling decisions need to be invoked for adjustment, which is inefficient and can easily cause oscillations in key operating parameters such as power grid frequency and voltage, affecting power supply quality.

[0004] In related technologies, static statistical analysis is used for load fluctuations, weather anomalies, and equipment anomalies. This cannot adapt to the dynamically changing power grid, resulting in weak risk prediction capabilities, poor accuracy of prediction results on which power grid dispatch relies, and low reliability of power grid dispatch. Summary of the Invention

[0005] This application addresses the problem that existing technologies for power grid dispatching are limited to single-objective scenarios, making it difficult to adapt to complex real-world scenarios in terms of scenario switching flexibility and power dispatching accuracy. It provides a multi-objective power grid dispatching decision-making method and system. This method constructs multi-objective functions for normal dispatching states, abnormal weather dispatching states, and abnormal device dispatching states, and adjusts and optimizes priorities based on the risk probability under different scenarios. This results in a unified output decision that can satisfy dispatching in the current high-risk scenario and can also be used as a benchmark for dispatching adjustments after risk changes in other scenarios. This improves the adaptability of dispatching decisions, reduces the dispatching complexity of switching between complex scenarios, and increases dispatching efficiency.

[0006] To achieve the above technical objectives, this application provides a technical solution: a multi-objective power grid dispatching decision-making method, comprising the following steps: constructing a normal dispatching objective based on minimum load fluctuation; constructing an abnormal weather dispatching objective based on minimum dispatching quantity and minimum dispatching loss; constructing a device abnormality dispatching objective based on minimum switching loss; constructing a multi-objective optimization function by combining the normal dispatching objective, the abnormal weather dispatching objective, and the device abnormality dispatching objective with initial priority weights; obtaining the current load fluctuation probability prediction value based on the fluctuation matching of regional historical load data and regional real-time load data; obtaining the current weather abnormality probability prediction value based on the matching of regional historical weather data and regional real-time weather data; obtaining the current device abnormality probability prediction value based on the matching of regional historical device maintenance data and regional real-time device data; obtaining the dispatching weights of the normal dispatching objective, the abnormal weather dispatching objective, and the device abnormality dispatching objective based on the current load fluctuation probability prediction value, the current weather abnormality probability prediction value, and the current device abnormality probability prediction value, respectively; correcting the initial priority weights with the dispatching weights, updating the multi-objective optimization function; obtaining a multi-objective power grid dispatching decision based on the updated multi-objective optimization function and regional real-time data; and executing power grid dispatching based on the multi-objective power grid dispatching decision.

[0007] Furthermore, the construction of the abnormal weather scheduling target based on minimum scheduling quantity and minimum scheduling loss includes: calculating the scheduling quantity by multiplying the negative value of the number of times distributed energy resources are scheduled by the scheduling frequency; calculating the scheduling loss by calculating the scheduling loss of distributed energy resources and the disaster resistance loss of retained distributed energy resources; and constructing the constraints of the abnormal weather scheduling target based on the line safety margin threshold and the renewable energy curtailment threshold.

[0008] Furthermore, the method of constructing an abnormal weather scheduling target with minimum scheduling amount and minimum scheduling loss also includes: setting scheduling amount weight and scheduling loss weight based on the difference between the average scheduling frequency and the median scheduling frequency of distributed energy resources in the actual area; and constructing the abnormal weather scheduling target with scheduling amount weight, scheduling loss weight, scheduling amount, and scheduling loss.

[0009] Furthermore, the step of constructing an abnormal device scheduling target with minimum switching loss includes: obtaining a first switching loss based on the line difference loss between the abnormal distributed energy resource device and the distributed energy resource device to be switched; obtaining a second switching loss based on the transient loss of the distributed energy resource device switching; and calculating the switching loss using the first switching loss and the second switching loss.

[0010] Furthermore, the step of obtaining the current load fluctuation probability prediction value by matching the fluctuations of regional historical load data and regional real-time load data includes: obtaining the previous load data of a preset preceding time period in the region; calculating the similarity between the previous load data and the historical load data of the region using a similarity algorithm; constructing a prediction trend set using historical load data with a similarity greater than a preset similarity threshold; obtaining the deviation ratio by comparing the regional real-time load data with the prediction trend set; and obtaining the current load fluctuation probability prediction value using the deviation ratio and the future load fluctuation situation in the prediction trend.

[0011] Furthermore, the step of obtaining the current weather anomaly probability prediction value based on the matching of regional historical weather data and regional real-time weather data includes: obtaining regional real-time weather data, filtering matching historical weather data that match the regional real-time weather data in the same time series, and obtaining the current weather anomaly probability prediction value based on the proportion of abnormal weather within a preset period in all matching historical weather data.

[0012] Furthermore, the step of obtaining the current device anomaly probability prediction value by matching regional historical device maintenance data with regional real-time device data includes: constructing device maintenance models for each device type based on the correlation between device parameters and maintenance time sequence in regional historical device maintenance data; obtaining the remaining lifespan using regional real-time device data and the corresponding device maintenance model; and obtaining the current device anomaly probability prediction value by using the ratio of the remaining lifespan to the initial lifespan.

