Mountain area distribution network new energy installed capacity planning method, system, equipment and medium

Abstract

The invention relates to the technical field of power distribution networks, and discloses a mountainous area power distribution network new energy installed capacity planning method, system, device and medium, and the method comprises the steps: firstly collecting the network structure, climate and new energy and load historical operation data of a mountainous area power distribution network; performing region division on the nodes based on a principal component analysis method, and extracting time sequence characteristics of new energy output of each region by using an improved spectral clustering algorithm fusing graph structure information; a multi-scene new energy output model covering normal and extreme meteorological conditions is constructed; on the basis, establishing a distribution network operation constraint system comprising power flow, voltage, current and equipment output constraints, and constructing a comprehensive objective function taking full life cycle economy as an objective; and finally, solving a new energy installed capacity configuration scheme which meets the constraint and is optimal in economical efficiency through iterative optimization. According to the method, the planning robustness, the power supply reliability and the new energy consumption capability of the mountain area power distribution network under extreme conditions are effectively improved.

Description

Technical Field

[0001] This invention relates to the field of power distribution network technology, and in particular to a method, system, equipment and medium for planning the installed capacity of new energy power distribution networks in mountainous areas. Background Technology

[0002] Currently, there are already a wealth of research findings both domestically and internationally regarding the planning of renewable energy in distribution networks. Mainstream research methods typically base their approaches on typical daily scenarios for renewable energy, then construct distributed generation models to comprehensively consider constraints such as grid power flow, thereby establishing corresponding optimization models. These models are then used to solve problems with objectives such as minimizing installed costs, maximizing renewable energy output, or maximizing installed capacity.

[0003] However, the research methods mentioned above usually only consider the output of new energy sources in typical scenarios, ignoring the impact of extreme conditions on the operation of distributed power sources. In addition, there are some extreme conditions that require more system backup capacity to ensure power supply reliability. Traditional single-scenario planning methods are difficult to take into account the above complex factors. Summary of the Invention

[0004] In view of the aforementioned existing problems, the present invention is proposed.

[0005] Therefore, this invention provides a method, system, equipment, and medium for planning the installed capacity of new energy power distribution networks in mountainous areas, which can solve the problem of... Existing technologies only consider the output of new energy sources in typical scenarios, ignore the impact of extreme conditions, and traditional single-scenario planning methods are unable to take into account complex factors.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution: In a first aspect, the present invention provides a method for planning the installed capacity of new energy power distribution networks in mountainous areas, including: Collect network structure parameters, local climate parameters, and historical operating data of wind power and photovoltaic power output and load at each node of the target mountain power distribution network; Based on principal component analysis, the historical operating data is dimensionality reduced and the main spatiotemporal features are extracted, thereby dividing the power distribution network into several new energy regions. For each new energy region, an improved spectral clustering algorithm is used to extract features from the historical output data of new energy sources to obtain coded feature values ​​that reflect the temporal variation patterns. Based on the encoded feature values, the operating scenarios of each new energy region are clustered and classified to construct new energy output models corresponding to normal operating conditions and extreme weather conditions, respectively. A distribution network operation constraint model is established under the condition of new energy access. The constraint model includes power flow balance constraints, node voltage constraints, branch current constraints, generator output upper and lower limit constraints, and line power transmission constraints. A comprehensive objective function is constructed with the goal of achieving economic efficiency throughout the entire life cycle. This comprehensive objective function takes into account the investment cost of new energy construction, operation and maintenance cost, network loss cost, and revenue from new energy power generation. Under the premise of satisfying the power distribution network operation constraint model, the optimal new energy installed capacity configuration scheme that makes the comprehensive objective function optimal is obtained by iterative optimization.

[0007] As a preferred embodiment of the mountainous distribution network renewable energy installed capacity planning method described in this invention, the step of dividing the distribution network into several renewable energy regions based on principal component analysis to perform dimensionality reduction processing on the historical operating data and extract the main spatiotemporal features includes: Centralized processing of historical output data for photovoltaic or wind power; Calculate the covariance matrix of the processed data, and perform eigenvalue decomposition on the covariance matrix to obtain multiple principal components and their corresponding contribution rates; Based on a preset cumulative contribution rate threshold, several principal components are selected to form a low-dimensional feature space. The original node data is mapped to the low-dimensional feature space to obtain the principal component feature values ​​of each node; The nodes are clustered based on the similarity of the principal component eigenvalues ​​to form several photovoltaic or wind power regions.

[0008] As a preferred embodiment of the mountainous distribution network new energy installed capacity planning method described in this invention, the improved spectral clustering algorithm includes constructing an autoencoder structure, wherein the autoencoder consists of an encoder and a decoder, both of which are implemented using a long short-term memory neural network. The historical output data of new energy sources and the similarity graph structure constructed based on Gaussian kernel function are used as input, and the autoencoder learns the inherent low-dimensional representation of the data. During training, both reconstruction error and graph structure preservation loss are minimized simultaneously. The output is the encoded feature value generated by the encoder after training and convergence, which is used for subsequent scene clustering.

[0009] As a preferred embodiment of the mountainous power grid renewable energy installed capacity planning method described in this invention, the step of constructing renewable energy output models corresponding to normal operating conditions and extreme weather conditions includes: The number of cluster categories is set based on the encoded feature values, and the cluster centroids are initialized. Each data point is assigned to the category of the nearest centroid, and the centroid position of each category is updated; Repeat the allocation and update operations until the centroid converges to obtain several typical operating scenarios; Based on meteorological information and historical power output characteristics, the typical operating scenarios are divided into normal climate scenarios and extreme weather scenarios, and the occurrence probability of each scenario is statistically analyzed. Based on the historical power output curves for each type of scenario, a corresponding probability model for new energy power output is established.

[0010] As a preferred embodiment of the mountainous distribution network new energy installed capacity planning method described in this invention, wherein: in the distribution network operation constraint model, the power flow balance constraint is used to ensure that the injected power of each node and the transmitted power of the branch satisfy Kirchhoff's laws at any time; Node voltage constraints are used to limit the voltage amplitude of each node to within the allowable operating range. Branch current constraints are used to prevent line overload. Generator output upper and lower limit constraints are used to ensure that wind power and photovoltaic equipment operate within their technical capabilities; Line power transmission constraints are used to ensure that the active and reactive power of a line does not exceed its thermal stability limit.

