Method, device and equipment for predicting openable capacity of power distribution network and storage medium

Abstract

The invention discloses a power distribution network openable capacity prediction method and device, equipment and a storage medium, and relates to the field of energy power, and the method comprises the steps: obtaining first power grid multi-source data of a target power distribution network, integrating the first power grid multi-source data to obtain corresponding integrated data, and storing the integrated data in a database; determining a target capacity curve corresponding to the target power distribution network based on the integrated data and a first load capacity corresponding to the target power distribution network; constructing an initial capacity prediction model corresponding to the target power distribution network based on the target capacity curve, and determining an initial prediction capacity corresponding to the target power distribution network based on a pre-acquired target adjustment request by using the initial capacity prediction model; and obtaining second power grid multi-source data of the target power distribution network, and obtaining a target capacity prediction result corresponding to the target power distribution network in the target time period by using the target prediction model based on the second power grid multi-source data and the initial prediction capacity corresponding to the target adjustment request. According to the invention, accurate estimation of the openable capacity of the power distribution network is realized.

Description

Technical Field

[0001] This invention relates to the field of energy and power, and in particular to a method, apparatus, equipment and storage medium for predicting the open capacity of a distribution network. Background Technology

[0002] Driven by the "dual carbon" goals, the penetration rate of distributed photovoltaic (PV) power in medium- and low-voltage distribution networks has grown rapidly. However, the intermittent and random nature of its output poses a severe challenge to distribution network planning and operation. Traditional methods for assessing the available capacity of distribution networks are typically based on static capacity calculations under the worst-case scenario. This method assumes extreme scenarios of maximum load and minimum output, leading to overly conservative assessment results that fail to reflect the true capacity of the distribution network. This results in low utilization of distribution network assets, rejection of numerous distributed power source integration applications, or the need for expensive capacity expansion and upgrades, severely restricting the large-scale consumption of renewable energy. Therefore, accurately predicting the available capacity of the distribution network after distributed power sources are integrated is an urgent problem to be solved. Summary of the Invention

[0003] In view of this, the purpose of this invention is to provide a method, apparatus, device, and storage medium for predicting the available capacity of a distribution network, capable of accurately predicting the available capacity of the distribution network. The specific solution is as follows:

[0004] In a first aspect, this application discloses a method for predicting the open capacity of a distribution network, including:

[0005] First-source grid data of the target distribution network is acquired, and the first-source grid data is integrated to obtain corresponding integrated data. Based on the integrated data and the first load capacity corresponding to the target distribution network, the target capacity curve corresponding to the target distribution network is determined. The grid multi-source data includes load data, output data of each distributed power source of the target distribution network, meteorological data, and grid topology data. The first load capacity is the maximum load capacity of the target distribution network under preset static conditions. The target capacity curve is the dynamic curve of the available capacity corresponding to the target distribution network.

[0006] Based on the target capacity curve, a preliminary capacity prediction model corresponding to the target distribution network is constructed. The preliminary capacity prediction model is then used to determine the initial predicted capacity of the target distribution network based on a pre-acquired target adjustment request. The target adjustment request is a request to adjust the distributed power sources of the target distribution network.

[0007] The second grid multi-source data of the target distribution network is obtained. Based on the second grid multi-source data and the initial predicted capacity corresponding to the target adjustment request, the target capacity prediction result of the target distribution network in the target time period is obtained using the target prediction model. The target prediction model is a prediction model built based on a long short-term memory network. The target capacity prediction result includes the load prediction result of the target distribution network, the output prediction result of the distributed power source, and the openable capacity prediction result of the target distribution network.

[0008] Optionally, determining the target capacity curve corresponding to the target distribution network based on the integrated data and the first load capacity corresponding to the target distribution network includes:

[0009] Target feature factors are determined based on the integrated data; the target feature factors include feature factors corresponding to different time scales.

[0010] The feature weights corresponding to each target feature factor are determined based on a preset dynamic weight allocation mechanism, and the comprehensive feature factor is determined based on the feature weights corresponding to the target feature factors.

[0011] Based on the comprehensive characteristic factors and the first load capacity corresponding to the target distribution network, the flexible adjustment increment corresponding to the target distribution network is determined, and the target capacity curve corresponding to the target distribution network is determined based on the flexible adjustment increment and the first load capacity.

[0012] Optionally, determining the initial predicted capacity of the target distribution network based on the pre-acquired target adjustment request using the preliminary capacity prediction model includes:

[0013] The preliminary capacity prediction model is used to determine the target adjustment node and the output power of the corresponding power source to be adjusted based on the pre-acquired target adjustment request. The power source to be adjusted is the distributed power source to be connected or the distributed power source to be removed corresponding to the target adjustment node.

[0014] Based on the target adjustment request, determine the target load status of the target adjustment node after the adjustment of the power supply to be adjusted;

[0015] The initial predicted capacity of the target distribution network is determined based on the target load status.

[0016] Optionally, determining the initial predicted capacity corresponding to the target distribution network based on the target load state includes:

[0017] Using the preliminary capacity prediction model, the local absorption coefficient corresponding to the target adjustment node is determined based on the target load state and the preset penalty coefficient; the preset penalty coefficient is the voltage over-limit penalty coefficient corresponding to the target distribution network.

[0018] Based on the output power of the power source to be adjusted corresponding to each target adjustment node and the local absorption coefficient, as well as the first load capacity corresponding to the target distribution network, the initial predicted capacity corresponding to the target distribution network is determined.

