Distributed photovoltaic bearing capacity evaluation method, device, equipment, medium and product

Abstract

The application discloses a kind of distributed photovoltaic carrying capacity evaluation method, device, equipment, medium and product, it is related to electric power system planning and new energy grid-connected field, the method includes obtaining multi-source operation data, pre-processing is carried out to multi-source operation data, obtains standard time series data, respectively establishes main network safety constraint condition and distribution network operation constraint condition, and coupling processing is carried out to main network safety constraint condition and distribution network operation constraint condition, generates collaborative constraint matrix, based on standard time series data and collaborative constraint matrix, time series power flow iteration calculation is carried out according to time series unit, obtains initial photovoltaic carrying capacity, through the long short-term memory network of pre-training, initial photovoltaic carrying capacity is dynamically revised, and target photovoltaic carrying capacity is obtained, based on target photovoltaic carrying capacity, generates distributed photovoltaic carrying capacity evaluation result.The application realizes the precise quantification and engineering application of grid photovoltaic carrying capacity.

Description

Technical Field

[0001] This application relates to the fields of power system planning and new energy grid connection technology, and in particular to a method, device, equipment, medium and product for assessing the carrying capacity of distributed photovoltaic power. Background Technology

[0002] Currently, with the deepening of the "dual carbon" goals and the accelerated construction of new power systems, distributed photovoltaic (PV) power generation is experiencing explosive growth. Its large-scale grid connection poses a severe challenge to the safe and stable operation of the power grid and the efficient absorption of new energy. Accurately assessing the power grid's carrying capacity for distributed PV has become a fundamental and critical technical issue that urgently needs to be addressed in power grid planning, PV grid connection approval, and distribution network upgrading and transformation.

[0003] Currently, existing distributed photovoltaic carrying capacity assessment technologies have the following main shortcomings: Existing assessment methods mostly rely on local short-term operating data or typical daily data for modeling and analysis, which fails to effectively cover the time-series fluctuation characteristics throughout the year. In addition, the data collection dimensions are limited, and there is a general lack of unified acquisition and integration of multi-source information from the main grid, distribution grid, photovoltaic, and load sides. This makes it difficult to truly reflect the dynamic operating characteristics of each link of the source, grid, and load, which fundamentally restricts the comprehensiveness and accuracy of the assessment results.

[0004] Current assessment methods generally treat main grid security constraints and distribution network operation constraints as two independent issues for calculation and analysis, lacking a coordinated coupling mechanism between the main grid and distribution network. This isolated analysis approach easily leads to a contradiction where the distribution network assessment considers the conditions for grid connection to be met, but the main grid cannot effectively absorb these conditions due to global security constraints such as transmission capacity limits and voltage deviations at key nodes. The assessment results cannot truly reflect the actual acceptance capacity of the power grid under the premise of global security, severely restricting the effective utilization of the power grid's security margin.

[0005] Traditional assessment models often employ static or quasi-static calculation methods based on typical operating modes. These methods are difficult to effectively respond to the strong intermittency and random fluctuations of photovoltaic output due to weather conditions, and they also cannot adapt to the dynamic changes in load at different times and seasons. This results in a large discrepancy between the assessment results and the actual operating status, and fails to meet the operational requirements for dynamic tracking and continuous assessment of carrying capacity in scenarios with a high proportion of renewable energy access.

[0006] Traditional power flow calculation methods driven by pure physical mechanisms inevitably introduce inherent errors during model simplification and boundary condition approximation. These errors accumulate during time-series iterations. Furthermore, the lack of an effective data-driven correction mechanism makes it difficult to further improve the evaluation accuracy.

[0007] In summary, existing distributed photovoltaic (PV) carrying capacity assessment technologies generally suffer from multiple defects, such as single data sources, isolated constraints of the main and distribution networks, static and rigid models, large calculation errors, and results that are difficult to support decision-making. As a result, the comprehensiveness, coordination, dynamism, accuracy, and engineering practicality of the assessment results are significantly insufficient, making it difficult to meet the actual needs of grid safety and stable operation and efficient consumption of new energy under the background of high proportion of distributed PV access. Summary of the Invention

[0008] The purpose of this application is to provide a method, device, equipment, medium, and product for assessing the carrying capacity of distributed photovoltaic (PV) power. Through the synergistic effect of multi-source data fusion, main and distribution network coordinated constraint coupling, time-series power flow iterative calculation, and long short-term memory network dynamic correction, it solves the technical problems of single data, isolated constraints, static models, large errors, and difficulty in supporting decision-making in traditional distributed PV carrying capacity assessment. It realizes a carrying capacity assessment with full cycle, high precision, dynamic continuity, and engineering output capability, and achieves accurate quantification and engineering application of grid PV carrying capacity under the background of high proportion of new energy access.