[0011] As a preferred embodiment of the mountainous area power grid renewable energy installed capacity planning method described in this invention, the process of constructing the comprehensive objective function includes: Calculate the initial construction investment cost of photovoltaic and wind power, which is related to the installed capacity and unit cost. Calculate the annual operation and maintenance costs, which are directly proportional to the installed capacity. Calculate network loss cost, which is determined based on line active power loss, electricity price, and time weighting under multiple scenarios; The revenue from renewable energy generation is calculated based on the actual amount of renewable energy generated and the on-grid electricity price in each scenario. The above costs and benefits are converted into annual economic indicators using the equivalent annual value method, and the optimization objective is to minimize the total cost.

[0012] As a preferred embodiment of the mountainous power grid renewable energy installed capacity planning method of the present invention, the step of solving the renewable energy installed capacity configuration scheme that optimizes the comprehensive objective function through iterative optimization includes: Set the convergence criteria and maximum number of iterations for the optimization algorithm; Randomly generate an initial planning scheme within the feasible installed capacity range; For each planning scheme, power flow calculations are performed by combining multi-scenario renewable energy output models, load data, and network topology. Determine whether the power flow calculation results satisfy all operational constraints. If they do, calculate the corresponding comprehensive objective function value; otherwise, discard the corresponding scheme. The planning scheme that yields a better overall objective function value is retained, and new schemes are generated iteratively. When the convergence condition or the maximum number of iterations is reached, the current optimal configuration scheme for new energy installed capacity is output.

[0013] Secondly, the present invention provides a new energy installed capacity planning system for mountain power distribution networks, comprising: The data acquisition module is used to collect network structure parameters, local climate parameters, and historical operating data of wind power and photovoltaic power generation output and load at each node in the target mountainous area. The partitioning module is used to perform dimensionality reduction processing on the historical operating data and extract the main spatiotemporal features based on principal component analysis, thereby dividing the distribution network into several new energy areas. The feature value acquisition module is used to extract features from the historical output data of new energy sources for each new energy region using an improved spectral clustering algorithm, and obtain coded feature values ​​that reflect the temporal variation patterns. The processing model construction module is used to cluster and classify the operating scenarios of each new energy region according to the encoded feature values, and construct new energy output models corresponding to normal operating conditions and extreme weather conditions respectively. The constraint model construction module is used to establish a distribution network operation constraint model under the condition of new energy access. The constraint model includes power flow balance constraints, node voltage constraints, branch current constraints, generator output upper and lower limit constraints, and line power transmission constraints. The objective function construction module is used to construct a comprehensive objective function with the goal of full life-cycle economic efficiency. The comprehensive objective function takes into account the investment cost of new energy construction, operation and maintenance cost, network loss cost, and the revenue of new energy power generation. The solution module is used to find the optimal new energy installed capacity configuration scheme that optimizes the comprehensive objective function by iterative optimization, under the premise of satisfying the distribution network operation constraint model.

[0014] Thirdly, the present invention provides an electronic device including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the method described above.

[0015] Fourthly, the present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method described above.

[0016] Compared with existing technologies, the beneficial effects of this invention are that it proposes a method for planning renewable energy installed capacity in mountainous power distribution networks. First, it collects historical operational data on the network structure, climate, and renewable energy and load of the mountainous power distribution network. Then, it divides the nodes into regions based on principal component analysis, and uses an improved spectral clustering algorithm that integrates graph structure information to extract the temporal characteristics of renewable energy output in each region. Next, it constructs a multi-scenario renewable energy output model covering both normal and extreme weather conditions. Based on this, it establishes a power distribution network operation constraint system including power flow, voltage, current, and equipment output constraints, and constructs a comprehensive objective function with full life-cycle economic efficiency as the goal. Finally, iterative optimization is used to find the renewable energy installed capacity configuration scheme that satisfies the constraints and achieves optimal economic efficiency. This invention effectively improves the planning robustness, power supply reliability, and renewable energy absorption capacity of mountainous power distribution networks under extreme conditions. Attached Figure Description

[0017] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 A flowchart of a method for planning the installed capacity of new energy power distribution networks in mountainous areas, provided as an embodiment of the present invention.

[0019] Figure 2 This is a diagram of a power distribution system considering the grid connection of new energy sources in mountainous areas, provided as an embodiment of a method for planning the installed capacity of new energy distribution networks in mountainous areas according to an embodiment of the present invention.

[0020] Figure 3 The following is an example of a method for planning the installed capacity of new energy power distribution networks in mountainous areas, provided as an embodiment of the present invention: the result of dividing the mountainous power distribution network area.

[0021] Figure 4 This is an internal structure diagram of an electronic device for a method of planning the installed capacity of new energy power distribution networks in mountainous areas, provided as an embodiment of the present invention. Detailed Implementation

[0022] To make the above-mentioned objects, features, and advantages of the present invention more apparent and understandable, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the protection scope of the present invention.

[0023] Example 1, referring to Figure 1 This is the first embodiment of the present invention, which provides a method for planning the installed capacity of new energy power distribution networks in mountainous areas, including: This invention provides a method that can effectively solve the problems mentioned above. The following will describe in detail how to implement the method for planning the installed capacity of new energy power distribution network in mountainous areas, using multiple embodiments. Figure 1 A flowchart illustrating a method for planning the installed capacity of new energy power distribution networks in mountainous areas is shown, including: S101, collect network structure parameters, local climate parameters, and historical operating data of wind power and photovoltaic power generation output and load of each node in the target mountainous area power distribution network; It should be noted that current renewable energy planning for mountainous power distribution networks generally relies on simplified typical daily power output curves or single meteorological scenarios for modeling, lacking the ability to systematically integrate multi-source heterogeneous historical operating data under complex terrain. For example, in mountainous environments, due to large altitude differences and significant micro-meteorological characteristics, the photovoltaic output of adjacent nodes may exhibit high asynchrony due to local cloud cover or different slope orientations. If only a unified output model is used or the impact of extreme weather is ignored, it can easily lead to an imbalance in installed capacity configuration, resulting in the risk of equipment curtailment or line overload.

[0024] Furthermore, traditional methods typically do not include the time-dimensional coupling relationship between load fluctuations and renewable energy output in the data collection scope, making it difficult for subsequent optimization models to accurately reflect the actual operating status, thereby weakening the adaptability and robustness of the planning scheme after actual operation.