[0019] Optionally, the step of obtaining the target capacity prediction result of the target distribution network in the target time period based on the second power grid multi-source data and the initial predicted capacity corresponding to the target adjustment request using the target prediction model includes:

[0020] The target prediction model is used to standardize, reduce dimensionality, and extract features from the multi-source data of the second power grid to obtain the processed data corresponding to the multi-source data of the second power grid.

[0021] The target dependency relationship is obtained based on the processed data using the preset gating mechanism of the target prediction model, and the target capacity prediction result of the target distribution network in the target time period is obtained based on the target dependency relationship and the initial prediction capacity corresponding to the target adjustment request.

[0022] Optionally, after obtaining the target capacity prediction result of the target distribution network in the target time period using the target prediction model, the method further includes:

[0023] Based on the objective constraints and preset optimization objectives, a multi-objective optimization model corresponding to the target capacity prediction results is determined.

[0024] The target power supply adjustment scheme corresponding to the target capacity prediction result is determined based on the target optimization algorithm and the multi-objective optimization model.

[0025] Optionally, determining the target power supply adjustment scheme corresponding to the target capacity prediction result based on the target optimization algorithm and the multi-objective optimization model includes:

[0026] Based on the objective optimization algorithm and the multi-objective optimization model, a power adjustment scheme corresponding to the target capacity prediction result is determined;

[0027] Simulation verification is performed based on the power adjustment scheme to determine whether the power adjustment scheme meets the preset adjustment target;

[0028] If the power adjustment scheme does not meet the preset adjustment target, a reacquisition request is initiated to obtain a new target adjustment request, and after obtaining the new target adjustment request, the process jumps to the step of using the preliminary capacity prediction model to determine the initial predicted capacity of the target distribution network based on the pre-acquired target adjustment request.

[0029] If the power adjustment scheme meets the preset adjustment target, then the power adjustment scheme is determined as the target power adjustment scheme.

[0030] Secondly, this application discloses a power distribution network openable capacity prediction device, comprising:

[0031] The capacity curve determination module is used to acquire first-source grid data of the target distribution network, integrate the first-source grid data to obtain corresponding integrated data, and determine the target capacity curve corresponding to the target distribution network based on the integrated data and the first load capacity corresponding to the target distribution network. The grid multi-source data includes load data, output data of each distributed power source of the target distribution network, meteorological data, and grid topology data. The first load capacity is the maximum load capacity of the target distribution network under preset static conditions. The target capacity curve is the dynamic curve of the openable capacity corresponding to the target distribution network.

[0032] The initial capacity prediction module is used to construct a preliminary capacity prediction model corresponding to the target distribution network based on the target capacity curve, and to determine the initial predicted capacity corresponding to the target distribution network based on the preliminary capacity prediction model and a pre-acquired target adjustment request; the target adjustment request is a request to adjust the distributed power sources of the target distribution network.

[0033] The prediction result acquisition module is used to acquire second grid multi-source data of the target distribution network, and based on the second grid multi-source data and the initial predicted capacity corresponding to the target adjustment request, use the target prediction model to obtain the target capacity prediction result of the target distribution network in the target time period; the target prediction model is a prediction model built based on a long short-term memory network, and the target capacity prediction result includes the load prediction result corresponding to the target distribution network, the output prediction result of the distributed power source, and the available capacity prediction result corresponding to the target distribution network.

[0034] Thirdly, this application discloses an electronic device, including:

[0035] Memory, used to store computer programs;

[0036] A processor is used to execute the computer program to implement the aforementioned method for predicting the open capacity of the distribution network.

[0037] Fourthly, this application discloses a computer-readable storage medium for storing a computer program, wherein the computer program, when executed by a processor, implements the aforementioned method for predicting the open capacity of a power distribution network.

[0038] In this application, when predicting the available capacity of a target distribution network, first multi-source data of the target distribution network is acquired, and the first multi-source data is integrated to obtain corresponding integrated data. Based on the integrated data and the first load capacity corresponding to the target distribution network, a target capacity curve corresponding to the target distribution network is determined. The multi-source data includes load data, output data of each distributed power source in the target distribution network, meteorological data, and grid topology data. The first load capacity is the maximum load capacity of the target distribution network under preset static conditions. The target capacity curve is the dynamic curve of the available capacity of the target distribution network. A preliminary capacity prediction model corresponding to the target distribution network is constructed based on the target capacity curve. The initial predicted capacity of the target distribution network is determined using the preliminary capacity prediction model based on a pre-acquired target adjustment request. The target adjustment request is a request to adjust the distributed generation of the target distribution network. Secondary grid multi-source data of the target distribution network is acquired. Based on the secondary grid multi-source data and the initial predicted capacity corresponding to the target adjustment request, the target prediction model is used to obtain the target capacity prediction result of the target distribution network for the target time period. The target prediction model is a prediction model constructed based on a Long Short-Term Memory (LSTM) network. The target capacity prediction result includes the load prediction result of the target distribution network, the output prediction result of the distributed generation, and the available capacity prediction result of the target distribution network. Therefore, this application first integrates and analyzes the characteristics of the target distribution network using multi-source data. Then, based on the integrated data, a target capacity curve describing the evolution of available capacity over time is constructed. Next, the preliminary capacity prediction model corresponding to the target distribution network is constructed using the target capacity curve as a collaborative evaluation model to quantify local absorption capacity and power backflow risk, accurately assess the impact of distributed generation adjustments corresponding to the target adjustment request on the target distribution network, and obtain the corresponding initial predicted capacity. Finally, a target prediction model based on a long short-term memory network is used to predict the available capacity and source load status for multiple future time periods. This yields the target capacity prediction results for the target distribution network during the target time period after adjustments are made based on the target adjustment request, thus achieving accurate estimation of the available capacity of the distribution network. Attached Figure Description