[0009] To achieve the above objectives, this application provides the following solution: Firstly, this application provides a method for assessing the carrying capacity of distributed photovoltaic power, including: Acquire multi-source operation data, which includes main grid side operation data, distribution network side operation data, photovoltaic side operation data, and load side operation data; The multi-source operational data is preprocessed to obtain standard time-series data. The preprocessing includes outlier removal, missing value completion, unit normalization, and time-series alignment. Based on the standard time-series data, main network security constraints and distribution network operation constraints are established respectively, and the main network security constraints and distribution network operation constraints are coupled to generate a collaborative constraint matrix. Based on the standard time series data and the cooperative constraint matrix, time series power flow iterative calculation is performed according to time series units to obtain the initial photovoltaic carrying capacity under each time series unit; The initial photovoltaic carrying capacity is dynamically corrected by a pre-trained long short-term memory network to obtain the target photovoltaic carrying capacity; Based on the target photovoltaic carrying capacity, a distributed photovoltaic carrying capacity assessment result is generated.

[0010] Optionally, the coupling process of the main network security constraints and the distribution network operation constraints to generate a collaborative constraint matrix specifically includes: A power flow transmission model for the main and distribution networks is established, using the main and distribution network interconnection nodes as the power interaction interface. Based on the main network power flow transmission model, the main network security constraints and the distribution network operation constraints are associated to form the collaborative constraint matrix; When the main network security constraints conflict with the distribution network operation constraints, the main network security constraints take precedence, and the distribution network operation constraints are reduced by the main network constraint correction coefficient to update the collaborative constraint matrix.

[0011] Optionally, the step of performing time-series power flow iterative calculations based on standard time-series data and the cooperative constraint matrix to obtain the initial photovoltaic carrying capacity under each time-series unit specifically includes: Using a preset duration as a time series unit, the annual data is divided into multiple typical operating scenarios based on meteorological conditions, load levels, and photovoltaic power output intensity; For each typical operating scenario, the Newton-Raphson method is used to iteratively solve the power flow equations using the aforementioned collaborative constraint matrix as boundary conditions, and the node voltage, line power, transformer load rate and distribution area voltage deviation are calculated time-by-time. The photovoltaic access capacity is gradually increased according to the preset step size until the voltage exceeds the limit, the line is overloaded or the transformer is overcapacitated, at which point the iterative calculation stops and the photovoltaic access capacity calculated in the previous step is used as the initial photovoltaic carrying capacity of the time series unit.

[0012] Optionally, the step of dynamically correcting the initial photovoltaic carrying capacity using a pre-trained long short-term memory network to obtain the target photovoltaic carrying capacity specifically includes: The initial photovoltaic load-bearing capacity is input into a pre-trained long short-term memory network to obtain the load-bearing capacity correction coefficient output by the long short-term memory network; The initial photovoltaic bearing capacity is weighted and adjusted according to the bearing capacity correction coefficient to obtain the target photovoltaic bearing capacity.

[0013] Optionally, the pre-training steps of the Long Short-Term Memory network include: Obtain historical operation datasets, which include historical time-series operation data, historical load data, historical photovoltaic output data, grid topology parameters, historical constraints, and historical evaluation biases; Using historical time-series operation data, historical load data, historical photovoltaic output data, power grid topology parameters, historical constraints, and historical evaluation deviations from the historical operation dataset as input features, and carrying capacity correction coefficient as output label, a training sample set is constructed. A long short-term memory network is constructed, which includes an input layer, a long short-term memory layer, a dropout layer, a fully connected layer, and an output layer connected in sequence. Using mean squared error as the loss function, the training sample set is input into the long short-term memory network for iterative training until the loss value is less than a preset loss threshold and the evaluation error is less than a preset error threshold, at which point training stops, and the trained long short-term memory network is obtained.

[0014] Optionally, generating distributed photovoltaic carrying capacity assessment results based on the target photovoltaic carrying capacity specifically includes: Based on the target photovoltaic carrying capacity, the constrained sensitive points are automatically identified. The constrained sensitive points include the main grid bottleneck section, the distribution network heavy-load feeder, the low-voltage distribution area and the overloaded transformer. The constrained sensitive points are sorted according to the degree of violation and the scope of impact to generate a priority list for power grid renovation. The target photovoltaic carrying capacity, the constrained sensitive points, and the grid transformation priority list are summarized according to four levels: distribution area level, feeder level, substation level, and regional level to generate the distributed photovoltaic carrying capacity assessment result.

[0015] Secondly, this application provides a distributed photovoltaic carrying capacity assessment device, comprising: The data acquisition module is used to acquire multi-source operation data, which includes main grid side operation data, distribution network side operation data, photovoltaic side operation data and load side operation data; The data preprocessing module is used to preprocess the multi-source running data to obtain standard time series data. The preprocessing includes outlier removal, missing value completion, unit normalization, and time series alignment. The constraint coupling module is used to establish main network security constraints and distribution network operation constraints based on the standard time series data, and to couple the main network security constraints and the distribution network operation constraints to generate a collaborative constraint matrix. The power flow calculation module is used to perform time-series power flow iterative calculations based on the standard time-series data and the cooperative constraint matrix, and to obtain the initial photovoltaic carrying capacity under each time-series unit. The dynamic correction module is used to dynamically correct the initial photovoltaic carrying capacity through a pre-trained long short-term memory network to obtain the target photovoltaic carrying capacity. The evaluation output module is used to generate distributed photovoltaic carrying capacity evaluation results based on the target photovoltaic carrying capacity.