[0025] Understandably, this solution constructs a high-spatiotemporal-resolution multidimensional input dataset by comprehensively collecting network structure parameters, local climate parameters, and historical operational data of wind power and photovoltaic power generation output and load at each node in the target mountainous distribution network. This provides a solid foundation for subsequent regional division and scenario modeling. The network structure parameters include line impedance, node connection relationships, transformer capacity, and topology hierarchy information. Local climate parameters cover key indicators such as average annual sunshine hours, wind speed distribution, extreme temperature records, precipitation frequency, and snow cover duration. Wind power and photovoltaic power generation output data at each node must be continuously recorded for at least one year at a granularity of no less than 15 minutes, and historical load operation data is collected synchronously to match the time scale of new energy output. These data collectively constitute a complete input system reflecting the operational characteristics of the mountainous distribution network, ensuring that subsequent principal component analysis and spectral clustering algorithms can effectively identify node clusters with similar output patterns and accurately distinguish between typical scenarios under normal operating conditions and extreme weather conditions. This supports optimized capacity decisions that balance economic efficiency and reliability.

[0026] Among them, network structure parameters refer to a set of quantitative indicators describing the physical connection and electrical characteristics of the distribution network, including but not limited to branch resistance and reactance values, node numbers and hierarchical relationships, substation locations and feeder segmentation information; local climate parameters refer to a series of environmental variables directly related to the performance of new energy power generation, provided by meteorological stations or reanalysis data, covering temperature, irradiance, wind speed, humidity and frequency of extreme events; historical operating data of wind power and photovoltaic power generation at each node refers to the time series of active power output recorded by metering devices during actual operation, and historical operating data of load refers to the time series of power demand recorded at the same node during the same period. Both must have time alignment and completeness to ensure the accuracy of subsequent spatiotemporal feature extraction.

[0027] S102, based on principal component analysis, performs dimensionality reduction on historical operating data and extracts the main spatiotemporal features, thereby dividing the distribution network into several new energy areas; In this embodiment of the invention, the steps of dividing the distribution network into several new energy areas based on principal component analysis to perform dimensionality reduction processing on historical operating data and extract the main spatiotemporal features include: Centralized processing of historical output data for photovoltaic or wind power; Calculate the covariance matrix of the processed data, and perform eigenvalue decomposition on the covariance matrix to obtain multiple principal components and their corresponding contribution rates; Based on a preset cumulative contribution rate threshold, several principal components are selected to form a low-dimensional feature space. The original node data is mapped to a low-dimensional feature space to obtain the principal component eigenvalues ​​of each node; The nodes are clustered based on the similarity of the principal component eigenvalues ​​to form several photovoltaic or wind power regions.

[0028] It should be noted that in mountainous power distribution networks, the output curves of new energy nodes in different geographical locations are highly heterogeneous due to terrain shielding, orientation differences, and local meteorological disturbances. Directly using the original high-dimensional time-series data for regional division is not only computationally complex but also susceptible to noise interference, leading to distorted clustering results. Therefore, principal component analysis is needed to perform structured dimensionality reduction on the original output data, extracting the spatiotemporal common features of the dominant output modes, thereby achieving new energy regional division with clear physical meaning and consistent electrical characteristics.

[0029] Specifically, the historical output data of all photovoltaic or wind power nodes in the target mountain area are first centralized to make the mean of the variables in each time dimension zero, eliminating the influence of dimensional and benchmark offset. Then, the covariance matrix is ​​calculated based on the centralized data matrix to reflect the linear correlation between output fluctuations at each time node. Next, eigenvalue decomposition is performed on the covariance matrix to obtain a set of orthogonal eigenvectors and their corresponding eigenvalues. Each set of eigenvectors represents a principal component direction, and the magnitude of the eigenvalue represents the proportion of information contained in the principal component, i.e., the contribution rate.

[0030] In some embodiments, a cumulative contribution rate threshold of 90% is first set to ensure that the retained principal components can cover most of the effective variation information in the original data. Then, the contribution rates corresponding to the feature values ​​are accumulated from large to small until the threshold condition is met, and the first k principal components are selected to form a low-dimensional feature space. Then, using the feature vectors corresponding to these k principal components, the original output sequence of each node is projected onto this low-dimensional space to generate the principal component feature value vector of each node. For example, in a distribution network in a mountainous area, there are 30 photovoltaic nodes. After principal component analysis, the first 3 principal components are retained, and each node is represented as a three-dimensional feature vector. Then, using these three-dimensional feature vectors as input, the K-means clustering algorithm is used to calculate the Euclidean distance between nodes, and nodes with similar features are grouped into the same category. For example, the principal component feature values ​​of nodes 7, 22, and 31 are closely clustered in the three-dimensional space and are assigned to photovoltaic region 1, while nodes 15, 18, and 24 form another cluster and are assigned to photovoltaic region 2. Finally, the same process is repeated for wind power nodes to form independent wind power region division results, thereby completing the structured grouping of new energy regions in the entire distribution network.

[0031] S103, for each new energy region, an improved spectral clustering algorithm is used to extract features from the historical output data of new energy sources to obtain coded feature values ​​that reflect the temporal variation pattern; In this embodiment of the invention, the improved spectral clustering algorithm includes constructing an autoencoder structure, which consists of an encoder and a decoder, both of which are implemented using a long short-term memory neural network. Historical power output data of new energy sources and a similarity graph structure built based on Gaussian kernel function are used as inputs, and an autoencoder learns the inherent low-dimensional representation of the data. During training, both reconstruction error and graph structure preservation loss are minimized simultaneously. The output is the encoded feature value generated by the encoder after training and convergence, which is used for subsequent scene clustering.

[0032] It should be noted that traditional spectral clustering methods, when processing time-series data of renewable energy output in mountainous areas, typically rely solely on static similarity matrices for graph embedding. This makes it difficult to capture the dynamic evolution patterns and long-term dependencies within the time series, resulting in clustering results that fail to accurately reflect the essential differences in output patterns under various meteorological conditions. Especially in mountainous environments where extreme weather events are sparse but have a significant impact, neglecting the joint modeling of temporal structure and graph topology information will make it difficult to distinguish between normal fluctuations and abnormal interruptions, thereby affecting the reliability of subsequent multi-scenario output models.

[0033] Specifically, an improved spectral clustering framework integrating temporal modeling capabilities and graph structure perception mechanisms is constructed. This framework uses an autoencoder as its core structure, with both the encoder and decoder implemented using unidirectional long short-term memory neural networks to handle new energy output sequences with strong temporal correlations. Simultaneously, a node similarity graph structure based on a Gaussian kernel function is introduced to explicitly inject spatial correlation into the learning process. During the training phase, two objectives are optimized simultaneously: first, to make the decoder output as close as possible to the original input sequence, i.e., minimizing the reconstruction error; and second, to constrain the low-dimensional representation of the encoder output to maintain local neighborhood relationships in the graph structure, i.e., minimizing the graph structure preservation loss. Finally, a joint loss function drives model convergence, extracting discriminative encoded feature values.