[0039] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0040] Figure 1 This is a flowchart of a method for predicting the open capacity of a distribution network disclosed in this application;

[0041] Figure 2 This is a schematic diagram of a specific distribution network open capacity prediction process disclosed in this application;

[0042] Figure 3 This is a schematic diagram of a specific target capacity curve determination process disclosed in this application;

[0043] Figure 4 This is a schematic diagram of a specific target distribution network for determining the initial predicted capacity disclosed in this application;

[0044] Figure 5 This is a schematic diagram of a specific closed-loop workflow for AI prediction and optimization disclosed in this application;

[0045] Figure 6 This is a schematic diagram of the structure of a power distribution network open capacity prediction device disclosed in this application;

[0046] Figure 7 This is a structural diagram of an electronic device disclosed in this application. Detailed Implementation

[0047] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0048] Driven by the "dual carbon" goals, the penetration rate of distributed photovoltaic (PV) power in medium- and low-voltage distribution networks has grown rapidly. However, the intermittent and random nature of its output poses a severe challenge to distribution network planning and operation. Traditional methods for assessing the available capacity of distribution networks are typically based on static capacity calculations under the worst-case scenario. This method assumes extreme scenarios of maximum load and minimum output, leading to overly conservative assessment results that fail to reflect the true capacity of the distribution network. This results in low utilization of distribution network assets, rejection of numerous distributed power source access applications, or the need for expensive capacity expansion and upgrades, severely restricting the large-scale consumption of renewable energy. To address these technical problems, this application discloses a method for predicting the available capacity of distribution networks, enabling accurate estimation of the available capacity of distribution networks.

[0049] See Figure 1 As shown in the figure, an embodiment of the present invention discloses a method for predicting the openable capacity of a distribution network, including:

[0050] Step S11: Obtain first grid multi-source data of the target distribution network, integrate the first grid multi-source data to obtain corresponding integrated data, and determine the target capacity curve corresponding to the target distribution network based on the integrated data and the first load capacity corresponding to the target distribution network; the grid multi-source data includes load data, output data of each distributed power source of the target distribution network, meteorological data, and grid topology data; the first load capacity is the maximum load capacity of the target distribution network under preset static conditions; the target capacity curve is the dynamic curve of the available capacity corresponding to the target distribution network.

[0051] In this embodiment, as Figure 2 As shown, the entire process of predicting the open capacity of a distribution network can be decomposed into five logical stages: data integration, dynamic modeling, collaborative evaluation, intelligent decision-making, and simulation verification, forming a closed loop from data acquisition to scheme output. The distribution network open capacity prediction system built based on this embodiment first performs multi-source data integration and characteristic analysis, collecting and processing multi-source data from the first power grid, including historical and real-time data of the target distribution network, to obtain the corresponding integrated data. It can be understood that the multi-source data from the first power grid may include load data (such as node active / reactive power). and Output data of distributed generation (DG, DG) (such as active power output) The system collects meteorological data (such as light intensity and temperature) and power grid topology data. This data is uploaded via smart meters, sensors, and SCADA (Supervisory Control and Data Acquisition) systems, and undergoes preprocessing such as data cleaning, alignment, and normalization to ensure spatiotemporal consistency. Specifically, multiple acquisition terminals can be integrated into the system's data sensing layer, and preprocessing algorithms such as unified time stamping, data cleaning, normalization, and feature extraction can be applied to achieve spatiotemporal alignment and feature fusion of multi-source heterogeneous data from meteorological, electrical, and load sources. This mechanism provides a high-quality, highly consistent input data foundation for dynamic modeling and intelligent decision-making, and is a key prerequisite for supporting the entire system to achieve precise collaborative analysis of the "source-grid-load" system.

[0052] In this embodiment, after obtaining the integrated data, the target capacity curve corresponding to the target distribution network can be determined based on the integrated data and the first load capacity corresponding to the target distribution network. Specifically, this can include: determining target characteristic factors based on the integrated data; the target characteristic factors include characteristic factors corresponding to different time scales; determining the characteristic weights corresponding to each target characteristic factor based on a preset dynamic weight allocation mechanism, and determining a comprehensive characteristic factor based on the characteristic weights corresponding to the target characteristic factors; determining the flexible adjustment increment corresponding to the target distribution network based on the comprehensive characteristic factors and the first load capacity corresponding to the target distribution network, so as to determine the target capacity curve corresponding to the target distribution network based on the flexible adjustment increment and the first load capacity, thereby realizing the multi-time-scale dynamic characteristic modeling of the openable capacity of the target distribution network.