[0016] Thirdly, this application provides a computer device, including: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the distributed photovoltaic carrying capacity assessment method described in any one of the above.

[0017] Fourthly, this application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the distributed photovoltaic carrying capacity assessment method described above.

[0018] Fifthly, this application provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the distributed photovoltaic carrying capacity assessment method described above.

[0019] According to the specific embodiments provided in this application, the following technical effects are disclosed: This application provides a method, apparatus, equipment, medium, and product for assessing the carrying capacity of distributed photovoltaic power. By acquiring multi-source operating data through the step of “acquiring multi-source operating data”, it solves the problem that traditional assessment methods rely on a single data source and only local short-term data, failing to cover multi-dimensional information of the source, grid, and load. It realizes unified perception and complete collection of operating data across the entire domain, including the main grid, distribution grid, photovoltaic, and load sides, laying a multi-dimensional data foundation covering the time-series characteristics of the whole year for subsequent accurate assessment of carrying capacity. By preprocessing multi-source operational data to obtain standard time-series data and establishing main grid security constraints and distribution network operational constraints based on the standard time-series data, and coupling these constraints to generate a collaborative constraint matrix, the system addresses issues such as outliers, missing items, inconsistent dimensions, and time-series misalignment in raw data due to factors like equipment accuracy, environmental interference, and transmission failures. This achieves high-quality cleaning, standardized mapping, and strict time-series alignment of multi-source heterogeneous data, forming a reliable data foundation directly usable for constraint modeling and power flow calculation. It breaks down the technical barriers of independent and isolated calculation of main grid and distribution network constraints in traditional assessments, resolving the contradiction of allowing access on the distribution network side while the main grid side cannot effectively absorb it. It achieves deep coupling and collaborative modeling of global main grid security constraints and local distribution network operational constraints. When constraints conflict, main grid security takes priority and is automatically reduced, ensuring that the assessment results simultaneously meet the dual requirements of global security and local acceptance. The step "based on standard time-series data and a collaborative constraint matrix, perform iterative calculations of time-series power flow by time-series unit to obtain the initial photovoltaic carrying capacity under each time-series unit" solves the problem that traditional static assessment models are unable to respond to the intermittency, randomness, and dynamic time-varying characteristics of photovoltaic output. It achieves refined decomposition of annual operating data at the time-series granularity and iterative solution of power flow for each time period, enabling the initial carrying capacity calculation results to fully cover the actual operating state of the power grid under different meteorological conditions and different load levels, significantly improving the time-series adaptability and dynamic response capability of the assessment. The step "dynamically correcting the initial photovoltaic carrying capacity using a pre-trained long short-term memory network to obtain the target photovoltaic carrying capacity" solves the problem of inherent error accumulation introduced by model simplification, parameter approximation, and idealization of boundary conditions in time-series power flow calculation. It realizes the use of the strong modeling capability of long short-term memory networks for time-series characteristics to accurately compensate for errors in the initial carrying capacity, significantly improving the accuracy and reliability of the quantitative results of carrying capacity.By generating distributed photovoltaic carrying capacity assessment results based on target photovoltaic carrying capacity, the problem of traditional assessment output being singular and unable to directly support power grid planning and grid connection approval decisions is solved. It realizes the output of comprehensive assessment results, including the maximum allowable capacity of each access point, constraint bottleneck identification, upgrade priority list, and four-level zoning assessment conclusions. This enables the assessment conclusions to directly serve actual business scenarios such as photovoltaic grid connection approval, distribution network upgrade and transformation, and source-load coordinated scheduling. It realizes the accurate quantification and engineering application of the grid photovoltaic carrying capacity under the background of high proportion of new energy access. Attached Figure Description

[0020] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0021] Figure 1 This is an application environment diagram of a distributed photovoltaic carrying capacity assessment method according to an embodiment of this application; Figure 2 A flowchart illustrating a distributed photovoltaic carrying capacity assessment method provided in an embodiment of this application; Figure 3 A schematic diagram of the functional modules of a distributed photovoltaic carrying capacity assessment device provided in an embodiment of this application; Figure 4 This is a schematic diagram of the structure of a computer device provided in an embodiment of this application. Detailed Implementation

[0022] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments 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] To make the above-mentioned objectives, features and advantages of this application more apparent and understandable, the application will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0024] The distributed photovoltaic carrying capacity assessment method provided in this application embodiment can be applied to, for example, Figure 1In the application environment shown, the data acquisition terminal 102 communicates with the server 104 via a network. The data storage system can store the data that the server 104 needs to process, such as multi-source operational data and long short-term memory network training sample data. The data storage system can be set up separately, integrated into the server 104, or placed in the cloud or on other servers. The data acquisition terminal 102 can send the collected multi-source operational data to the server 104. After receiving the multi-source operational data, the server 104 preprocesses the acquired multi-source operational data to obtain standard time-series data; based on the standard time-series data, it establishes main grid security constraints and distribution network operational constraints, and performs coupling processing to generate a collaborative constraint matrix; based on the standard time-series data and the collaborative constraint matrix, it performs time-series power flow iterative calculations according to time-series units to obtain the initial photovoltaic carrying capacity under each time-series unit; it dynamically corrects the initial photovoltaic carrying capacity through a pre-trained long short-term memory network to obtain the target photovoltaic carrying capacity; and based on the target photovoltaic carrying capacity, it generates a distributed photovoltaic carrying capacity assessment result. Server 104 can feed back the obtained distributed photovoltaic carrying capacity assessment results to data acquisition terminal 102 or third-party systems such as dispatch centers and planning departments. Furthermore, in some embodiments, the distributed photovoltaic carrying capacity assessment method can also be implemented independently by server 104 or data acquisition terminal 102. For example, data acquisition terminal 102 with computing capabilities can directly perform carrying capacity assessment processing on the collected multi-source operating data, or server 104 can obtain multi-source operating data from the data storage system and perform carrying capacity assessment processing on the obtained multi-source operating data.