[0034] In some embodiments, the historical power output data of all nodes in a photovoltaic area are first expanded by time steps to form a multivariate time series matrix. Then, the Gaussian kernel function is used to calculate the similarity between the power output curves of any two nodes to construct a symmetric similarity matrix. For example, nodes 7 and 22 are both located on the south slope and have no shading, so their annual power output curves are highly similar, and their corresponding similarity values ​​are close to 1. Next, the similarity matrix is ​​converted into a degree matrix and a normalized Laplacian matrix as graph structure constraints. Then, the original power output sequence and the Laplacian matrix are input into an autoencoder, where the encoder updates the hidden state step by step through long short-term memory units, gradually compressing the input to a low-dimensional latent space. For example, if the latent dimension is set to 8, each node is mapped to an encoded feature value vector of length 8. Then, the mean squared reconstruction error and the graph Laplacian regularization term are weighted and combined in the loss function. For example, the graph loss weight is set to 0.3 to ensure a balance between time series fidelity and structural consistency. Finally, when the total loss decreases by less than a preset threshold for 10 consecutive iterations, the model is considered to have converged, and the encoded feature values ​​generated by the encoder are output for subsequent K-means clustering to classify typical operating scenarios.

[0035] S104. Based on the coding feature values, the operation scenarios of each new energy region are clustered and classified to construct new energy output models corresponding to normal operation conditions and extreme weather conditions, respectively. In this embodiment of the invention, the steps of constructing new energy output models corresponding to normal operating conditions and extreme weather conditions respectively include: The number of cluster categories is set based on the encoded feature values, and the cluster centroids are initialized. Each data point is assigned to the category of the nearest centroid, and the centroid position of each category is updated; Repeat the allocation and update operations until the centroid converges to obtain several typical operating scenarios; By combining meteorological information and historical power output characteristics, typical operating scenarios are divided into normal climate scenarios and extreme weather scenarios, and the occurrence probability of each type of scenario is statistically analyzed. Based on the historical power output curves for each type of scenario, a corresponding probability model for new energy power output is established.

[0036] It should be noted that in mountainous power distribution network planning, using only a single or averaged renewable energy output model will fail to reflect the drastic fluctuations and structural interruptions in output under complex meteorological conditions. Especially under rare but high-impact events such as extreme high temperatures, strong winds, or snow accumulation, traditional models tend to significantly overestimate the generating capacity, leading to redundant installed capacity or insufficient system safety margins. Therefore, it is essential to identify representative typical operating scenarios from historical operational data and clearly distinguish between normal and extreme meteorological conditions to provide multi-scenario support for subsequent robust optimization.

[0037] Specifically, the K-means clustering algorithm is used to classify the encoded feature values ​​of the aforementioned improved spectral clustering output into scenarios. This process uses Euclidean distance as a metric and achieves automatic grouping of data points through iterative optimization. The number of clusters is predetermined based on the elbow rule or silhouette coefficient analysis to ensure that the main output modes are covered while avoiding overfitting noise disturbances. In each iteration, all data points are assigned to the cluster corresponding to the nearest centroid, and then the mean of all points in each cluster is recalculated as the new centroid until the change in the centroid position is less than the preset convergence threshold.

[0038] In some embodiments, cluster analysis is first performed on the coded feature values ​​of the historical power output of a photovoltaic area over 2000 days, and the number of clusters K is determined to be 5 based on the peak value of the silhouette coefficient. Then, 5 three-dimensional centroid vectors are randomly initialized, and the K-means iteration process is started. For example, the coded feature value of day 127 is calculated to be closest to centroid number 3, so it is assigned to the 3rd class. Then, the centroid of the 3rd class is updated to the average vector of the feature values ​​of all days in that class. The above allocation and update operation is repeated until the centroid displacement is less than 0.001 for 5 consecutive rounds. Finally, 5 typical operating scenarios are obtained, and each scenario corresponds to a set of power outputs with similar temporal patterns. The system generates a set of curves; then, it combines these curves with concurrent meteorological records to label different scenarios. For example, scenarios 1, 2, and 3 correspond to normal weather conditions such as sunny, cloudy, and rainy days; scenario 4 corresponds to continuous snow cover causing almost no photovoltaic output; and scenario 5 corresponds to extreme high temperatures in summer causing a decrease in module efficiency. Finally, the system calculates the frequency of each scenario in historical data. For example, scenario 1 occurred 788 days and scenario 4 occurred 38 days. Based on this, the probabilities of occurrence are calculated to be 0.394 and 0.019, respectively. Based on the daily power output curves of all scenarios in each category, a Beta distribution or piecewise linear probability density function is fitted to construct the corresponding new energy power output probability model.

[0039] S105. Establish a distribution network operation constraint model under the condition of new energy access. The constraint model includes power flow balance constraints, node voltage constraints, branch current constraints, generator output upper and lower limit constraints, and line power transmission constraints. In this embodiment of the invention, in the power distribution network operation constraint model, the power flow balance constraint is used to ensure that the injected power of each node and the transmitted power of the branch satisfy Kirchhoff's laws at any time. Node voltage constraints are used to limit the voltage amplitude of each node to within the allowable operating range. Branch current constraints are used to prevent line overload. Generator output upper and lower limit constraints are used to ensure that wind power and photovoltaic equipment operate within their technical capabilities; Line power transmission constraints are used to ensure that the active and reactive power of a line does not exceed its thermal stability limit.

[0040] It should be noted that if the synergistic effect of multi-dimensional operational constraints is ignored during the planning stage after a high proportion of new energy sources are connected to the distribution network in mountainous areas, serious problems such as voltage exceeding limits, line overload, and even power flow non-convergence will occur in actual operation. For example, if a node injects a large amount of active power during a peak solar power generation period on a sunny midday, without power flow balance constraints, the branch power flow direction cannot be accurately calculated, potentially leading to misjudgments of the line load level. Without node voltage constraints, the node voltage may rise above 1.08 per-unit value, exceeding the distribution network's allowable range of 0.95 to 1.05 per-unit value, threatening user equipment safety. Without branch current constraints, the feeder end is prone to current overruns due to long-distance transmission and high impedance characteristics, accelerating line aging and even causing tripping. Ignoring generator output upper and lower limit constraints might lead to the optimization model allocating 6 MW of installed capacity to a 5 MW solar power plant, resulting in long-term inverter overload operation. Without line power transmission constraints, concentrated wind power output combined with low load during extreme wind conditions could cause the active power of a 35 kV line to exceed its 20 MW thermal stability limit, resulting in conductor overheating and deformation.

