Electric power resource allocation method, device and system for intra-domain charging piles and medium

Abstract

The invention discloses a power resource allocation method, device and system for intra-domain charging piles and a medium, and relates to the technical field of data processing. The method comprises the following steps: acquiring historical use data corresponding to a plurality of charging piles in a target area, real-time use data corresponding to the plurality of charging piles and power resource supply data of the target area; dividing each charging pile into at least two different charging pile groups according to the historical use data and the real-time use data; on the basis of real-time traffic information in the target area and the charging pile group, constructing a correlation function used for representing the relationship between the traffic condition in the target area and the grouping condition of the charging pile group; and obtaining a target power resource allocation scheme based on the correlation function, the power resource supply data and the real-time use data. The invention aims to minimize the comprehensive cost of power resource allocation of the charging equipment in the target area and ensure the balance of power supply and demand.

Description

Technical Field

[0001] This application relates to the field of data processing technology, and in particular to a method, apparatus, system and medium for allocating power resources for charging piles within a region. Background Technology

[0002] In the current charging service scenario within the domain, charging piles are treated as independent individuals or fixed groups, and power is allocated in a fixed proportion based on preset rules (such as the upper limit of charging pile power and the order of user access), such as "first come, first served" or power is allocated equally according to the rated power of the charging piles.

[0003] However, the current technical means of power resource allocation for charging piles in the region have the following shortcomings: they rely solely on the usage data of charging piles for power resource allocation, without integrating multi-source data such as traffic flow and power supply status, making it difficult to cope with complex scenarios (such as sudden traffic congestion or a surge in charging demand in a certain area due to time-phased traffic changes), and failing to accurately reduce the overall cost of power resource allocation while ensuring the balance of power supply and demand for charging piles. Summary of the Invention

[0004] The main purpose of this application is to provide a method, device, system and medium for power resource allocation of charging piles within a region, aiming to minimize the overall cost of power resource allocation for charging equipment within the target area and ensure the balance between power supply and demand.

[0005] To achieve the above objectives, this application provides a method for allocating power resources for charging piles within a region, the method comprising:

[0006] Acquire historical usage data, real-time usage data, and power supply data of multiple charging piles within the target area;

[0007] Based on the historical usage data and the real-time usage data, each charging pile is divided into at least two different charging pile groups;

[0008] Based on real-time traffic information within the target area and the charging pile group, a correlation function is constructed to characterize the relationship between traffic conditions within the target area and the grouping status of the charging pile group.

[0009] Based on the correlation function, the power supply data, and the real-time usage data, a target power resource allocation plan is obtained.

[0010] Specifically, based on the historical usage data and the real-time usage data, the charging piles are divided into at least two different groups of charging piles, including:

[0011] Based on the historical usage data and the real-time usage data, the charging pile group and the corresponding group label are obtained through a first preset neural network model.

[0012] Specifically, the first preset neural network model includes an input layer, an encoding layer, and a clustering layer;

[0013] The step of obtaining the charging pile group and its corresponding group label based on the historical usage data and the real-time usage data through a first preset neural network model includes:

[0014] Based on the historical usage data and the real-time usage data, an input feature vector is obtained through the input layer;

[0015] Based on the input feature vector, a low-dimensional latent space vector is obtained through the coding layer;

[0016] Based on the low-dimensional latent space vector, the charging pile group and the group label are obtained through the clustering layer.

[0017] Specifically, obtaining the charging pile group and the group label through the clustering layer based on the low-dimensional latent space vector includes:

[0018] The clustering layer transforms the low-dimensional latent space vector into a set of data points, where each data point in the set corresponds one-to-one with each charging pile in the target area.

[0019] A preset number of data points in the data point set are randomly selected as the initial data point centers, wherein the preset number is used to characterize the preset number of the initial charging pile group;

[0020] The first Euclidean distance between each data point and the center of each initial data point is calculated.

[0021] Each data point is assigned to the initial charging pile group corresponding to the smallest Euclidean distance in the first Euclidean distance;

[0022] The mean of the initial charging pile group is calculated using the preset mean calculation formula set in the clustering layer.

[0023] Repeat the above steps until the change threshold of the mean remains within the preset change threshold range, and obtain the center of the target data point based on the current mean.