[0025] The data acquisition terminal 102 can be, but is not limited to, various Supervisory Control and Data Acquisition (SCADA) terminals, power distribution automation terminals, phasor measurement units (PMUs), smart meters, photovoltaic grid-connected monitoring terminals, desktop computers, laptops, server workstations, etc. The server 104 can be implemented using a standalone server or a server cluster composed of multiple servers, or it can be a cloud server.

[0026] In one exemplary embodiment, such as Figure 2 As shown, a method for assessing the carrying capacity of distributed photovoltaic power is provided. This method is executed by computer equipment, specifically by a terminal or server alone, or by both a terminal and a server. In this embodiment, the method is applied to... Figure 1 Taking server 104 as an example, the explanation includes the following steps 201 to 206. Wherein: Step 201: Obtain multi-source operation data, which includes main grid side operation data, distribution network side operation data, photovoltaic side operation data, and load side operation data; Specifically, based on the actual operational needs of the power grid, this approach breaks away from the traditional single-dimensional data collection model and collects comprehensive operational information from four dimensions: the main grid side, the distribution network side, the photovoltaic side, and the load side. This acquires multi-source operational data and constructs a complete data foundation covering power generation, grid, load, and storage. The multi-source operational data includes at least the following: The main grid side operation data includes: 110kV substation bus voltage amplitude, line current, active power, reactive power, transmission section limit, hub node voltage deviation, equipment rated parameters, and topology correlation matrix.

[0027] The distribution network operation data includes: 10kV feeder operating electrical quantities, 0.4kV transformer substation voltage, current, transformer load rate, three-phase unbalance, and total harmonic distortion rate.

[0028] The photovoltaic side operation data includes: installed capacity, inverter control parameters, grid connection point coordinates, real-time power output, solar irradiance, and ambient temperature.

[0029] Load-side operational data includes time-series load curves for industrial, commercial, residential, and agricultural categories.

[0030] The time resolution of the aforementioned multi-source operating data was determined to be 15 minutes to meet the subsequent timing synchronization requirements.

[0031] Step 202: Preprocess the multi-source operational data to obtain standard time series data. The preprocessing includes outlier removal, missing value completion, unit normalization, and time series alignment. Specifically, the raw data collected on-site is affected by factors such as the accuracy of the acquisition equipment, environmental interference, and transmission failures. The raw data may contain outliers, missing items, inconsistent dimensions, and time series misalignments. If it is used directly for calculation, it will seriously affect the accuracy and reliability of the evaluation. Therefore, it is necessary to complete the data quality control through a standardized preprocessing process to form standard time series data that meets the calculation requirements.

[0032] Specifically, outlier identification is performed on voltage, current, and power sequences. Based on statistical principles, a discrimination rule is established: if a data point deviates from a reasonable range between the sequence mean and standard deviation, the data is identified as outlier and removed to avoid interfering with the calculation results. The sequence mean is... The standard deviation is .

[0033] To address missing data segments, weighted imputation based on similar days from the same period is used to fill in the missing sections in the removed data. The complete data sequence is then reconstructed using the imputation formula: ; in, These are supplementary values ​​for missing data points. This represents the data for the kth similar day in the same period.

[0034] After anomaly removal and missing data imputation, the multi-source operational data is normalized to map data of different dimensions and orders of magnitude to a standard interval, eliminating calculation biases caused by numerical differences. The normalization formula is: ; in, The original data values, and These are the minimum and maximum values ​​of the data sequence, respectively.

[0035] Strict time-series alignment is performed on the normalized main grid, distribution grid, photovoltaic, and load-side data to eliminate data with mismatched timestamps and form standard time-series data, providing high-quality data input for subsequent constraint modeling and power flow calculation.

[0036] Step 203: Based on standard time-series data, establish main grid security constraints and distribution network operation constraints respectively, and couple the main grid security constraints and distribution network operation constraints to generate a collaborative constraint matrix; Specifically, in order to overcome the technical challenges of independent and fragmented constraints and insufficient coordination between the main and distribution networks in traditional assessments, this step establishes main network security constraints and distribution network operation constraints separately through a hierarchical quantification approach.

[0037] Establish mainnet security constraint functions The constraints included are: (1) The operating current of the transmission line shall not exceed the rated current carrying capacity; (2) The load factor does not exceed the upper limit; (3) The voltage deviation of the 110kV hub nodes shall be controlled within ±5%; (4) The cross-sectional transmission power shall not exceed the dispatch limit; (5) The system retains sufficient safety and stability reserves to prevent voltage collapse and power angle instability, and maintains voltage stability margin.