[0041] S106. Construct a comprehensive objective function with the goal of full life cycle economic efficiency. The comprehensive objective function takes into account the investment cost of new energy construction, operation and maintenance cost, network loss cost and the revenue of new energy power generation. In this embodiment of the invention, the process of constructing the comprehensive objective function includes: Calculate the initial construction investment cost of photovoltaic and wind power, which is related to the installed capacity and unit cost. Calculate the annual operation and maintenance costs, which are directly proportional to the installed capacity. Calculate network loss cost, which is determined based on line active power loss, electricity price, and time weighting under multiple scenarios; The revenue from renewable energy generation is calculated based on the actual amount of renewable energy generated and the on-grid electricity price in each scenario. The above costs and benefits are converted into annual economic indicators using the equivalent annual value method, and the optimization objective is to minimize the total cost.

[0042] It should be noted that in the planning of new energy distribution networks in mountainous areas, if the sole objective is to maximize installed capacity or minimize initial investment, the operational economics and system losses throughout the entire life cycle may be overlooked. This can lead to a situation where the initial investment is low but the long-term operation and maintenance and network loss costs are high, or the power generation is high but the revenue is lost due to forced curtailment caused by voltage exceeding limits. Therefore, it is necessary to construct a comprehensive objective function that covers investment, operation and maintenance, losses, and revenue to achieve a unified assessment of economic efficiency and technical feasibility.

[0043] In some embodiments, the initial construction investment costs of photovoltaic (PV) and wind power are first calculated separately. This cost is equal to the sum of the planned installed capacity of each node multiplied by the corresponding unit cost. For example, if node 7 plans to install 1.2531 MW of PV with a unit cost of 3.5 million yuan per MW, its investment cost is 4.38585 million yuan. If node 36 plans to install 0.5319 MW of wind power with a unit cost of 6 million yuan per MW, its investment cost is 3.1914 million yuan. Next, the annual operation and maintenance costs are calculated. This cost is linearly calculated based on the installed capacity. For example, if the annual maintenance unit price for PV is 12,000 yuan per MW and for wind power it is 18,000 yuan per MW, then the annual maintenance costs for the two nodes are 15,037 million yuan and 9,574 million yuan, respectively. Then, based on the multi-scenario power flow calculation results, the active power loss of the lines in each time period under each typical operating scenario is statistically analyzed. Combined with time-of-use pricing and the probability of scenario occurrence, a weighted sum is obtained to obtain the average annual network loss cost. For example, in a PV field... In Scenario 1 (sunny day), the average daily grid loss is 120 kWh, the electricity price is 0.45 yuan per kWh, and the probability of this scenario is 0.394. Therefore, its annual contribution loss cost is approximately 120 × 0.45 × 365 × 0.394 ≈ 7758 yuan. Then, the revenue from renewable energy generation is calculated. This revenue equals the actual power generation of each node in each scenario multiplied by the grid-connected electricity price, then weighted by probability and summed. For example, node 7 generates an average of 5000 kWh per day in Scenario 1, with a grid-connected electricity price of 0.38 yuan per kWh. If the annual revenue contribution is 5000×0.38×365×0.394≈274,319 yuan, then all cost items are subtracted from the revenue items and converted into annual net economic indicators using the equivalent annual value method. The initial investment is calculated with a discount rate of 5%, a photovoltaic lifespan of 25 years, and a wind power lifespan of 20 years, respectively, and converted into equivalent annual investment costs. Finally, the optimization objective is to minimize the total cost in that year to ensure that the scheme is economically optimal throughout its entire life cycle.

[0044] S107, under the premise of satisfying the distribution network operation constraint model, solves the new energy installed capacity configuration scheme that makes the comprehensive objective function optimal through iterative optimization.

[0045] In this embodiment of the invention, the step of finding the optimal new energy installed capacity configuration scheme that optimizes the comprehensive objective function through iterative optimization includes: Set the convergence criteria and maximum number of iterations for the optimization algorithm; Randomly generate an initial planning scheme within the feasible installed capacity range; For each planning scheme, power flow calculations are performed by combining multi-scenario renewable energy output models, load data, and network topology. Determine whether the power flow calculation results satisfy all operational constraints. If they do, calculate the corresponding comprehensive objective function value; otherwise, discard the corresponding scheme. The planning scheme that yields a better overall objective function value is retained, and new schemes are generated iteratively. When the convergence condition or the maximum number of iterations is reached, the current optimal configuration scheme for new energy installed capacity is output.

[0046] This invention proposes a method for planning renewable energy installed capacity in mountainous power distribution networks. First, it collects historical operational data on the network structure, climate, and renewable energy and load of the mountainous power distribution network. Then, it uses principal component analysis to divide nodes into regions, and an improved spectral clustering algorithm incorporating graph structure information is used to extract the temporal characteristics of renewable energy output in each region. Next, it constructs a multi-scenario renewable energy output model covering both normal and extreme weather conditions. Based on this, it establishes a power distribution network operation constraint system including power flow, voltage, current, and equipment output constraints, and constructs a comprehensive objective function with full life-cycle economic efficiency as the goal. Finally, iterative optimization is used to find the renewable energy installed capacity configuration scheme that satisfies the constraints and achieves optimal economic efficiency. This invention effectively improves the planning robustness, power supply reliability, and renewable energy absorption capacity of mountainous power distribution networks under extreme conditions.

[0047] Example 2, refer to Figures 2-3 Based on the above embodiments, a specific implementation method for planning the installed capacity of new energy power distribution networks in mountainous areas can be designed as follows: like Figure 2 As shown, the network structure parameters, climate parameters, and historical annual operation data of wind power, photovoltaic power and load at each node of the target mountain power distribution network were collected. Based on the actual situation of the target, it is planned to install photovoltaic power at 3 nodes and wind power at 2 nodes.

[0048] Furthermore, based on principal component analysis, new energy regions were divided, and the results are as follows: Figure 3 As shown in the figure, a distribution network is divided into two photovoltaic regions and two wind power regions according to the annual characteristics of new energy sources. The power output model of new energy sources in the same region can be considered to be consistent.

[0049] Furthermore, the specific steps for dividing new energy regions based on principal component analysis are as follows: The collected photovoltaic data is centrally processed to make the mean of each individual variable zero, as shown in the following formula: In the formula, The original value of the j-th data in node i. Let be the mean of the j-th variable. This is the processed value of the j-th data in node i.

[0050] Furthermore, the covariance matrix among the target variables is calculated using the following formula: In the formula, Let be the covariance matrix we are looking for. The total number of nodes. For the centralized data matrix, This is the transpose of the data matrix after centralization.

[0051] Furthermore, eigenvalue decomposition is performed on the covariance matrix to find the principal components and their importance. The calculation formula is as follows: In the formula, The eigenvector represents the direction of the corresponding principal component. The eigenvalue represents the importance of the principal component corresponding to the eigenvector.