[0024] The second Euclidean distance between each data point and the center of each target data point is calculated.

[0025] Each data point is assigned to a charging pile group corresponding to the smallest Euclidean distance in the second Euclidean distance, and the group label is determined according to the data characteristics corresponding to the charging pile group.

[0026] Specifically, the real-time traffic information includes real-time traffic flow information within the target area and traffic road information within the target area;

[0027] Based on real-time traffic information within the target area and the charging pile group, a correlation function is constructed to characterize the relationship between traffic conditions within the target area and the grouping status of the charging pile group, including:

[0028] Based on the real-time traffic flow information and the traffic road information, a spatiotemporal feature matrix is ​​constructed;

[0029] The correlation function is obtained through a second preset neural network model based on the group label and the spatiotemporal feature matrix.

[0030] Specifically, the second preset neural network model includes an encoder self-attention layer, a feedforward network layer, a decoder attention layer, and a fully connected output layer;

[0031] The step of obtaining the association function based on the group label and the spatiotemporal feature matrix through a second preset neural network model includes:

[0032] Based on the group labels and the spatiotemporal feature matrix, an attention feature matrix is ​​obtained through the encoder's self-attention layer;

[0033] Based on the attention feature matrix, an enhanced feature matrix is ​​obtained through the feedforward network layer;

[0034] Based on the enhanced feature matrix and the group label, the decoded feature matrix is ​​obtained through the decoder attention layer;

[0035] The correlation function is obtained through the fully connected output layer based on the decoded feature matrix.

[0036] Specifically, obtaining the target power resource allocation plan based on the correlation function, the power resource supply data, and the real-time usage data includes:

[0037] Based on the correlation function, the power resource supply data, and the real-time usage data, construct the objective function and the corresponding constraints for the objective function;

[0038] Solve the objective function to obtain the objective decision variables, wherein the objective decision variables include the target real-time power allocation of each charging pile in the target area and the power allocation of each charging pile group;

[0039] The target decision variable is determined as the target power resource allocation scheme.

[0040] To achieve the above objectives, this application also provides a power resource allocation device for charging piles within a region, the device comprising:

[0041] The first unit is used to acquire historical usage data, real-time usage data, and power supply data of multiple charging piles within the target area;

[0042] The second unit is used to divide each charging pile into at least two different charging pile groups based on the historical usage data and the real-time usage data.

[0043] The third unit is used to construct a correlation function based on real-time traffic information within the target area and the charging pile group to characterize the relationship between traffic conditions within the target area and the grouping status of the charging pile group.

[0044] The fourth unit is used to obtain the target power resource allocation plan based on the correlation function, the power resource supply data, and the real-time usage data.

[0045] To achieve the above objectives, this application also provides a system comprising a memory storing a plurality of instructions; the processor loading instructions from the memory to execute the steps of any of the methods provided in this application.

[0046] To achieve the above objectives, this application also provides a medium storing a plurality of instructions adapted for loading by a processor to execute the steps in any of the methods provided in this application.

[0047] This application provides a method, apparatus, system, and medium for power resource allocation of charging piles within a target area. First, it acquires historical usage data, real-time usage data, and power supply data for multiple charging piles within a target area. Then, based on the historical and real-time usage data, each charging pile is divided into at least two different charging pile groups. Next, based on real-time traffic information within the target area and the charging pile groups, a correlation function is constructed to characterize the relationship between traffic conditions and charging pile group grouping within the target area. Finally, based on the correlation function, the power supply data, and the real-time usage data, a target power resource allocation scheme is obtained, thereby minimizing the overall cost of power resource allocation for charging equipment within the target area and ensuring a balance between power supply and demand. Attached Figure Description

[0048] Figure 1A flowchart illustrating the method provided in the embodiments of this application;

[0049] Figure 2 This is a schematic diagram of the application process of the first preset neural network model provided in the embodiments of this application;

[0050] Figure 3 This is a schematic diagram of the application process of the second preset neural network model provided in the embodiments of this application;

[0051] Figure 4 A schematic diagram of the power resource allocation device for charging piles within the domain provided in this application embodiment;

[0052] Figure 5 This is a schematic diagram of the system provided in an embodiment of this application. Detailed Implementation

[0053] 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.