[0038] Meanwhile, the mainnet security constraints prioritize global security as the top priority.

[0039] Establish distribution network operation constraint functions The constraints included are: (1) The 10kV feeder current shall not exceed the rated value, and the load rate shall not exceed 90%; (2) The load rate of the distribution transformer shall not exceed 85%; (3) The voltage deviation of the 0.4kV distribution area is controlled within ±7%; (4) The three-phase imbalance is no greater than 4%; (5) The total harmonic distortion rate of the public network is no greater than 5%; (6) The power factor at the photovoltaic grid connection point shall not be lower than 0.9 to ensure that there is no long-term backflow of power that would cause the protection limit to be exceeded.

[0040] As an optional implementation, the main network security constraints and distribution network operation constraints are coupled to generate a collaborative constraint matrix, specifically including: A power flow transmission model for the main and distribution networks is established, using the main and distribution network interconnection nodes as the power interaction interface. Based on the main grid and distribution network power flow transmission model, the main grid security constraints and distribution network operation constraints are associated to form a collaborative constraint matrix; When the main grid security constraints conflict with the distribution network operation constraints, the main grid security constraints take priority, and the distribution network operation constraints are reduced by the main grid constraint correction coefficient to update the collaborative constraint matrix.

[0041] Step 204: Based on standard time series data and collaborative constraint matrix, perform time series power flow iterative calculation by time series unit to obtain the initial photovoltaic carrying capacity under each time series unit; As an optional implementation method, the annual data is divided into multiple typical operating scenarios based on meteorological conditions, load levels, and photovoltaic power output intensity, with a preset duration as a time series unit; For each typical operating scenario, the Newton-Raphson method is used to iteratively solve the power flow equations with the collaborative constraint matrix as the boundary condition, and the node voltage, line power, transformer load rate and distribution area voltage deviation are calculated time-by-time. The photovoltaic access capacity is gradually increased according to the preset step size until the voltage exceeds the limit, the line is overloaded or the transformer is overcapacitated, at which point the iterative calculation stops and the photovoltaic access capacity calculated in the previous step is used as the initial photovoltaic carrying capacity of the time series unit.

[0042] Specifically, to adapt to the time-varying characteristics of power grid operation, a preset duration (e.g., 15 minutes) is used as a time series unit. The annual data is divided into multiple typical operating scenarios based on meteorological conditions (sunny days with strong sunlight, cloudy days with weak sunlight, rainy days with no sunlight), load levels (peak load, flat load, low load), and photovoltaic output intensity. Through time series decomposition, the continuous operating state is transformed into discrete computing units that can be calculated independently.

[0043] For each typical operating scenario, the Newton-Raphson method is used to iteratively solve the power flow equations with the collaborative constraint matrix as the boundary condition, and the node voltage, line power, transformer load rate and distribution area voltage deviation are accurately calculated for each time period.

[0044] Assuming all calculated indicators do not exceed limits, the photovoltaic (PV) access capacity is gradually increased according to preset step sizes, with the step size set at 50kW for the distribution area level, 500kW for the feeder level, and 5MW for the substation level. Iterative calculations stop when voltage exceeds limits, line overloads, or transformer overcapacity occurs, and the PV access capacity calculated in the previous step is used as the initial PV carrying capacity for that time series unit.

[0045] Step 205: The initial photovoltaic carrying capacity is dynamically corrected using a pre-trained long short-term memory network to obtain the target photovoltaic carrying capacity; Specifically, the initial photovoltaic carrying capacity obtained from the time-series power flow calculation is input into a pre-trained Long Short-Term Memory (LSTM) network to obtain the carrying capacity correction coefficient output by the LSTM network. According to the bearing capacity correction factor The initial photovoltaic carrying capacity is weighted and adjusted to obtain the target photovoltaic carrying capacity.

[0046] As an optional implementation, the pre-training steps of the Long Short-Term Memory (LSTM) network include: Obtain historical operating datasets, which include historical time-series operating data, historical load data, historical photovoltaic output data, grid topology parameters, historical constraints, and historical assessment biases.

[0047] Using historical time-series operational data, historical load data, historical photovoltaic output data, grid topology parameters, historical constraints, and historical assessment biases from the historical operational dataset as input features, and the carrying capacity correction coefficient as the output label, a training sample set is constructed. The training sample set contains no less than 3 years of historical operational data and is divided into a training set, a validation set, and a test set in a 7:2:1 ratio.

[0048] Construct a Long Short-Term Memory (LSTM) network, which consists of an input layer, an LSTM layer, a Dropout layer, a fully connected layer, and an output layer connected sequentially. Set the learning rate (lr), batch size (batch_size), and number of iterations (epochs).

[0049] Using the mean squared error (MSE) as the loss function, the training sample set is input into the long short-term memory network for iterative training until the model loss value is less than a preset loss threshold (e.g., 0.001) and the evaluation error is less than a preset error threshold (e.g., 5%). Training is then stopped, and the trained long short-term memory network is obtained.

[0050] Step 206: Based on the target photovoltaic carrying capacity, generate the distributed photovoltaic carrying capacity assessment results.