[0052] Furthermore, the eigenvalues ​​are ranked according to their importance, and principal components are selected based on a preset contribution rate threshold. The cumulative contribution rate is calculated using the following formula: In the formula, This represents the cumulative contribution rate of the first k principal components after sorting. The total number of data points for a single node. The feature corresponding to the j-th principal component.

[0053] Furthermore, based on the obtained principal components, the data is mapped to a new low-dimensional space using the corresponding eigenvectors, as calculated by the following formula: In the formula, This is the principal component matrix obtained after mapping the original data, where each row represents the corresponding new eigenvalues. This is the eigenvector corresponding to the k-th principal component.

[0054] Furthermore, based on the obtained principal component matrix, the nodes are divided into regions, meaning nodes with similar eigenvalues ​​belong to the same region. The formula for calculating the node eigenvalue is as follows: In the formula, Let m be the node characteristic value of node m. The elements in row m and column i of the principal component matrix.

[0055] Furthermore, wind power data is used to replace photovoltaic data, and the aforementioned steps are repeated to complete the division of distribution network areas.

[0056] Furthermore, feature values ​​of new energy data in each region are extracted based on an improved spectral clustering algorithm. The specific steps based on the improved spectral clustering algorithm are as follows: A similarity matrix is ​​constructed based on the collected dataset, using the following formula: In the formula, This is a similarity matrix constructed using the Gaussian kernel function. For elements of the similarity matrix, , For the data points in the dataset, It is a scale parameter that controls the width of the neighborhood.

[0057] Furthermore, the degree matrix is ​​constructed, and the calculation formula is as follows: In the formula, the diagonal matrix It is a degree matrix.

[0058] Furthermore, the Laplace matrix is ​​constructed as follows: In the formula, For a symmetric normalized Laplace matrix, It is the identity matrix. It is the inverse of the square root of the degree matrix.

[0059] Furthermore, an autoencoder model consisting of an encoder and a decoder is constructed, where both the encoder and decoder are unidirectional long short-term memory neural networks, and the hidden layer parameters of the two neural networks are initialized.

[0060] Furthermore, using the collected data and the Laplacian matrix as input, the encoded feature values ​​are calculated based on the encoder network parameters, using the following formula: In the formula, t represents the time corresponding to the current data. This is the input value at the current moment. The value of the forget gate. , , For the corresponding network parameters, Candidate cell state, , , For the corresponding network parameters, For the value of the input gate, , , For the corresponding network parameters, For activation function, This represents the state value of the memory unit at the current moment. For Hadama accumulation, The value of the output gate. , , For the corresponding network parameters, The hyperbolic tangent activation function is used. The feature values ​​output by the encoder. , This is the calculated value corresponding to the previous moment.

[0061] Furthermore, the encoded feature values ​​output by the encoder are used as input for the decoder calculation. The calculation process is the same as the step of using the acquired data and the Laplacian matrix as input to calculate the encoded feature values ​​based on the encoder network parameters. The parameters are synchronously adopted from the decoder network parameters.

[0062] Furthermore, the autoencoder loss function is calculated using the following formula: In the formula, For the total loss function, To reconstruct the loss, To balance the losses, Here, b is a hyperparameter representing importance, and b is the number of input parameters. This is the encoder's raw input. For decoder output, Let be the Laplace matrix, Tr denote the trace of the matrix, and F denote the desired dimensionality-reduced feature representation.

[0063] Furthermore, if the total loss function value reaches the preset value, the encoded feature value is output; otherwise, the neural network parameters are updated, and the aforementioned steps are repeated.

[0064] Furthermore, classification is performed based on the characteristic values ​​of new energy data, and new energy output models are constructed for each region under normal and extreme conditions. The specific steps for classification based on the characteristic values ​​of new energy data are as follows: Furthermore, clustering is performed based on the obtained encoded feature values, and the required number of categories K is selected according to the actual situation.

[0065] Furthermore, K centroids are randomly initialized.

[0066] Furthermore, each data point is assigned to the cluster of its nearest centroid, as shown in the following formula: In the formula, Represents the index of the assigned cluster, and argmin indicates minimizing the parameter. This represents the i-th data point. Let k be the k-th centroid.

[0067] Furthermore, the centroid of each cluster is recalculated using the following formula: In the formula, For the new center of mass, The number of data points The allocated subset of data.

[0068] Furthermore, determine whether the centroids have converged. If they have not converged, repeat the aforementioned steps. If they have converged, output the clustering results.

[0069] Furthermore, the scenarios and probabilities of wind and solar power are shown in Tables 1 and 2. Solar power under normal conditions is categorized into three types: cloudy, sunny, and rainy, each corresponding to a different power output curve. Extreme weather conditions are further divided into snowy and high-temperature weather. During snowy weather, solar power output is minimal, while under extreme high-temperature conditions, the power generation efficiency decreases due to the high temperature affecting the modules. Wind power under normal conditions is also categorized into three types, each corresponding to a different power output curve. Extreme weather conditions are divided into two types: one is reduced power generation efficiency of wind turbines under extreme temperature conditions, and the other is extreme wind conditions, which can lead to shutdown and zero output.

[0070] Table 1 Probabilities of various photovoltaic scenarios

[0071] Table 2 Probabilities of Wind Power Scenarios

[0072] Furthermore, a constraint model for the target mountainous distribution network under the condition of new energy grid connection is constructed, including distribution network power flow constraints, as shown in the following formula: In the formula, , For node voltage, For nodes , Line impedance between, For nodes , Line current between For nodes , Line power between This represents node power.

[0073] Furthermore, it also includes generator power constraints, node voltage and power constraints, and branch current and power constraints, as shown in the following formulas: In the formula , These represent the current active and reactive power outputs of the generator. , , , These represent the upper and lower limits of the generator's active and reactive power, respectively. The current node voltage. This represents the current branch current. , These are the upper and lower limits of the node voltage. , These are the upper and lower limits of the line current. , Let i be the active and reactive loads. , Let the active and reactive power of the generator at node i be denoted as . , For the current active and reactive power of the line, , , , These are the upper and lower limits for the active and reactive power of the line.

[0074] Furthermore, a comprehensive objective function considering economic benefits is proposed. The smaller the value of this objective function, the better the economic benefits, and the better the corresponding planning scheme. The calculation formula is as follows: In the formula, The annual economic objective function is... For photovoltaic construction investment costs, For photovoltaic maintenance costs, For wind power construction investment costs, For wind power maintenance costs, For line loss costs, The specific calculation methods for various indicators of photovoltaic power generation revenue are as follows: In the formula, For photovoltaic and other annual value coefficients, For the discount rate, For the lifespan of photovoltaic systems, for Node photovoltaic capacity, For the number of photovoltaic nodes, This refers to the unit investment cost for photovoltaic construction.