[0053] In one specific implementation, such as Figure 3 As shown, the static capacity input (i.e., the first load capacity) is the base capacity value calculated based on the traditional worst-case operating conditions. First, based on the integrated data, the calculations for each time scale are performed. Target feature factors (including time scales such as day, week, month, and year) For example, day-scale factor. This can be expressed as the daily volatility of load and DG output:

[0054] ;

[0055] in, The standard deviation of the nodal net power of the target distribution network. The mean value of the node net power of the target distribution network reflects the intensity of intraday net power fluctuations. and These refer to the total load power and total output power of distributed generation in the power grid system at that time scale, respectively. Similarly, the week-scale factor... The calculation can be based on the differences in load patterns between weekdays and weekends. First, the weekday and weekend scale factors can be determined separately based on the integrated data. Then, the weekday and weekend scale factors are weighted and summed to obtain the weekly scale factor. Monthly and annual scale factors can be obtained using the same method. Specifically, when calculating the annual scale factor, the influence of the monthly scale factor on the annual scale factor can be adjusted based on the seasonal impact on the distribution network's load status. After obtaining all the required target scale factors, feature weights can be dynamically assigned to each target scale factor based on its importance (e.g., higher weight for recent data). ,satisfy Feature weights can be determined using the entropy weight method or the AHP (Analytic Hierarchy Process), and finally, a comprehensive feature factor is synthesized. :

[0056] ;

[0057] The comprehensive characteristic factors comprehensively reflect the dynamic characteristics of source-load interaction across multiple time scales. Based on these comprehensive characteristic factors and the first load capacity corresponding to the target distribution network, the flexible regulation increment for the target distribution network can be determined. :

[0058] ;

[0059] in, The adjustment coefficients set based on the grid's flexible regulation capabilities (such as energy storage and demand response) ultimately yield the dynamic open capacity curve of the target distribution network. (That is, the target capacity curve):

[0060] ;

[0061] The output curve is a time function, intuitively displaying the available capacity at different future time periods, providing dynamic input for subsequent evaluation and optimization. This embodiment innovatively introduces a multi-timescale dynamic characteristic model, extracting dynamic influencing factors at daily, weekly, monthly, and yearly scales in parallel, and using the entropy weight method for weighted fusion, transforming the static capacity benchmark into a time-varying dynamic available capacity curve. This breakthrough enables the evaluation results to accurately depict the evolution of the distribution network's true capacity under seasonal changes and load fluctuations, achieving a fundamental shift from post-event static calculation to pre-event dynamic perception, and providing a scientific basis for tapping the grid's potential and delaying capacity expansion investment.

[0062] Step S12: Construct a preliminary capacity prediction model for the target distribution network based on the target capacity curve, and use the preliminary capacity prediction model to determine the initial predicted capacity of the target distribution network based on a pre-acquired target adjustment request; the target adjustment request is a request to adjust the distributed power sources of the target distribution network.

[0063] In this embodiment, as Figure 4 As shown, the preliminary capacity prediction model for the target distribution network, constructed based on the target capacity curve, is the collaborative evaluation model. This model quantifies the net impact of DG (distributed generation) integration or removal, ensuring that the evaluation results balance local consumption and grid security. When using the preliminary capacity prediction model to determine the initial predicted capacity of the target distribution network based on pre-acquired target adjustment requests, the specific process may include: using the preliminary capacity prediction model based on pre-acquired target adjustment requests; determining the target adjustment nodes and their corresponding output power to be adjusted based on the target adjustment requests; the power sources to be adjusted are either distributed generation sources to be integrated or to be removed from the target adjustment nodes; determining the target load state of the target adjustment nodes after the adjustment of the power sources to be adjusted based on the target adjustment requests; and determining the initial predicted capacity of the target distribution network based on the target load state. The determination of the initial predicted capacity of the target distribution network based on the target load status may include: using a preliminary capacity prediction model to determine the local absorption coefficient of the target adjustment node based on the target load status and a preset penalty coefficient; the preset penalty coefficient is the voltage over-limit penalty coefficient of the target distribution network; and determining the initial predicted capacity of the target distribution network based on the output power and local absorption coefficient of the power source to be adjusted corresponding to each target adjustment node, as well as the first load capacity of the target distribution network.

[0064] In one specific implementation, the target adjustment request may be "to add a distributed generation (DG) to the target adjustment node i," and the input to the collaborative evaluation model is the output power of the DG and the local load data at the target adjustment node i. First, it is determined that after adjusting the distributed generation at the target adjustment node i, the DG output at the target adjustment node i will... Load at target adjustment node i Match: If If local consumption reduces the pressure on the upstream power grid, then the DG output is fully absorbed, which has a positive impact on the power grid; if This could lead to power backfeed causing voltage overshoot, requiring calculation of the backfeed power. And verify whether it causes node voltage Exceeding the limit To quantify the net impact, the collaborative assessment model calculates the local absorption coefficient. :

[0065] ;

[0066] in, This is the voltage over-limit penalty coefficient, which can be set based on the voltage sensitivity obtained from power flow calculations. This comprehensively reflects both the positive (absorption benefits) and negative (backflow risk) impacts of DG integration. Ultimately, the collaborative evaluation model outputs the initial predicted capacity corresponding to the target distribution network. :

[0067] ;

[0068] Through the above process, this embodiment establishes a collaborative evaluation mechanism. By judging the matching relationship between DG output and local load, a local absorption coefficient that incorporates voltage over-limit penalties is introduced to quantify the risks of local absorption and power backfeed. A coupling impact model between DG and load is established to accurately quantify the net impact of DG access, ensuring that operational risks are avoided while improving absorption levels. Finally, the maximum carrying capacity that balances absorption benefits and grid security is output as the initial predicted capacity.