[0013] Furthermore, the step of correcting the initial priority weights with scheduling weights, updating the multi-objective optimization function, and obtaining multi-objective power grid dispatching decisions based on the updated multi-objective optimization function and regional real-time data includes: obtaining special scheduling relationships based on human-caused scheduling data and weather anomaly types under historical abnormal weather conditions in the region; obtaining the current weather anomaly type by matching regional historical weather data with regional real-time weather data; obtaining special scheduling correction weights based on the special scheduling relationships using the current weather anomaly type; correcting the scheduling weights with the special scheduling correction weights; updating the multi-objective optimization function with the corrected scheduling weights; and obtaining multi-objective power grid dispatching decisions based on the updated multi-objective optimization function and regional real-time data.

[0014] Furthermore, the construction of a normal scheduling target based on minimum load fluctuation includes: calculating load fluctuation using the standard deviation of load fluctuations from regional real-time load data and regional average load data; and constructing constraints for the normal scheduling target based on regional equipment power balance and equipment capacity.

[0015] Another technical solution provided in this application is a multi-objective power grid dispatching decision system for implementing the method described above, comprising: a storage layer for receiving and storing regional load data, weather data, maintenance data, and device data; a development layer for storing the construction logic of normal dispatching objectives, abnormal weather dispatching objectives, device abnormality dispatching objectives, and dispatching weights; and an application layer for calling the data in the storage layer and the construction logic in the development layer to perform power grid dispatching decision calculations.

[0016] The beneficial effects of this application are as follows: 1. The scheduling objectives under normal conditions, under abnormal weather conditions, and under equipment failure conditions are pre-constructed. The scheduling weights of each objective are adjusted by the probability of load fluctuation in the current area, the probability of abnormal weather, and the probability of equipment failure. This adjusts the priority of each objective under different scenarios, so that the power grid scheduling decision can not only meet the maximum demand of the current scenario, but also ensure the adjustment efficiency when other demands occur, thereby improving the global adaptability of power grid scheduling.

[0017] 2. By assigning higher priority to distributed energy resources that are frequently dispatched, and taking advantage of the fact that distributed energy resources that are dispatched more frequently usually have better response mechanisms and higher stability, the reliability and stability of dispatching under abnormal weather conditions can be improved, thus avoiding the failure caused by the excessive fragility of distributed energy resources that are dispatched less frequently.

[0018] 3. By fusing multiple trends, prediction biases caused by random deviations of a single trend are avoided. At the same time, the reliability of predictions is further enhanced by credibility. In extreme cases, when the deviation between the regional real-time load data and each prediction trend is large, the overall credibility is dispersed, and the prediction trend with large fluctuations will be more prominent, which can avoid misjudgment of fluctuations in extreme cases.

[0019] 4. By calculating the ratio of remaining life to initial life, the reliability of the current remaining life can be reflected. The closer the device is to the maintenance schedule, the lower the reliability and the more likely it is to fail. Therefore, the probability of failure can be judged based on the ratio of remaining life to initial life. Attached Figure Description

[0020] Figure 1 This is a flowchart illustrating a multi-objective power grid dispatching decision-making method according to this application.

[0021] Figure 2 This is a schematic diagram of the load fluctuation probability prediction process in a multi-objective power grid dispatch decision-making method according to this application. Detailed Implementation

[0022] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description of this application is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely one preferred embodiment of this application and are only used to explain this application. They do not limit the scope of protection of this application. All other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0023] like Figure 1 As shown in the first embodiment of this application, a multi-objective power grid dispatching decision-making method includes the following steps:

[0024] The normal scheduling objective is constructed with the minimum load fluctuation, the abnormal weather scheduling objective is constructed with the minimum scheduling amount and the minimum scheduling loss, the abnormal device scheduling objective is constructed with the minimum switching loss, and the multi-objective optimization function is constructed by combining the normal scheduling objective, the abnormal weather scheduling objective and the abnormal device scheduling objective with the initial priority weight.

[0025] The current load fluctuation probability prediction value is obtained by matching the fluctuation of regional historical load data with regional real-time load data; the current weather anomaly probability prediction value is obtained by matching regional historical weather data with regional real-time weather data; and the current device anomaly probability prediction value is obtained by matching regional historical device maintenance data with regional real-time device data.

[0026] The scheduling weights for normal scheduling targets, abnormal weather scheduling targets, and equipment abnormality scheduling targets are obtained based on the current load fluctuation probability prediction value, the current weather abnormality probability prediction value, and the current equipment abnormality probability prediction value, respectively.

[0027] The initial priority weights are corrected by adjusting the scheduling weights, the multi-objective optimization function is updated, and the updated multi-objective optimization function and real-time regional data are used to obtain multi-objective power grid scheduling decisions. Power grid scheduling is then executed based on the multi-objective power grid scheduling decisions.

[0028] In this embodiment, a multi-objective optimization function is pre-constructed based on initial priority weights. This function is applied to scheduling scenarios under normal conditions, abnormal weather conditions, and equipment malfunctions. The initial priority weights guide the final grid dispatch decision towards higher-priority scenarios without compromising its adaptability to other lower-priority scenarios. Furthermore, the dispatch weights of each objective are adjusted based on the current regional load fluctuation probability, abnormal weather probability, and equipment malfunction probability. These dispatch weights are then used to update the initial priority weights, making the grid dispatch decision more suitable for the current high-risk scenario while maintaining adaptability to switching to other scenarios. This ensures the accuracy of power dispatch and the flexibility to switch between complex scenarios.