[0051] As an optional implementation method, based on the target photovoltaic carrying capacity, constrained sensitive points are automatically identified. These constrained sensitive points include main grid bottleneck sections, heavily loaded distribution network feeders, low-voltage distribution areas, and overloaded transformers. The constrained sensitive points are then ranked according to the degree of exceeding limits and the scope of impact, forming a priority list for grid upgrades.

[0052] The target photovoltaic carrying capacity, constrained sensitive points, and grid transformation priority lists are summarized at four levels: distribution area level, feeder level, substation level, and regional level to generate distributed photovoltaic carrying capacity assessment results. The assessment results include the maximum allowable grid connection capacity of distributed photovoltaics for each time unit, each grid connection point, each feeder, and each distribution area, as well as the grid transformation priority list, which can be directly used for planning and design, grid connection system review, and project implementation.

[0053] Implementing steps 201 to 206 addresses the problem of traditional assessment methods relying on a single data source and limited to short-term local data, failing to cover multi-dimensional information about the source, grid, and load. It achieves unified perception and complete collection of operational data across the entire network, including the main grid, distribution grid, photovoltaic grid, and load sides, laying a multi-dimensional data foundation covering year-round temporal characteristics for subsequent accurate capacity assessment. It also resolves issues such as outliers, missing items, inconsistent dimensions, and time-series misalignments in raw data due to factors like equipment accuracy, environmental interference, and transmission failures. This results in high-quality cleaning, standardized mapping, and strict time-series alignment of multi-source heterogeneous data, forming a reliable data foundation directly usable for constraint modeling and power flow calculation. Furthermore, it breaks down the technical barriers of independent and isolated calculation of main grid and distribution grid constraints in traditional assessments, resolving the contradiction of allowing distribution grid access while the main grid cannot effectively absorb it. It achieves deep coupling and collaborative modeling of global security constraints of the main grid and local operational constraints of the distribution grid. When constraints conflict, automatic reduction prioritizes main grid security, ensuring that the assessment results simultaneously meet the dual requirements of global security and local acceptance. Finally, it addresses the limitations of traditional static assessment models. In response to the intermittent, random, and dynamically time-varying characteristics of photovoltaic (PV) power output, this system achieves refined decomposition of annual operating data at the time-series granularity and iterative solution of power flow in each time period. This ensures that the initial carrying capacity calculation results can fully cover the actual operating status of the power grid under different meteorological conditions and load levels, significantly improving the time-series adaptability and dynamic response capability of the assessment. It also solves the inherent error accumulation problem introduced by model simplification, parameter approximation, and idealization of boundary conditions in time-series power flow calculation. By utilizing the strong modeling capability of long short-term memory networks for time-series characteristics, it achieves accurate error compensation for the initial carrying capacity, significantly improving the accuracy and reliability of the quantitative results of carrying capacity. Furthermore, it addresses the problem of traditional assessment output being singular and unable to directly support power grid planning and grid connection approval decisions. It achieves comprehensive assessment results output, including the maximum allowable capacity of each access point, bottleneck identification, upgrade priority list, and four-level zoning assessment conclusions. This allows the assessment conclusions to directly serve practical business scenarios such as PV grid connection approval, distribution network upgrades, and source-load coordinated scheduling, realizing the accurate quantification and engineering application of the grid's PV carrying capacity under the background of high-proportion renewable energy access.

[0054] In another exemplary embodiment of this application, a verification step is further included in addition to the foregoing embodiments.

[0055] To ensure the reliability and generalizability of the model under different power grid application scenarios, three typical scenarios—urban distribution network, rural distribution network, and industrial park distribution network—were selected as verification objects, representing application environments with high load density, weak grid structure, and severe load fluctuations, respectively.

[0056] Establish a verification indicator system, which includes: assessment error not exceeding 5%, voltage over-limit rate, line overload rate, and transformer overload rate.

[0057] The model's output load-bearing capacity results are compared with actual on-site grid-connected measured data. Simultaneously, power flow simulation verification is conducted using OpenDSS (Open Distribution System Simulator) or DIgSILENT (Digital Simulator for Electrical Network) to comprehensively evaluate errors, the number of overloads, and equipment overload conditions, generating a complete verification report. When all indicators meet the preset standards, the model is deemed to have engineering applicability and can be used for actual grid-connected assessment and planning design work.

[0058] In another exemplary embodiment of this application, an online dynamic evaluation step is further included in addition to the foregoing embodiments.

[0059] The evaluation model is integrated with SCADA systems, distribution automation systems, and photovoltaic grid-connected monitoring systems. It acquires real-time power grid operation data with a sampling period of 15 minutes, automatically completes data cleaning, constraint updates, carrying capacity calculations, and result output, thus upgrading the evaluation model from offline calculation to online application.

[0060] Based on adjustments to grid operation modes, equipment maintenance status, new photovoltaic access capacity, or load changes, the collaborative constraint matrix is ​​updated in real time to dynamically correct the photovoltaic carrying capacity of each region, supporting grid connection approval, real-time scheduling, and operation mode arrangement.