[0069] Step S13: Obtain the second grid multi-source data of the target distribution network. Based on the second grid multi-source data and the initial predicted capacity corresponding to the target adjustment request, use the target prediction model to obtain the target capacity prediction result of the target distribution network in the target time period. The target prediction model is a prediction model built based on a long short-term memory network. The target capacity prediction result includes the load prediction result of the target distribution network, the output prediction result of the distributed power source, and the openable capacity prediction result of the target distribution network.

[0070] In this embodiment, as Figure 5 The diagram illustrates the closed-loop workflow of AI prediction and optimization. After acquiring the second-generation multi-source data of the target distribution network, based on the second-generation multi-source data and the initial predicted capacity corresponding to the target adjustment request, the target prediction model is used to obtain the target capacity prediction result of the target distribution network in the target time period. This includes: using the target prediction model to standardize, reduce dimensionality, and extract features from the second-generation multi-source data to obtain the processed data; using the preset gating mechanism of the target prediction model to obtain the target dependency relationship based on the processed data; and based on the target dependency relationship and the initial predicted capacity corresponding to the target adjustment request, obtaining the target capacity prediction result of the target distribution network in the target time period.

[0071] In one specific implementation, the target prediction model first standardizes, reduces dimensionality, and extracts features from the input data through data preprocessing and feature engineering. Then, it employs an LSTM (Long Short-Term Memory) rolling prediction algorithm. The LSTM model captures long-term dependencies through gating mechanisms (input gate, forget gate, output gate), and its core unit state update formula is:

[0072] ;

[0073] in, For the Gate of Oblivion For input gate, Candidate state This represents the current state. The LSTM model uses historical sequences... As input, output the prediction results for multiple future time periods (i.e., the target capacity prediction results). Including load forecasting DG output prediction and available capacity prediction .

[0074] In this embodiment, after obtaining the target capacity prediction result of the target distribution network in the target time period using the target prediction model, the method further includes: determining the multi-objective optimization model corresponding to the target capacity prediction result based on the target constraint relationship and the preset optimization objective; and determining the target power supply adjustment scheme corresponding to the target capacity prediction result based on the target optimization algorithm and the multi-objective optimization model. Specifically, determining the target power supply adjustment scheme corresponding to the target capacity prediction result based on the target optimization algorithm and the multi-objective optimization model includes: determining the power supply adjustment scheme corresponding to the target capacity prediction result based on the target optimization algorithm and the multi-objective optimization model; performing simulation verification based on the power supply adjustment scheme to determine whether the power supply adjustment scheme meets the preset adjustment objective; if the power supply adjustment scheme does not meet the preset adjustment objective, initiating a re-acquisition request to obtain a new target adjustment request, and after obtaining the new target adjustment request, jumping to the step of determining the initial predicted capacity corresponding to the target distribution network based on the pre-acquired target adjustment request using the preliminary capacity prediction model; and if the power supply adjustment scheme meets the preset adjustment objective, then determining the power supply adjustment scheme as the target power supply adjustment scheme.

[0075] In one specific implementation, the parameters from the target prediction results can be substituted into the target constraints to construct a multi-objective optimization model for the target distribution network, aiming to maximize distributed generation (DG) absorption and minimize voltage deviation and network loss. This multi-objective optimization model can be expressed as:

[0076] ;

[0077] in To minimize total cost, , , The weighting coefficients are as follows: the first term maximizes DG absorption, the second term minimizes voltage deviation, and the third term minimizes network loss. The target constraints include voltage safety. Operational safety constraints include line thermal stability and transformer capacity. The target prediction model uses an optimization algorithm to solve this multi-objective optimization model and outputs the optimal access scheme and control strategy (such as DG access location, capacity, and inverter power factor settings). Simulation verification is then performed to check whether the scheme meets the preset adjustment objectives (such as voltage safety, line capacity, and absorption rate). If the solution does not meet the requirements, the decision variables (such as DG access point or capacity) are adjusted and re-optimized; otherwise, the final scheme is output. Alternatively, iterative optimization can be performed by adjusting the multi-objective optimization model (such as constraint weights and objective function coefficients). Finally, the system outputs a dynamic evaluation report and access optimization scheme, providing a scientific basis for distribution network planning and operation. By integrating LSTM prediction and multi-objective optimization, this embodiment forms a self-learning, adaptive intelligent decision-making closed loop from prediction and optimization to simulation verification, and the final output target power supply adjustment scheme combines safety and optimality.

[0078] Based on the above process, this embodiment constructs a four-layer analysis architecture of "data-driven, dynamic modeling, collaborative evaluation, and intelligent decision-making," systematically improving the efficiency of distribution network resource utilization and the level of renewable energy absorption, and providing a reliable data foundation for the system. While ensuring grid security, this embodiment fully releases the acceptance potential of existing distribution assets, effectively supporting accurate decisions for "no capacity expansion" or "micro-capacity expansion" access, providing key technical support for high-proportion renewable energy access and the construction of new power systems, and realizing a fundamental shift from static calculation to dynamic prediction, optimization, and verification.