[0075] In the formula, for Node wind power capacity, For the number of wind power nodes, The unit cost of wind power construction investment.

[0076] In the formula, This represents the annual maintenance cost per unit of photovoltaic power.

[0077] In the formula, This refers to the annual maintenance cost per unit of wind power.

[0078] In the formula, For the number of scenes, For intraday time points, Weighting of operating costs This is an operating cost correction factor. for The price at which electricity is purchased from the upper-level power grid at all times. For distribution network The active power loss of the line at any given time.

[0079] In the formula, for The price of renewable energy electricity that is always connected to the grid. for time Scene Total renewable energy generation at each node.

[0080] Furthermore, by combining the constraint model and the objective function, the optimal installed capacity of new energy is solved. The specific steps for solving the optimal installed capacity of new energy are as follows: ① Determine the convergence conditions and the upper limit of the number of iterations based on the actual situation.

[0081] ② Plan the grid-connected capacity of new energy sources at each node using random sampling methods.

[0082] ③ Perform power flow calculations based on the grid-connected capacity of new energy sources, load data, and network topology data of each node.

[0083] ④ Verify the power flow calculation results based on the constraint model. If the constraints are met, proceed to step ⑤; otherwise, skip to step ②.

[0084] ⑤ Calculate the comprehensive objective function value. If it is greater than the previous calculation result, record the corresponding planning scheme.

[0085] ⑥ Check whether the solution has converged or whether the upper limit of the number of iterations has been reached. If so, end the solution and output the last recorded planning scheme. Otherwise, repeat steps ② to ③.

[0086] Furthermore, the planning scheme is shown in the table below: Table 3 New Energy Installation Planning Scheme

[0087] In summary, this invention proposes a method for planning renewable energy installed capacity in mountainous power distribution networks considering extreme conditions. It employs multi-stage classification for both potential extreme and general scenarios, and balances the economic efficiency and reliability of renewable energy planning in mountainous power distribution networks, thus possessing significant practical value. The above are merely preferred embodiments of this invention and are not intended to limit the scope of protection of this invention. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of this invention are included within the scope of protection of this invention.

[0088] Example 3, referring to Figure 4 This embodiment also provides a new energy installed capacity planning system for mountain power distribution networks, including: The data acquisition module is used to collect network structure parameters, local climate parameters, and historical operating data of wind power and photovoltaic power generation output and load at each node in the target mountainous area. The partitioning module is used to perform dimensionality reduction processing on historical operating data and extract the main spatiotemporal features based on principal component analysis, thereby dividing the power distribution network into several new energy areas. The feature value acquisition module is used to extract features from the historical output data of new energy sources for each new energy region using an improved spectral clustering algorithm, and obtain coded feature values ​​that reflect the temporal variation patterns. The processing model building module is used to cluster and classify the operating scenarios of each new energy region based on the encoded feature values, and build new energy output models corresponding to normal operating conditions and extreme weather conditions respectively. The constraint model construction module is used to establish a distribution network operation constraint model under the condition of new energy access. The constraint model includes power flow balance constraints, node voltage constraints, branch current constraints, generator output upper and lower limit constraints, and line power transmission constraints. The objective function construction module is used to construct a comprehensive objective function with the goal of full life cycle economic efficiency. The comprehensive objective function takes into account the investment cost of new energy construction, operation and maintenance cost, network loss cost, and the revenue of new energy power generation. The solution module is used to find the optimal new energy installed capacity configuration scheme by iterative optimization under the premise of satisfying the distribution network operation constraint model.

[0089] The above-mentioned unit modules can be embedded in the processor of the electronic device in hardware form or independent of it, or they can be stored in the memory of the electronic device in software form, so that the processor can call and execute the corresponding operations of the above modules.

[0090] This embodiment also provides an electronic device, which can be a terminal, and its internal structure diagram can be as follows: Figure 4As shown, the electronic device includes a processor, memory, communication interface, display screen, and input device connected via a system bus. The processor provides computing and control capabilities. The memory includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, carrier networks, NFC (Near Field Communication), or other technologies. When the computer program is executed by the processor, it implements a method for planning the installed capacity of new energy power distribution networks in mountainous areas. The display screen can be an LCD screen or an e-ink screen. The input device can be a touch layer covering the display screen, buttons, a trackball, or a touchpad on the device's casing, or an external keyboard, touchpad, or mouse.

[0091] This embodiment also provides a computer-readable storage medium on which a computer program is stored, and when the computer program is executed by a processor, it performs the following steps: Collect network structure parameters, local climate parameters, and historical operating data of wind power and photovoltaic power output and load at each node of the target mountain power distribution network; Based on principal component analysis, the historical operating data is dimensionality reduced and the main spatiotemporal features are extracted, thereby dividing the distribution network into several new energy areas. For each new energy region, an improved spectral clustering algorithm is used to extract features from the historical output data of new energy sources to obtain coded feature values ​​that reflect the temporal variation patterns. Based on the coding feature values, the operation scenarios of each new energy region are clustered and classified, and new energy output models corresponding to normal operation conditions and extreme weather conditions are constructed respectively. Establish a distribution network operation constraint model under the condition of new energy access. The constraint model includes power flow balance constraints, node voltage constraints, branch current constraints, generator output upper and lower limit constraints, and line power transmission constraints. A comprehensive objective function is constructed with the goal of achieving full life-cycle economic efficiency. The comprehensive objective function takes into account the investment cost of new energy construction, operation and maintenance cost, network loss cost, and the revenue of new energy power generation. Under the premise of satisfying the distribution network operation constraint model, the optimal new energy installed capacity configuration scheme that maximizes the comprehensive objective function is obtained through iterative optimization.

[0092] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

[0093] Although preferred embodiments of the invention have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including both the preferred embodiments and all changes and modifications falling within the scope of the invention.