[0029] Specifically, establishing normal scheduling objectives with minimal load fluctuations includes:

[0030] Load fluctuation is calculated using the standard deviation of load fluctuations from regional real-time load data and regional average load data.

[0031] Constraints for normal scheduling objectives are constructed based on regional equipment power balance and equipment capacity.

[0032] In normal scenarios, power grid dispatching needs to prioritize smoothing out load peak-valley differences and minimize frequent unit start-ups and shutdowns or repeated charging and discharging of energy storage caused by sudden load changes. Therefore, obtaining real-time regional load data can be achieved by connecting to the distribution network SCADA system and user electricity consumption acquisition terminals via a data access adapter module to obtain both real-time and historical regional load data. The regional average load data can then be obtained based on the historical load data that matches the real-time regional load data.

[0033] The normal scheduling objective is:

[0034] ;

[0035] in, This represents the normal scheduling objective function. This represents the real-time load data for the region at time t. This represents the regional average load data.

[0036] Since power grid dispatching typically needs to satisfy load balance within a dispatching cycle, therefore, within a dispatching cycle... For example, the normal scheduling target is:

[0037] ;

[0038] in, express The normal scheduling objective function for a periodic period, Indicates the relationship between regional historical load data and Regional average load data with the same cycle.

[0039] Meanwhile, regional power balance constraints are constructed based on the principle that the input power and output power within the region are the same, and regional equipment capacity constraints are constructed based on the upper and lower limits of generator output and the upper and lower limits of energy storage system storage.

[0040] The objectives for scheduling abnormal weather conditions, based on minimizing scheduling volume and minimizing scheduling loss, include:

[0041] The scheduling quantity is calculated by multiplying the number of times distributed energy resources are scheduled by the negative value of the scheduling frequency.

[0042] Dispatch losses are calculated based on the dispatch losses of distributed energy resources and the disaster resistance losses of retained distributed energy resources.

[0043] Constraints for abnormal weather dispatch targets are constructed using line safety margin thresholds and renewable energy curtailment thresholds.

[0044] In abnormal weather conditions, the output of renewable energy may drop sharply due to weather interference. Therefore, it is necessary to reduce unnecessary dispatch operations, but at the same time, it is necessary to avoid damage to grid equipment under abnormal weather conditions. For example, in typhoon weather, it is necessary to control the wind turbine to adjust the blade angle to avoid damage to the blade, but at the same time, it is necessary to ensure the power supply as much as possible.

[0045] Specifically, the objectives for managing abnormal weather conditions are:

[0046] ;

[0047] in, This represents the objective function for scheduling during abnormal weather. Indicates the first Number of times distributed energy resource scheduling is performed. This represents the total number of distributed energy resources. Indicates the first The truth value of distributed energy resource scheduling Indicates the first Frequency of distributed energy resource dispatch Indicates the scheduling quantity weight. Indicates the scheduling loss weight. This indicates the loss in the dispatching of distributed energy resources. This indicates that distributed energy resources should be retained to mitigate disaster losses. .

[0048] When the When the state of a distributed energy resource changes, it is considered that the distributed energy resource has been dispatched. When the first If the state of a distributed energy resource remains unchanged, it is considered that the distributed energy resource has not been dispatched. Therefore, when distributed energy resources are dispatched, the number of dispatches and the frequency of dispatch of the distributed energy resources are calculated simultaneously, as well as the dispatch losses of the dispatched distributed energy resources and the disaster resistance losses of the undispatched distributed energy resources.

[0049] It is understandable that Distributed Energy Resources (DER) include distributed renewable energy, energy storage systems, and controllable loads.

[0050] Under abnormal weather conditions, every scheduling change can potentially damage distributed energy resources or trigger equipment failures already at their limits. Therefore, it's crucial to minimize the number of equipment scheduling events during abnormal weather. Furthermore, the scheduling quantity is calculated by considering the negative value of scheduling frequency; more frequently scheduled distributed energy resources will have lower scheduling quantities, making them more likely to be scheduled. In this embodiment, frequently scheduled distributed energy resources are given higher priority for dispatch. This leverages the generally higher response mechanisms and stability of frequently scheduled distributed energy resources, ensuring reliability and stability during abnormal weather conditions and preventing failures caused by the vulnerability of low-frequency scheduled distributed energy resources. Simultaneously, a higher scheduling frequency corresponds to a lower scheduling quantity. Using the scheduling frequency to reflect the single scheduling quantity of distributed energy resources avoids grid fluctuations caused by excessively high single scheduling power, ensuring grid stability.

[0051] In this embodiment, the loss of distributed energy resources under abnormal weather conditions is divided into distributed energy resource scheduling loss and retained distributed energy resource disaster resistance loss. Distributed energy resource scheduling loss is obtained based on the state change loss of scheduled distributed energy resources and the disaster loss after scheduling. Retained distributed energy resource disaster resistance loss is obtained based on the disaster loss of unscheduled distributed energy resources. A disaster loss model for each type of distributed energy resource can be built based on the historical disaster situation of distributed energy resources under various states during abnormal weather conditions. When calculating scheduling losses, the corresponding disaster loss data in the disaster loss model is retrieved.