[0061] An assessment report is generated, which includes details such as the spatiotemporal distribution of carrying capacity, a list of constraint bottlenecks, recommended access capacity, priority of power grid transformation, analysis of absorption potential, and economic benefit calculation. It can be directly used for planning and design, access system review, and project implementation.

[0062] This application also provides an application scenario in which the above-mentioned distributed photovoltaic (PV) carrying capacity assessment method is applied. Specifically, the distributed PV carrying capacity assessment method provided in this embodiment can be applied in the distributed PV grid connection approval and grid planning scenario. Distributed PV projects enter the grid access assessment processing link from the application stage, and obtain corresponding carrying capacity assessment results and access scheme suggestions through human-machine collaboration, before entering the downstream grid connection approval and grid planning and modification stages. The distributed PV carrying capacity assessment method provided in this embodiment belongs to the machine assessment stage in the PV grid connection access assessment process. Specifically, in the assessment process for PV grid connection access, the carrying capacity of the PV grid connection access point can be assessed based on a collaborative approach of machine assessment and manual review, thereby generating corresponding distributed PV carrying capacity assessment results, providing quantitative basis for grid connection approval decisions and grid planning and modification.

[0063] Based on the same inventive concept, this application also provides a distributed photovoltaic carrying capacity assessment device for implementing the distributed photovoltaic carrying capacity assessment method described above. The solution provided by this device is similar to the implementation scheme described in the above method; therefore, the specific limitations in one or more embodiments of the distributed photovoltaic carrying capacity assessment device provided below can be found in the limitations of the distributed photovoltaic carrying capacity assessment method described above, and will not be repeated here.

[0064] In one exemplary embodiment, such as Figure 3 As shown, a distributed photovoltaic carrying capacity assessment device is provided, the device comprising: The data acquisition module is used to acquire multi-source operation data, including main grid side operation data, distribution network side operation data, photovoltaic side operation data and load side operation data; The data preprocessing module is used to preprocess multi-source runtime data to obtain standard time series data. The preprocessing includes outlier removal, missing value completion, unit normalization, and time series alignment. The constraint coupling module is used to establish main grid security constraints and distribution network operation constraints based on standard time series data, and to couple the main grid security constraints and distribution network operation constraints to generate a collaborative constraint matrix. The power flow calculation module is used to perform iterative calculations of time-series power flow based on standard time-series data and cooperative constraint matrices, and to obtain the initial photovoltaic carrying capacity under each time-series unit. The dynamic correction module is used to dynamically correct the initial photovoltaic carrying capacity through a pre-trained long short-term memory network to obtain the target photovoltaic carrying capacity. The evaluation output module is used to generate distributed photovoltaic carrying capacity evaluation results based on the target photovoltaic carrying capacity.

[0065] In one exemplary embodiment, a computer device is provided, which may be a server or a terminal, and its internal structure diagram may be as follows. Figure 4 As shown, this computer device includes a processor, memory, input / output (I / O) interfaces, and a communication interface. The processor, memory, and I / O interfaces are connected via a system bus, and the communication interface is also connected to the system bus via the I / O interfaces. The processor provides computational and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and a database. The internal memory provides the environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The database stores multi-source operational data and long short-term memory network training sample data, etc. The I / O interfaces are used for exchanging information between the processor and external devices. The communication interface is used for communication with external terminals via a network connection. When the computer program is executed by the processor, it implements a method for assessing the carrying capacity of distributed photovoltaic power generation.

[0066] Those skilled in the art will understand that Figure 4 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0067] In one exemplary embodiment, a computer device is also provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the above-described method embodiments.

[0068] In one exemplary embodiment, a computer-readable storage medium is provided storing a computer program that, when executed by a processor, implements the steps in the above-described method embodiments.

[0069] In one exemplary embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps in the above-described method embodiments.

[0070] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data must comply with relevant regulations.

[0071] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments described above. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM).

[0072] The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0073] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0074] This document uses specific examples 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. Furthermore, those skilled in the art will recognize that, based on the ideas of this application, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A method for assessing the carrying capacity of distributed photovoltaic power, characterized in that, The method for assessing the carrying capacity of distributed photovoltaic power includes: Acquire multi-source operation data, which includes main grid side operation data, distribution network side operation data, photovoltaic side operation data, and load side operation data; The multi-source operational data is preprocessed to obtain standard time-series data. The preprocessing includes outlier removal, missing value completion, unit normalization, and time-series alignment. Based on the standard time-series data, main network security constraints and distribution network operation constraints are established respectively, and the main network security constraints and distribution network operation constraints are coupled to generate a collaborative constraint matrix. Based on the standard time series data and the cooperative constraint matrix, time series power flow iterative calculation is performed according to time series units to obtain the initial photovoltaic carrying capacity under each time series unit; The initial photovoltaic carrying capacity is dynamically corrected by a pre-trained long short-term memory network to obtain the target photovoltaic carrying capacity; Based on the target photovoltaic carrying capacity, a distributed photovoltaic carrying capacity assessment result is generated.