[0079] As can be seen, this application first integrates and analyzes the characteristics of the target distribution network through multi-source data. Then, based on the integrated data, it constructs a target capacity curve describing the evolution of available capacity over time. Next, it uses the target capacity curve to construct a preliminary capacity prediction model for the target distribution network as a collaborative evaluation model to quantify local absorption capacity and power backflow risk, accurately assess the impact of distributed generation adjustments corresponding to the target adjustment request on the target distribution network, and obtain the corresponding initial predicted capacity. Finally, it uses a target prediction model based on a long short-term memory network to predict the available capacity and source-load status for future multi-period periods, thereby obtaining the target capacity prediction result of the target distribution network in the target period after adjustments are made based on the target adjustment request, achieving accurate estimation of the available capacity of the distribution network.

[0080] See Figure 6As shown, this application discloses a power distribution network openable capacity prediction device, comprising:

[0081] The capacity curve determination module 11 is used to acquire first grid multi-source data of the target distribution network, integrate the first grid multi-source data to obtain corresponding integrated data, and determine the target capacity curve corresponding to the target distribution network based on the integrated data and the first load capacity corresponding to the target distribution network. The grid multi-source data includes load data, output data of each distributed power source of the target distribution network, meteorological data, and grid topology data. The first load capacity is the maximum load capacity of the target distribution network under preset static conditions. The target capacity curve is the dynamic curve of the openable capacity corresponding to the target distribution network.

[0082] The initial capacity prediction module 12 is used to construct a preliminary capacity prediction model corresponding to the target distribution network based on the target capacity curve, and to determine the initial predicted capacity corresponding to the target distribution network based on a pre-acquired target adjustment request using the preliminary capacity prediction model; the target adjustment request is a request to adjust the distributed power sources of the target distribution network.

[0083] The prediction result acquisition module 13 is used to acquire the second grid multi-source data of the target distribution network, and based on the second grid multi-source data and the initial predicted capacity corresponding to the target adjustment request, use the target prediction model to obtain the target capacity prediction result of the target distribution network in the target time period; the target prediction model is a prediction model constructed based on a long short-term memory network, and the target capacity prediction result includes the load prediction result corresponding to the target distribution network, the output prediction result of the distributed power source, and the openable capacity prediction result corresponding to the target distribution network.

[0084] As can be seen, this application first integrates and analyzes the characteristics of the target distribution network through multi-source data. Then, based on the integrated data, it constructs a target capacity curve describing the evolution of available capacity over time. Next, it uses the target capacity curve to construct a preliminary capacity prediction model for the target distribution network as a collaborative evaluation model to quantify local absorption capacity and power backflow risk, accurately assess the impact of distributed generation adjustments corresponding to the target adjustment request on the target distribution network, and obtain the corresponding initial predicted capacity. Finally, it uses a target prediction model based on a long short-term memory network to predict the available capacity and source-load status for future multi-period periods, thereby obtaining the target capacity prediction result of the target distribution network in the target period after adjustments are made based on the target adjustment request, achieving accurate estimation of the available capacity of the distribution network.

[0085] In one specific embodiment, the capacity curve determination module 11 may include:

[0086] The feature factor determination submodule is used to determine target feature factors based on the integrated data; the target feature factors include feature factors corresponding to different time scales.

[0087] The feature factor synthesis submodule is used to determine the feature weights corresponding to each target feature factor based on a preset dynamic weight allocation mechanism, and to determine the comprehensive feature factor based on the feature weights corresponding to the target feature factors.

[0088] The capacity curve determination submodule is used to determine the flexible adjustment increment corresponding to the target distribution network based on the comprehensive characteristic factor and the first load capacity corresponding to the target distribution network, so as to determine the target capacity curve corresponding to the target distribution network based on the flexible adjustment increment and the first load capacity.

[0089] In one specific embodiment, the initial capacity prediction module 12 may include:

[0090] The adjustment request analysis submodule is used to utilize the preliminary capacity prediction model to determine the target adjustment node and the output power of the corresponding power source to be adjusted based on the pre-acquired target adjustment request of the target distribution network; the power source to be adjusted is the distributed power source to be connected or the distributed power source to be removed corresponding to the target adjustment node.

[0091] The load status determination submodule is used to determine the target load status of the target adjustment node after the adjustment of the power supply to be adjusted, based on the target adjustment request.

[0092] The first capacity prediction submodule is used to determine the initial predicted capacity corresponding to the target distribution network based on the target load status.

[0093] In one specific implementation, the first capacity prediction submodule may specifically include:

[0094] The absorption coefficient determination unit is used to determine the local absorption coefficient corresponding to the target adjustment node based on the target load state and the preset penalty coefficient using the preliminary capacity prediction model; the preset penalty coefficient is the voltage over-limit penalty coefficient corresponding to the target distribution network.

[0095] The first capacity prediction unit is used to determine the initial predicted capacity of the target distribution network based on the output power of the power source to be adjusted corresponding to each target adjustment node, the local absorption coefficient, and the first load capacity corresponding to the target distribution network.

[0096] In one specific embodiment, the prediction result acquisition module 13 may specifically include:

[0097] The data processing submodule is used to standardize, reduce dimensionality, and extract features from the second power grid multi-source data using the target prediction model, so as to obtain the processed data corresponding to the second power grid multi-source data.

[0098] The second capacity prediction submodule is used to obtain the target dependency relationship based on the processed data using the preset gating mechanism of the target prediction model, and to obtain the target capacity prediction result of the target distribution network in the target time period based on the target dependency relationship and the initial prediction capacity corresponding to the target adjustment request.

[0099] In one specific embodiment, the device may further include:

[0100] The optimization model determination module is used to determine the multi-objective optimization model corresponding to the target capacity prediction result based on the target constraint relationship and the preset optimization objective.