[0052] By dividing the loss of distributed energy resources under abnormal weather conditions into the loss of distributed energy resource scheduling and the loss of retained distributed energy resources for disaster resistance, the calculation of losses is improved by avoiding the neglect of inactive distributed energy resources when calculating the loss.

[0053] The line safety margin threshold is set at the load that the line can retain for disaster relief, so as to avoid the line being directly overloaded during a disaster. The acceptable curtailment value of renewable energy is set at the threshold for curtailment of renewable energy, so as to avoid excessive waste of resources when avoiding abnormal weather.

[0054] In this embodiment, the scheduling quantity weight and scheduling loss weight are set based on the difference between the average scheduling frequency and the median scheduling frequency of distributed energy resources in the actual area. The larger the difference, the larger the scheduling quantity weight; the smaller the difference, the larger the scheduling loss weight. The difference between the average scheduling frequency and the median scheduling frequency reflects the uneven distribution of scheduling frequency. When the difference is large, it indicates that a few distributed energy resources are scheduled more frequently, while most are scheduled less frequently. In this case, the scheduling quantity weight is larger, prioritizing the scheduling of distributed energy resources with high frequency of scheduling to avoid resource overload or poor scheduling effect caused by scheduling too many distributed energy resources that are not frequently adjusted. When the difference is small, it indicates that the distributed energy resources in the current area have high adjustability. In this case, the scheduling loss weight is increased to reduce the loss generated during the scheduling process. It can be understood that in some cases, the ratio of the difference to the average scheduling frequency can be directly used as the scheduling quantity weight.

[0055] The objectives for constructing device anomaly scheduling with minimum switching loss include:

[0056] The first switching loss is obtained based on the line difference loss between the abnormal distributed energy resource device and the distributed energy resource device to be switched.

[0057] The second switching loss is obtained based on the transient loss of the switching of distributed energy resource devices;

[0058] The switching loss is calculated using the first switching loss and the second switching loss.

[0059] In the event of an anomaly in a distributed energy resource device, it is necessary to ensure normal power supply through device switching. Therefore, it is necessary to minimize the additional losses during the switching process.

[0060] Specifically, the device anomaly scheduling objective is:

[0061] ;

[0062] in, This represents the objective function for abnormal device scheduling. This represents the total number of distributed energy resource devices. Indicates the first The line length of each distributed energy resource device Indicates the corresponding to the first The line length of the distributed energy resource device to be switched is [length missing]. This represents the line loss per unit length. Indicates the corresponding to the first The transient switching loss of the distributed energy resource device to be switched.

[0063] Understandably, line losses often differ little within a given area, so the change in line length before and after the switch is used to obtain the losses caused by the switch. The transient switching losses of the distributed energy resource devices to be switched can be simulated and pre-stored using transient simulation software, and then retrieved when calculating the switching losses.

[0064] In this embodiment, constructing the device anomaly scheduling objective with minimum switching loss further includes:

[0065] The constraints for abnormal device scheduling targets are constructed using preset switching times and the capacity of the distributed energy resource devices to be switched.

[0066] According to safety standards, the preset switching time is set to 3 minutes. In practical applications, the preset switching time can also be set according to actual needs.

[0067] The multi-objective optimization function, which combines the normal scheduling objective, the abnormal weather scheduling objective, and the abnormal equipment scheduling objective with initial priority weights, includes:

[0068] The initial priority weights are constructed based on the ratio of historical abnormal weather days, historical device abnormal days, and normal days.

[0069] In this embodiment, a year is used as a calculation cycle. The initial priority weight for the current year is obtained based on the ratio of the number of days with abnormal weather, the number of days with abnormal equipment, and the number of normal days to the total number of days in the previous year. This initial priority weight is then used to calculate and output the current initial power grid dispatch decision. It is understood that in other cases, the initial priority weight can also be set based on expert experience.

[0070] like Figure 2 As shown, the probability prediction value of current load fluctuation is obtained by matching the fluctuations of historical load data and real-time load data of the region, including:

[0071] Obtain the preceding load data of the region within a preset preceding time period, calculate the similarity between the preceding load data and the historical load data of the region using a similarity algorithm, and construct a prediction trend set using historical load data with similarity greater than a preset similarity threshold;

[0072] The deviation ratio is obtained by comparing the real-time regional load data with the predicted trend set.

[0073] The predicted probability of current load fluctuation is obtained by using the deviation ratio and the future load fluctuation situation in the forecast trend.

[0074] In this embodiment, the preset preceding time period is 24 hours, meaning that the load data from the 24 hours prior to the current time series is used as the preceding load data. The similarity between the preceding load data and historical load data can be calculated using a dynamic time warping algorithm or a cosine similarity algorithm. The historical load data curve with a similarity greater than a preset similarity threshold is used as the prediction trend. The regional load is acquired in real time, and the load fluctuation probability prediction value is calculated based on the deviation ratio between the real-time regional load and the predicted values ​​in each prediction trend.

[0075] Specifically, the deviation ratio is obtained by comparing the real-time regional load data with the predicted trend set, including:

[0076] Obtain the predicted load data corresponding to the time series of real-time load data for each predicted trend in the predicted trend set and the region;

[0077] The deviation ratio is obtained by comparing the difference between each predicted load data and the regional real-time load data with the ratio of the regional real-time load data.