[0094] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A method for planning the installed capacity of new energy power distribution networks in mountainous areas, characterized in that, include: Collect network structure parameters, local climate parameters, and historical operating data of wind power and photovoltaic power output and load at each node of the target mountain power distribution network; Based on principal component analysis, the historical operating data is dimensionality reduced and the main spatiotemporal features are extracted, thereby dividing the power distribution network into several new energy regions. For each new energy region, an improved spectral clustering algorithm is used to extract features from the historical output data of new energy sources to obtain coded feature values ​​that reflect the temporal variation patterns. Based on the encoded feature values, the operating scenarios of each new energy region are clustered and classified to construct new energy output models corresponding to normal operating conditions and extreme weather conditions, respectively. A distribution network operation constraint model is established under the condition of new energy access. The constraint model includes power flow balance constraints, node voltage constraints, branch current constraints, generator output upper and lower limit constraints, and line power transmission constraints. A comprehensive objective function is constructed with the goal of achieving full life-cycle economic efficiency. This comprehensive objective function takes into account the investment cost of new energy construction, operation and maintenance cost, network loss cost, and revenue from new energy power generation. Under the premise of satisfying the power distribution network operation constraint model, the optimal new energy installed capacity configuration scheme that makes the comprehensive objective function optimal is obtained by iterative optimization.

2. The method for planning the installed capacity of new energy power distribution networks in mountainous areas as described in claim 1, characterized in that, The steps of performing dimensionality reduction processing on the historical operating data and extracting the main spatiotemporal features based on principal component analysis, and dividing the distribution network into several new energy areas accordingly, include: Centralized processing of historical output data for photovoltaic or wind power; Calculate the covariance matrix of the processed data, and perform eigenvalue decomposition on the covariance matrix to obtain multiple principal components and their corresponding contribution rates; Based on a preset cumulative contribution rate threshold, several principal components are selected to form a low-dimensional feature space. The original node data is mapped to the low-dimensional feature space to obtain the principal component feature values ​​of each node; The nodes are clustered based on the similarity of the principal component eigenvalues ​​to form several photovoltaic or wind power regions.

3. The method for planning the installed capacity of new energy power distribution networks in mountainous areas as described in claim 2, characterized in that, The improved spectral clustering algorithm includes constructing an autoencoder structure, which consists of an encoder and a decoder, both of which are implemented using a long short-term memory neural network. The historical output data of new energy sources and the similarity graph structure constructed based on Gaussian kernel function are used as input, and the autoencoder learns the inherent low-dimensional representation of the data. During training, both reconstruction error and graph structure preservation loss are minimized simultaneously. The output is the encoded feature value generated by the encoder after training and convergence, which is used for subsequent scene clustering.

4. The method for planning the installed capacity of new energy power distribution networks in mountainous areas as described in claim 3, characterized in that, The steps for constructing new energy output models corresponding to normal operating conditions and extreme weather conditions include: The number of cluster categories is set based on the encoded feature values, and the cluster centroids are initialized. Each data point is assigned to the category of the nearest centroid, and the centroid position of each category is updated; Repeat the allocation and update operations until the centroid converges to obtain several typical operating scenarios; Based on meteorological information and historical power output characteristics, the typical operating scenarios are divided into normal climate scenarios and extreme weather scenarios, and the occurrence probability of each scenario is statistically analyzed. Based on the historical power output curves for each type of scenario, a corresponding probability model for new energy power output is established.

5. The method for planning the installed capacity of new energy power distribution networks in mountainous areas as described in claim 4, characterized in that, In the power distribution network operation constraint model, the power flow balance constraint is used to ensure that the injected power of each node and the transmitted power of the branch at any time satisfy Kirchhoff's laws. Node voltage constraints are used to limit the voltage amplitude of each node to within the allowable operating range. Branch current constraints are used to prevent line overload. Generator output upper and lower limit constraints are used to ensure that wind power and photovoltaic equipment operate within their technical capabilities; Line power transmission constraints are used to ensure that the active and reactive power of a line does not exceed its thermal stability limit.

6. The method for planning the installed capacity of new energy power distribution networks in mountainous areas as described in claim 5, characterized in that, The process of constructing the comprehensive objective function includes: Calculate the initial construction investment cost of photovoltaic and wind power, which is related to the installed capacity and unit cost; Calculate the annual operation and maintenance costs, which are directly proportional to the installed capacity. Calculate network loss cost, which is determined based on line active power loss, electricity price, and time weighting under multiple scenarios; The revenue from renewable energy generation is calculated based on the actual amount of renewable energy generated and the on-grid electricity price in each scenario. The above costs and benefits are converted into annual economic indicators using the equivalent annual value method, and the optimization objective is to minimize the total cost.

7. The method for planning the installed capacity of new energy power distribution networks in mountainous areas as described in claim 6, characterized in that, The steps for finding the optimal new energy installed capacity configuration scheme that optimizes the comprehensive objective function through iterative optimization include: Set the convergence criteria and maximum number of iterations for the optimization algorithm; Randomly generate an initial planning scheme within the feasible installed capacity range; For each planning scheme, power flow calculations are performed by combining multi-scenario renewable energy output models, load data, and network topology. Determine whether the power flow calculation results satisfy all operational constraints. If they do, calculate the corresponding comprehensive objective function value; otherwise, discard the corresponding scheme. The planning scheme that yields a better overall objective function value is retained, and new schemes are generated iteratively. When the convergence condition or the maximum number of iterations is reached, the current optimal configuration scheme for new energy installed capacity is output.

8. A new energy installed capacity planning system for mountain power distribution networks, using the method described in any one of claims 1 to 7, characterized in that, include: The data acquisition module is used to collect network structure parameters, local climate parameters, and historical operating data of wind power and photovoltaic power generation output and load at each node in the target mountainous area. The partitioning module is used to perform dimensionality reduction processing on the historical operating data and extract the main spatiotemporal features based on principal component analysis, thereby dividing the distribution network into several new energy areas. The feature value acquisition module is used to extract features from the historical output data of new energy sources for each new energy region using an improved spectral clustering algorithm, and obtain coded feature values ​​that reflect the temporal variation patterns. The processing model construction module is used to cluster and classify the operating scenarios of each new energy region according to the encoded feature values, and construct new energy output models corresponding to normal operating conditions and extreme weather conditions respectively. The constraint model construction module is used to establish a distribution network operation constraint model under the condition of new energy access. The constraint model includes power flow balance constraints, node voltage constraints, branch current constraints, generator output upper and lower limit constraints, and line power transmission constraints. The objective function construction module is used to construct a comprehensive objective function with the goal of full life-cycle economic efficiency. The comprehensive objective function takes into account the investment cost of new energy construction, operation and maintenance cost, network loss cost, and the revenue of new energy power generation. The solution module is used to find the optimal new energy installed capacity configuration scheme that optimizes the comprehensive objective function by iterative optimization, under the premise of satisfying the distribution network operation constraint model.

9. An electronic device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method for planning the installed capacity of new energy power distribution network in mountainous areas as described in any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the method for planning the installed capacity of new energy power distribution network in mountainous areas as described in any one of claims 1 to 7.

Images