2. The method for assessing the carrying capacity of distributed photovoltaic power according to claim 1, characterized in that, The coupling process between the main network security constraints and the distribution network operation constraints to generate a collaborative constraint matrix specifically includes: A power flow transmission model for the main and distribution networks is established, using the main and distribution network interconnection nodes as the power interaction interface. Based on the main network power flow transmission model, the main network security constraints and the distribution network operation constraints are associated to form the collaborative constraint matrix; When the main network security constraints conflict with the distribution network operation constraints, the main network security constraints take precedence, and the distribution network operation constraints are reduced by the main network constraint correction coefficient to update the collaborative constraint matrix.

3. The method for assessing the carrying capacity of distributed photovoltaic power according to claim 1, characterized in that, Based on standard time-series data and the cooperative constraint matrix, iterative calculations of time-series power flow are performed on a time-series basis to obtain the initial photovoltaic carrying capacity under each time-series unit, specifically including: Using a preset duration as a time series unit, the annual data is divided into multiple typical operating scenarios based on meteorological conditions, load levels, and photovoltaic power output intensity; For each typical operating scenario, the Newton-Raphson method is used to iteratively solve the power flow equations using the aforementioned collaborative constraint matrix as boundary conditions, and the node voltage, line power, transformer load rate and distribution area voltage deviation are calculated time-by-time. The photovoltaic access capacity is gradually increased according to the preset step size until the voltage exceeds the limit, the line is overloaded or the transformer is overcapacitated, at which point the iterative calculation stops and the photovoltaic access capacity calculated in the previous step is used as the initial photovoltaic carrying capacity of the time series unit.

4. The method for assessing the carrying capacity of distributed photovoltaic power according to claim 1, characterized in that, The step of dynamically correcting the initial photovoltaic carrying capacity using a pre-trained long short-term memory network to obtain the target photovoltaic carrying capacity specifically includes: The initial photovoltaic load-bearing capacity is input into a pre-trained long short-term memory network to obtain the load-bearing capacity correction coefficient output by the long short-term memory network; The initial photovoltaic bearing capacity is weighted and adjusted according to the bearing capacity correction coefficient to obtain the target photovoltaic bearing capacity.

5. The method for assessing the carrying capacity of distributed photovoltaic power according to claim 1, characterized in that, The pre-training steps of the Long Short-Term Memory network include: Obtain historical operation datasets, which include historical time-series operation data, historical load data, historical photovoltaic output data, grid topology parameters, historical constraints, and historical evaluation biases; Using historical time-series operation data, historical load data, historical photovoltaic output data, power grid topology parameters, historical constraints, and historical evaluation deviations from the historical operation dataset as input features, and carrying capacity correction coefficient as output label, a training sample set is constructed. A long short-term memory network is constructed, which includes an input layer, a long short-term memory layer, a dropout layer, a fully connected layer, and an output layer connected in sequence. Using mean squared error as the loss function, the training sample set is input into the long short-term memory network for iterative training until the loss value is less than a preset loss threshold and the evaluation error is less than a preset error threshold, at which point training stops, and the trained long short-term memory network is obtained.

6. The method for assessing the carrying capacity of distributed photovoltaic power according to claim 1, characterized in that, The process of generating distributed photovoltaic carrying capacity assessment results based on the target photovoltaic carrying capacity specifically includes: Based on the target photovoltaic carrying capacity, the constrained sensitive points are automatically identified. The constrained sensitive points include the main grid bottleneck section, the distribution network heavy-load feeder, the low-voltage distribution area and the overloaded transformer. The constrained sensitive points are sorted according to the degree of violation and the scope of impact to generate a priority list for power grid renovation. The target photovoltaic carrying capacity, the constrained sensitive points, and the grid transformation priority list are summarized according to four levels: distribution area level, feeder level, substation level, and regional level to generate the distributed photovoltaic carrying capacity assessment result.

7. A distributed photovoltaic carrying capacity assessment device, characterized in that, The distributed photovoltaic carrying capacity assessment device includes: The data acquisition module is used to acquire multi-source operation data, which includes main grid side operation data, distribution network side operation data, photovoltaic side operation data and load side operation data; The data preprocessing module is used to preprocess the multi-source running data to obtain standard time series data. The preprocessing includes outlier removal, missing value completion, unit normalization, and time series alignment. The constraint coupling module is used to establish main network security constraints and distribution network operation constraints based on the standard time series data, and to couple the main network security constraints and the distribution network operation constraints to generate a collaborative constraint matrix. The power flow calculation module is used to perform time-series power flow iterative calculations based on the standard time-series data and the cooperative constraint matrix, and to obtain the initial photovoltaic carrying capacity under each time-series unit. The dynamic correction module is used to dynamically correct the initial photovoltaic carrying capacity through a pre-trained long short-term memory network to obtain the target photovoltaic carrying capacity. The evaluation output module is used to generate distributed photovoltaic carrying capacity evaluation results based on the target photovoltaic carrying capacity.

8. A computer device, comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that the processor executes the computer program to implement the steps of the distributed photovoltaic carrying capacity assessment method according to any one of claims 1-6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by a processor, the computer program implements the steps of the distributed photovoltaic carrying capacity assessment method according to any one of claims 1-6.

10. A computer program product, comprising a computer program, characterized in that, When executed by a processor, the computer program implements the steps of the distributed photovoltaic carrying capacity assessment method according to any one of claims 1-6.

Images