[0078] The predicted probability value of current load fluctuation is obtained by using the deviation ratio and the future load fluctuation situation in the forecast trend, including:

[0079] The reliability of each forecast trend is obtained by the deviation ratio of each forecast trend, and the standard deviation of future load fluctuation within the preset period of each forecast trend is calculated.

[0080] The overall load fluctuation standard deviation is calculated based on the reliability of all predicted trends and the standard deviation of future load fluctuations. The current load fluctuation probability prediction value is obtained by using the ratio of the overall load fluctuation standard deviation to the regional historical maximum load fluctuation standard deviation.

[0081] In this embodiment, load forecasting for the current time point is performed based on load curves exhibiting the same trend as those in previous time periods. The deviation ratio is obtained by comparing the predicted value with the real-time load data of the region. The smaller the deviation ratio, the closer the predicted trend is to the actual trend, and the higher the reliability. The reliability of each predicted trend can be obtained using an exponential decay function and a normalization algorithm. Furthermore, the standard deviation of load fluctuation for each predicted trend within a preset period is calculated. In this embodiment, the preset period is 24 hours. Using reliability as a weight, the future load fluctuation standard deviations of all predicted trends are weighted and summed to obtain the overall load fluctuation standard deviation. The ratio of the overall load fluctuation standard deviation to the region's historical maximum load fluctuation standard deviation is used as the predicted probability value of the current load fluctuation. This multi-trend fusion calculation avoids prediction deviations caused by random deviations from a single trend. Simultaneously, reliability further enhances prediction reliability. In extreme cases where the deviation between the real-time load data of the region and each predicted trend is large, the overall reliability is dispersed, making the more volatile predicted trends more prominent, thus avoiding misjudgments of fluctuations in extreme situations.

[0082] Among them, the standard deviation of the region's historical maximum load fluctuation reflects the fluctuation state under the most extreme historical conditions. By comparing the current overall load fluctuation standard deviation with the regional historical maximum load fluctuation standard deviation, the current risk position is reflected. If the current overall load fluctuation standard deviation is much lower than the regional historical maximum load fluctuation standard deviation, then even if fluctuations exist, they are more consistent with the stable load and low fluctuation situation of a normal working day. At this time, the fluctuations do not require too much targeted scheduling, so they can be considered as low fluctuation probability.

[0083] The predicted probability of current weather anomalies is obtained by matching historical weather data with real-time weather data for the region, including:

[0084] Obtain real-time weather data for the region, filter historical weather data that match the real-time weather data for the same time period in the region's historical weather data, and obtain the current weather anomaly probability prediction value based on the proportion of abnormal weather within a preset period in all matched historical weather data.

[0085] In this embodiment, the weather data includes at least temperature and humidity. Historical weather data that is matched with the real-time regional weather data in terms of time sequence and weather data matching is used as the historical weather data for matching. The probability of future weather anomalies is determined by the proportion of abnormal weather within a preset period in the historical weather data.

[0086] The predicted value of the current device anomaly probability is obtained by matching historical device maintenance data with real-time device data in the region, including:

[0087] Based on the correlation between device parameters and maintenance sequence in the historical device maintenance data of the region, device maintenance models for each device type are constructed;

[0088] The remaining lifespan is obtained by using real-time regional device data and the corresponding device maintenance model, and the current device anomaly probability prediction value is obtained by using the ratio of the remaining lifespan to the initial lifespan.

[0089] In this embodiment, historical device maintenance data includes at least device type, device maintenance sequence, and device maintenance cycle parameters. Regional real-time device data includes at least real-time device parameters. Device maintenance models corresponding to each device type are constructed based on the device maintenance sequence and device maintenance cycle parameters. The remaining lifespan of the device is obtained based on the device maintenance model under the regional real-time device data conditions. The predicted probability of current device anomaly is obtained by using the ratio of the remaining lifespan to the initial lifespan when the device is not in use. The calculation of the ratio of remaining lifespan to initial lifespan reflects the reliability of the current remaining lifespan. The closer the device is to the maintenance sequence, the lower its reliability and the more likely it is to fail. Therefore, the failure probability is determined based on the ratio of remaining lifespan to initial lifespan.

[0090] The scheduling weights for normal scheduling targets, abnormal weather scheduling targets, and equipment abnormality scheduling targets are obtained based on the current load fluctuation probability prediction, the current weather abnormality probability prediction, and the current equipment abnormality probability prediction, respectively.

[0091] The current load fluctuation probability prediction value, the current weather abnormality probability prediction value, and the current equipment abnormality probability prediction value are normalized to obtain the scheduling weights of normal scheduling targets, abnormal weather scheduling targets, and equipment abnormality scheduling targets. The sum of the scheduling weights of normal scheduling targets, abnormal weather scheduling targets, and equipment abnormality scheduling targets is 1.

[0092] In this embodiment, the scheduling priority in multi-objective optimization is determined by the probability of each risk occurring, ensuring that the scheduling requirements of high-risk scenarios can be met, and reserving adjustment margins for possible occurrence of other risk scenarios, so that adjustments can be made quickly when other risk scenarios occur.