[0101] The adjustment scheme determination module is used to determine the target power supply adjustment scheme corresponding to the target capacity prediction result based on the target optimization algorithm and the multi-objective optimization model.

[0102] In one specific implementation, the adjustment scheme determination module may include:

[0103] The scheme acquisition submodule is used to determine the power adjustment scheme corresponding to the target capacity prediction result based on the target optimization algorithm and the multi-objective optimization model.

[0104] The scheme verification submodule is used to perform simulation verification based on the power adjustment scheme to determine whether the power adjustment scheme meets the preset adjustment target.

[0105] The scheme adjustment submodule is used to initiate a reacquisition request to obtain a new target adjustment request if the power supply adjustment scheme does not meet the preset adjustment target, and after obtaining the new target adjustment request, jump to the step of determining the initial predicted capacity of the target distribution network based on the pre-acquisition target adjustment request using the preliminary capacity prediction model.

[0106] The scheme determination submodule is used to determine the power adjustment scheme as the target power adjustment scheme if the power adjustment scheme meets the preset adjustment target.

[0107] Furthermore, embodiments of this application also disclose an electronic device, Figure 7 This is a structural diagram of an electronic device 20 according to an exemplary embodiment. The content of the diagram should not be construed as limiting the scope of this application.

[0108] Figure 7This is a schematic diagram of the structure of an electronic device 20 provided in an embodiment of this application. Specifically, the electronic device 20 may include: at least one processor 21, at least one memory 22, a power supply 23, a communication interface 24, an input / output interface 25, and a communication bus 26. The memory 22 stores a computer program, which is loaded and executed by the processor 21 to implement the relevant steps in the power distribution network open capacity prediction method disclosed in any of the foregoing embodiments. Alternatively, the electronic device 20 in this embodiment may specifically be a computer.

[0109] In this embodiment, the power supply 23 is used to provide operating voltage for each hardware device on the electronic device 20; the communication interface 24 can create a data transmission channel between the electronic device 20 and external devices, and the communication protocol it follows can be any communication protocol applicable to the technical solution of this application, and is not specifically limited here; the input / output interface 25 is used to acquire external input data or output data to the outside world, and its specific interface type can be selected according to specific application needs, and is not specifically limited here.

[0110] In addition, the memory 22, as a carrier for resource storage, can be a read-only memory, random access memory, disk, or optical disk, etc. The resources stored thereon can include an operating system 221, computer programs 222, etc., and the storage method can be temporary storage or permanent storage.

[0111] The operating system 221 is used to manage and control the various hardware devices on the electronic device 20 and the computer program 222, which may be Windows Server, Netware, Unix, Linux, etc. In addition to including a computer program capable of performing the power distribution network open capacity prediction method executed by the electronic device 20 as disclosed in any of the foregoing embodiments, the computer program 222 may further include computer programs capable of performing other specific tasks.

[0112] Furthermore, this application also discloses a computer-readable storage medium for storing a computer program; wherein, when the computer program is executed by a processor, it implements the aforementioned method for predicting the openable capacity of a distribution network. Specific steps of this method can be found in the corresponding content disclosed in the foregoing embodiments, and will not be repeated here.

[0113] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the apparatus disclosed in the embodiments, since it corresponds to the method disclosed in the embodiments, the description is relatively simple; relevant parts can be referred to in the method section.

[0114] Those skilled in the art will further recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0115] The steps of the methods or algorithms described in conjunction with the embodiments disclosed herein can be implemented directly by hardware, a software module executed by a processor, or a combination of both. The software module can be located in random access memory (RAM), main memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, removable disk, CD-ROM, or any other form of storage medium known in the art.

[0116] Finally, it should be noted that in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0117] The technical solutions provided in this application have been described in detail above. Specific examples have been used to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the methods and core ideas of this application. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of this application. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A method for predicting the open capacity of a distribution network, characterized in that, include: First-source power grid data of the target distribution network is acquired, the first-source power grid data is integrated to obtain the corresponding integrated data, and the target capacity curve of the target distribution network is determined based on the integrated data and the first load capacity of the target distribution network. The multi-source data of the power grid includes load data, output data of each distributed power source of the target distribution network, meteorological data, and power grid topology data; the first load capacity is the maximum load capacity of the target distribution network under preset static conditions; the target capacity curve is the dynamic curve of the available capacity of the target distribution network. Based on the target capacity curve, a preliminary capacity prediction model is constructed for the target distribution network. The initial predicted capacity of the target distribution network is determined using the preliminary capacity prediction model based on the pre-acquired target adjustment request. The target adjustment request is a request to adjust the distributed power sources of the target distribution network; The second grid multi-source data of the target distribution network is obtained. Based on the second grid multi-source data and the initial predicted capacity corresponding to the target adjustment request, the target capacity prediction result of the target distribution network in the target time period is obtained using the target prediction model. The target prediction model is a prediction model built based on a long short-term memory network. The target capacity prediction result includes the load prediction result of the target distribution network, the output prediction result of the distributed power source, and the openable capacity prediction result of the target distribution network.