[0093] Specifically, the initial priority weights are corrected using scheduling weights, the multi-objective optimization function is updated, and multi-objective power grid scheduling decisions are obtained using the updated multi-objective optimization function and real-time regional data, including:

[0094] The initial priority weights are replaced with scheduling weights, the multi-objective optimization function is updated, and the calculation is iteratively performed based on the updated multi-objective optimization function and real-time regional data until the minimum weighted sum and the corresponding power grid scheduling decision are obtained.

[0095] In the power grid dispatch decision-making process based on the initial priority weights, real-time regional load data, real-time weather data, and real-time device data are acquired. The probability of potential risks in different future scenarios is calculated, and dispatch weights are obtained based on the probabilities. The initial priority weights are then adjusted. The optimal dispatch scheme under multi-objective conditions is selected by minimizing the weighted sum of each objective function value and its corresponding dispatch weight. This eliminates the need for complex single-scenario calculations and comprehensive solutions, improving dispatch efficiency. Simultaneously, the optimization priority of each objective is determined by the dispatch weights, ensuring that the power grid dispatch decision is more adaptable to the current high-risk scenario without discarding other scenarios. This multi-objective optimization approach ensures that the current power grid dispatch decision is not only adaptable to the current high-risk scenario but also to other scenarios. When the risk value of other scenarios increases, adjustments can be made based on the current power grid dispatch decision, reducing dispatch during risk changes and improving power grid stability.

[0096] In this scenario, the regional real-time data includes at least regional real-time load data, regional real-time weather data, and regional real-time device data. The process involves revising the initial priority weights using scheduling weights, updating the multi-objective optimization function, and then using the updated multi-objective optimization function and regional real-time data to obtain multi-objective power grid dispatch decisions. This also includes:

[0097] The current power grid dispatch decision is used as the basis for iteration. The dispatch weights corresponding to the current power grid dispatch decision are used as the initial priority weight combination. The latest dispatch weights are calculated in real time. The initial priority weights are corrected with the latest dispatch weights, and the multi-objective optimization function is updated. The power grid dispatch decision method is updated iteratively based on the updated multi-objective optimization function.

[0098] At this point, each update of the scheduling weight is based on the original power grid scheduling decision. Since the original power grid scheduling decision itself meets the three risk scenarios, even if the risk probability of one of the risk scenarios increases, the device used to compensate for the risk in the original power grid scheduling decision can remain unchanged, and the scheduling of the other devices can be carried out. This reduces the number of device schedulings during the risk scenario switching process, improves scheduling efficiency, and can also compensate for the efficiency reduction caused by multiple schedulings of high-frequency scheduling devices in abnormal weather scenarios, while taking into account both reliability and efficiency.

[0099] In other cases, adjusting the initial priority weights with scheduling weights, updating the multi-objective optimization function, and then using the updated multi-objective optimization function and real-time regional data to obtain multi-objective power grid scheduling decisions also include:

[0100] Special scheduling relationships are obtained based on historical human scheduling data under abnormal weather conditions in the region and the types of weather anomalies.

[0101] The current weather anomaly type is obtained by matching regional historical weather data with regional real-time weather data;

[0102] Based on the current weather anomaly type, special scheduling correction weights are obtained according to special scheduling correlations;

[0103] The scheduling weights are corrected using special scheduling correction weights, and the multi-objective optimization function is updated with the corrected scheduling weights. The updated multi-objective optimization function and regional real-time data are used to obtain multi-objective power grid scheduling decisions.

[0104] In this situation, considering that the harm caused by power outages may be greater than equipment damage under special abnormal weather conditions, such as the need to ensure power supply under extreme cold weather, the scheduling weights under special weather conditions are adjusted based on the regional historical data on human scheduling under abnormal weather conditions. The scheduling weights of abnormal weather scheduling targets and abnormal equipment scheduling targets are reduced, while the scheduling weights of normal scheduling targets are increased, to ensure that the scheduling decisions meet the needs under actual conditions.

[0105] Specifically, special scheduling correlations are obtained based on historical human scheduling data during abnormal weather events in the region and the types of weather anomalies, including:

[0106] Obtain the scheduling type, corresponding weather anomaly type, weather range, and distance between the scheduling device and the extreme weather center from the human-managed scheduling data. Construct a special scheduling relationship based on the distance range between the scheduling device and the extreme weather center for each scheduling type under different weather ranges for each weather anomaly type.

[0107] In this embodiment, the distance range between the scheduling device corresponding to the same scheduling type and the extreme weather center is used as the correlation benchmark. The number of scheduling devices and the number of unscheduled devices within this distance range are used to obtain a special scheduling correction ratio. The corrected scheduling weight is obtained by multiplying the special scheduling correction ratio by the scheduling weight. If there is a probability of weather anomalies in the current area, the current weather anomaly type and the current extreme weather center are obtained by matching the area's historical weather data and real-time weather data. The extreme weather center can be obtained based on the center of the area most severely affected by the abnormal weather; for example, in extremely cold weather, the center of the lowest temperature is used as the extreme weather center. The area to be corrected is obtained based on the distance range between the extreme weather center and the special scheduling correlation. The scheduling weight of all devices in the area to be corrected is adjusted according to the special scheduling correction ratio corresponding to the scheduling type, reducing the number of devices shut down for disaster avoidance within the extreme weather range, ensuring the basic power supply needs of the current area, and avoiding power imbalance caused by excessive disaster avoidance, which could lead to more severe disaster losses.

[0108] As a second embodiment of this application, a multi-objective power grid dispatching decision-making system includes:

[0109] The storage layer is used to receive and store regional load data, weather data, maintenance data, and equipment data;

[0110] The development layer is used to store the construction logic for normal scheduling targets, abnormal weather scheduling targets, device abnormality scheduling targets, and scheduling weights;

[0111] The application layer is used to call data in the storage layer and the construction logic in the development layer to perform power grid dispatching decision calculations.

[0112] The construction logic for normal scheduling targets, abnormal weather scheduling targets, equipment abnormality scheduling targets, and scheduling weights includes at least the following:

[0113] The normal scheduling objective is constructed with the minimum load fluctuation, the abnormal weather scheduling objective is constructed with the minimum scheduling amount and the minimum scheduling loss, the abnormal device scheduling objective is constructed with the minimum switching loss, and the multi-objective optimization function is constructed by combining the normal scheduling objective, the abnormal weather scheduling objective and the abnormal device scheduling objective with the initial priority weight.

[0114] The current load fluctuation probability prediction value is obtained by matching the fluctuation of regional historical load data with regional real-time load data; the current weather anomaly probability prediction value is obtained by matching regional historical weather data with regional real-time weather data; and the current device anomaly probability prediction value is obtained by matching regional historical device maintenance data with regional real-time device data.

[0115] The scheduling weights of normal scheduling targets, abnormal weather scheduling targets, and equipment abnormality scheduling targets are obtained based on the current load fluctuation probability prediction value, the current weather abnormality probability prediction value, and the current equipment abnormality probability prediction value.

[0116] Specifically, the storage layer receives real-time data via Kafka / RabbitMQ. Kafka and RabbitMQ form a message queue system to enable asynchronous communication, task triggering, and event-driven mechanisms across modules. Historical data is synchronized with the power grid system via an interface. The storage layer utilizes a relational database (RDS / DM) to store historical load, weather, and equipment maintenance data, and Redis to store real-time load, weather, and equipment data. It also temporarily stores probability predictions, scheduling weights, and decision results. Data is uniformly orchestrated through a metadata management service to ensure the application layer can quickly retrieve the required data.

[0117] In the development layer, Python is used to write calculation models for minimum load fluctuation, minimum scheduling amount, minimum scheduling loss, and minimum switching loss, as well as mapping algorithms between probabilistic prediction values ​​and scheduling weights. The algorithms are then encapsulated into standardized interfaces in Java for application layer to call.

[0118] In the application layer, the system calls the initial priority weights, regional historical load data, regional real-time load data, regional historical weather data, regional real-time weather data, regional historical device maintenance data, and regional real-time device data from the storage layer. It then calls the probability calculation interface from the development layer to calculate and store the predicted probabilities of current load fluctuations, current weather anomalies, and current device anomalies. Next, it calls the probability predictions from the storage layer and the weight calculation interface from the development layer to calculate and store the scheduling weights for normal scheduling objectives, abnormal weather scheduling objectives, and device anomaly scheduling objectives. Finally, it calls the weight results from the storage layer and the multi-objective decision-making interface from the development layer to integrate the three scheduling objectives, calculate the power grid scheduling decision, store it in the storage layer, and output it to the execution end.

[0119] In other cases, a multi-objective power grid dispatching decision-making system also includes:

[0120] The deployment layer is used to encapsulate the build logic in the development layer in Docker containers, so as to provide a containerized runtime environment for the application layer.

[0121] In the deployment layer, a distributed containerized deployment approach is adopted. The front-end service is handled by Nginx for static resource hosting and access routing reverse proxy, while the back-end service is based on Tomcat to deploy various microservice components, supporting service-level load balancing and hot deployment. Different build logics in the development layer are encapsulated into independent Docker containers to ensure the isolation of the runtime environment of each logical module and avoid dependency conflicts.

[0122] In this embodiment, a storage layer, development layer, application layer, and deployment layer are constructed based on the Java EE architecture. The development layer builds service components based on Spring Boot and Spring Cloud, supports mixed development using languages ​​such as Java, Python, and C++, and uses Vue and HTML to build the front-end interface. The deployment layer includes an Nginx front-end service, a Tomcat back-end service, and an algorithm container environment based on Docker, supporting online calls to multiple models and isolated operation in heterogeneous environments.

[0123] The specific embodiments described above are preferred embodiments of a multi-objective power grid dispatch decision-making method and system of this application, and are not intended to limit the specific implementation scope of this application. The scope of this application includes but is not limited to the specific embodiments described above. All equivalent changes made in accordance with the shape and structure of this application are within the protection scope of this application.