2. The method for predicting the openable capacity of a distribution network according to claim 1, characterized in that, The step of determining the target capacity curve corresponding to the target distribution network based on the integrated data and the first load capacity corresponding to the target distribution network includes: Target feature factors are determined based on the integrated data; the target feature factors include feature factors corresponding to different time scales. The feature weights corresponding to each target feature factor are determined based on a preset dynamic weight allocation mechanism, and the comprehensive feature factor is determined based on the feature weights corresponding to the target feature factors. Based on the comprehensive characteristic factors and the first load capacity corresponding to the target distribution network, the flexible adjustment increment corresponding to the target distribution network is determined, and the target capacity curve corresponding to the target distribution network is determined based on the flexible adjustment increment and the first load capacity.

3. The method for predicting the openable capacity of a distribution network according to claim 1, characterized in that, The step of determining the initial predicted capacity of the target distribution network based on the pre-acquired target adjustment request using the preliminary capacity prediction model includes: The preliminary capacity prediction model is used to determine the target adjustment node and the output power of the corresponding power source to be adjusted based on the pre-acquired target adjustment request. The power source to be adjusted is the distributed power source to be connected or the distributed power source to be removed corresponding to the target adjustment node. Based on the target adjustment request, determine the target load status of the target adjustment node after the adjustment of the power supply to be adjusted; The initial predicted capacity of the target distribution network is determined based on the target load status.

4. The method for predicting the openable capacity of a distribution network according to claim 3, characterized in that, Determining the initial predicted capacity of the target distribution network based on the target load state includes: Using the preliminary capacity prediction model, the local absorption coefficient corresponding to the target adjustment node is determined based on the target load state and the preset penalty coefficient; the preset penalty coefficient is the voltage over-limit penalty coefficient corresponding to the target distribution network. Based on the output power of the power source to be adjusted corresponding to each target adjustment node and the local absorption coefficient, as well as the first load capacity corresponding to the target distribution network, the initial predicted capacity corresponding to the target distribution network is determined.

5. The method for predicting the openable capacity of a distribution network according to claim 1, characterized in that, The step of obtaining the target capacity prediction result of the target distribution network in the target time period based on the initial predicted capacity corresponding to the second power grid multi-source data and the target adjustment request, using the target prediction model, includes: The target prediction model is used to standardize, reduce dimensionality, and extract features from the multi-source data of the second power grid to obtain the processed data corresponding to the multi-source data of the second power grid. The target dependency relationship is obtained based on the processed data using the preset gating mechanism of the target prediction model, and the target capacity prediction result of the target distribution network in the target time period is obtained based on the target dependency relationship and the initial prediction capacity corresponding to the target adjustment request.

6. The method for predicting the openable capacity of a distribution network according to claim 1, characterized in that, After obtaining the target capacity prediction result of the target distribution network in the target time period using the target prediction model, the method further includes: Based on the objective constraints and preset optimization objectives, a multi-objective optimization model corresponding to the target capacity prediction results is determined. The target power supply adjustment scheme corresponding to the target capacity prediction result is determined based on the target optimization algorithm and the multi-objective optimization model.

7. The method for predicting the openable capacity of a distribution network according to claim 6, characterized in that, The step of determining the target power supply adjustment scheme corresponding to the target capacity prediction result based on the target optimization algorithm and the multi-objective optimization model includes: Based on the objective optimization algorithm and the multi-objective optimization model, a power adjustment scheme corresponding to the target capacity prediction result is determined; Simulation verification is performed based on the power adjustment scheme to determine whether the power adjustment scheme meets the preset adjustment target; If the power adjustment scheme does not meet the preset adjustment target, a reacquisition request is initiated to obtain a new target adjustment request, and after obtaining the new target adjustment request, the process jumps to the step of using the preliminary capacity prediction model to determine the initial predicted capacity of the target distribution network based on the pre-acquired target adjustment request. If the power adjustment scheme meets the preset adjustment target, then the power adjustment scheme is determined as the target power adjustment scheme.

8. A power distribution network openable capacity prediction device, characterized in that, include: The capacity curve determination module is used to acquire first grid multi-source data of the target distribution network, integrate the first grid multi-source data to obtain corresponding integrated data, and determine the target capacity curve corresponding to the target distribution network based on the integrated data and the first load capacity corresponding to the target distribution network. The multi-source data of the power grid includes load data, output data of each distributed power source of the target distribution network, meteorological data, and power grid topology data; the first load capacity is the maximum load capacity of the target distribution network under preset static conditions; the target capacity curve is the dynamic curve of the available capacity of the target distribution network. The initial capacity prediction module is used to construct a preliminary capacity prediction model corresponding to the target distribution network based on the target capacity curve, and use the preliminary capacity prediction model to determine the initial predicted capacity corresponding to the target distribution network based on the pre-acquired target adjustment request. The target adjustment request is a request to adjust the distributed power sources of the target distribution network; The prediction result acquisition module is used to acquire second grid multi-source data of the target distribution network, and based on the second grid multi-source data and the initial predicted capacity corresponding to the target adjustment request, use the target prediction model to obtain the target capacity prediction result of the target distribution network in the target time period; the target prediction model is a prediction model built based on a long short-term memory network, and the target capacity prediction result includes the load prediction result corresponding to the target distribution network, the output prediction result of the distributed power source, and the available capacity prediction result corresponding to the target distribution network.

9. An electronic device, characterized in that, include: Memory, used to store computer programs; A processor for executing the computer program to implement the distribution network open capacity prediction method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, Used to store a computer program, wherein the computer program, when executed by a processor, implements the distribution network open capacity prediction method as described in any one of claims 1 to 7.

Images