Claims

1. A multi-objective oriented power grid scheduling decision method, characterized in that: The method comprises the following steps: constructing a normal scheduling target with minimum load fluctuation, an abnormal weather scheduling target with minimum scheduling quantity and minimum scheduling loss, and a device abnormal scheduling target with minimum switching loss; combining the normal scheduling target, the abnormal weather scheduling target and the device abnormal scheduling target with initial priority weights to construct a multi-objective optimization function; obtaining a current load fluctuation probability prediction value according to the matching of regional historical load data and regional real-time load data, a current weather abnormality probability prediction value according to the matching of regional historical weather data and regional real-time weather data, and a current device abnormality probability prediction value according to the matching of regional historical device maintenance data and regional real-time device data; obtaining scheduling weights of the normal scheduling target, the abnormal weather scheduling target and the device abnormal scheduling target respectively according to the current load fluctuation probability prediction value, the current weather abnormality probability prediction value and the current device abnormality probability prediction value; correcting the initial priority weights with the scheduling weights, updating the multi-objective optimization function, obtaining a multi-objective power grid scheduling decision with the updated multi-objective optimization function and regional real-time data, and executing power grid scheduling according to the multi-objective power grid scheduling decision; the construction of the abnormal weather scheduling target with minimum scheduling quantity and minimum scheduling loss comprises: calculating the scheduling quantity by multiplying the scheduling frequency of the distributed energy resource by the negative value of the scheduling frequency; calculating the scheduling loss by using the scheduling loss of the distributed energy resource and the disaster resistance loss of the reserved distributed energy resource; constructing constraint conditions of the abnormal weather scheduling target by using the line safety margin threshold and the renewable energy curtailment threshold; the construction of the device abnormal scheduling target with minimum switching loss comprises: obtaining a first switching loss according to the line difference loss of the abnormal distributed energy resource device and the distributed energy resource device to be switched; obtaining a second switching loss according to the transient loss of the switching of the distributed energy resource device; calculating the switching loss by using the first switching loss and the second switching loss.

2. The multi-objective power grid scheduling decision method of claim 1, wherein: the construction of the abnormal weather scheduling target with minimum scheduling quantity and minimum scheduling loss further comprises: setting the scheduling quantity weight and the scheduling loss weight according to the difference between the average value and the median value of the scheduling frequency of the actual regional distributed energy resource; constructing the abnormal weather scheduling target by using the scheduling quantity weight, the scheduling loss weight, the scheduling quantity and the scheduling loss.

3. The multi-objective power grid scheduling decision method of claim 1, wherein: the obtaining of the current load fluctuation probability prediction value according to the matching of the regional historical load data and the regional real-time load data comprises: obtaining precursor load data of a preset precursor time period, calculating the similarity between the regional precursor load data and the regional historical load data by using a similarity algorithm, and constructing a prediction trend set by using the historical load data with a similarity greater than a preset similarity threshold; obtaining a deviation ratio by comparing the regional real-time load data with the prediction trend set; obtaining the current load fluctuation probability prediction value by using the deviation ratio and the future load fluctuation in the prediction trend.

4. The power grid dispatching decision-making method oriented towards multiple objectives as described in claim 1, characterized in that: The step of obtaining the current weather anomaly probability prediction value by matching regional historical weather data with regional real-time weather data includes: Obtain real-time weather data for the region, filter historical weather data that match the real-time weather data for the same time period in the region's historical weather data, and obtain the current weather anomaly probability prediction value based on the proportion of abnormal weather within a preset period in all matched historical weather data.

5. The power grid dispatching decision-making method oriented towards multiple objectives as described in claim 1, characterized in that: The step of obtaining the current device anomaly probability prediction value by matching regional historical device maintenance data with regional real-time device data includes: Based on the correlation between device parameters and maintenance sequence in the historical device maintenance data of the region, device maintenance models for each device type are constructed; The remaining lifespan is obtained by using real-time regional device data and the corresponding device maintenance model, and the current device anomaly probability prediction value is obtained by using the ratio of the remaining lifespan to the initial lifespan.

6. The power grid dispatching decision-making method oriented towards multiple objectives as described in claim 1, characterized in that: The process of correcting the initial priority weights with scheduling weights, updating the multi-objective optimization function, and obtaining multi-objective power grid scheduling decisions based on the updated multi-objective optimization function and real-time regional data includes: Special scheduling relationships are obtained based on historical human scheduling data under abnormal weather conditions in the region and the types of weather anomalies. The current weather anomaly type is obtained by matching regional historical weather data with regional real-time weather data; Based on the current weather anomaly type, special scheduling correction weights are obtained according to special scheduling correlations; The scheduling weights are corrected using special scheduling correction weights, and the multi-objective optimization function is updated with the corrected scheduling weights. The updated multi-objective optimization function and regional real-time data are used to obtain multi-objective power grid scheduling decisions.

7. The power grid dispatching decision-making method oriented towards multiple objectives as described in claim 1, characterized in that: The goal of establishing normal scheduling with minimal load fluctuations includes: Load fluctuation is calculated using the standard deviation of load fluctuations from regional real-time load data and regional average load data. Constraints for normal scheduling objectives are constructed based on regional equipment power balance and equipment capacity.

8. A multi-objective oriented power grid dispatching decision system for implementing the method according to any one of claims 1 to 7, characterized in that: include: The storage layer is used to receive and store regional load data, weather data, maintenance data, and equipment data; The development layer is used to store the construction logic for normal scheduling targets, abnormal weather scheduling targets, device abnormality scheduling targets, and scheduling weights; The application layer is used to call data in the storage layer and the construction logic in the development layer to perform power grid dispatching decision calculations.

Images