[0054] The current technical means of allocating power resources for charging piles within the region have the following shortcomings: they rely solely on the usage data of charging piles for power resource allocation, without integrating multi-source data such as traffic flow and power supply status, making it difficult to cope with complex scenarios (such as sudden traffic congestion or a surge in charging demand in a certain area due to time-phased traffic changes), and failing to accurately reduce the overall cost of power resource allocation while ensuring the balance of power supply and demand for charging piles.

[0055] Therefore, this application provides a method, device, system, and medium for allocating power resources for charging piles within a region to solve practical technical problems.

[0056] In some embodiments, the device may be integrated into an electronic system, such as a system, server, or similar system.

[0057] In some embodiments, the server may also be implemented as a system.

[0058] The server can be a standalone physical server, a server cluster or distributed system consisting of multiple physical servers, or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, CDN (Content Delivery Network), and big data and artificial intelligence platforms.

[0059] The system can be a smartphone, tablet, laptop, desktop computer, smart speaker, smartwatch, etc., but is not limited to these. The system and server can be connected directly or indirectly via wired or wireless communication, which is not limited herein.

[0060] The following sections provide detailed descriptions of each example. It should be noted that the sequence numbers of the following embodiments are not intended to limit the preferred order of the embodiments.

[0061] This application provides a method for allocating power resources for charging piles within a region. The method can minimize the overall cost of allocating power resources for charging equipment within the target area and ensure the balance between power supply and demand.

[0062] To facilitate understanding of the technical solutions described in this application by those skilled in the art, an application scenario will be assumed below, such as: the target area is a new urban area containing 50 charging piles, surrounded by 3 transportation hubs, 2 commercial areas, and 5 residential areas. The total power supply capacity of the target area is 2000kW, and the power company provides time-of-use pricing.

[0063] like Figure 1 The specific process of the method can be as follows:

[0064] S110. Obtain historical usage data, real-time usage data and power supply data of multiple charging piles in the target area.

[0065] In some embodiments, the historical usage data may be the charging pile usage records for the past 6 months, including data such as the number of daily charging times, peak hours (e.g., 8 am and 6 pm), and average charging time (45 minutes in residential areas and 20 minutes in commercial areas).

[0066] In some embodiments, the real-time usage data may be the current usage rate of each charging pile obtained through an IoT interface (e.g., charging pile A currently has an 85% usage rate and 3 vehicles in the queue).

[0067] In some embodiments, the power supply data for the target area can be real-time data connected to the power grid, obtaining the current total available power (e.g., 1800kW) and the time-of-use electricity price (1.2 yuan / kWh during the day and 0.5 yuan / kWh at night).

[0068] In some embodiments, the historical usage data, the real-time usage data, and the power resource supply data may be preprocessed as shown below:

[0069] Missing value handling: Use time series interpolation to fill missing values ​​in historical usage data;

[0070] Outlier filtering: Identifies and corrects abnormal charging records using the IsolationForest algorithm;

[0071] Feature standardization: Z-score standardization is performed on numerical feature data.

[0072] Specifically, based on the hypothetical scenarios described above, we can analyze the historical time series data (taking "daily charging frequency" as an example) in the historical usage data. We find that missing values ​​are mostly continuous (e.g., a charging station has no data for 3 consecutive days due to a malfunction) or random (e.g., data is missing on a single day). Then, we use the ARIMA model for interpolation, leveraging the time series correlation of the historical data to predict the missing values. For example, if a charging station has 25, 28, 32, and 30 charging times on days t-2, t-1, t+1, and t+2 respectively, the ARIMA(1,1,1) model predicts 29 missing values ​​for day t. Alternatively, linear interpolation can be used, filling in the missing values ​​with the mean of adjacent time points. For example, if a charging station has missing data on day 5, and the charging times on days 4 and 6 are 22 and 24 respectively, the interpolation result is 23.

[0073] Specifically, for the historical usage data, the IsolationForest algorithm can also be used to construct a binary tree to segment the data and calculate the "isolation score" of each sample (the easier a sample is to be isolated, the more likely it is to be an outlier). Samples with isolation scores higher than the threshold (such as the top 5%) are judged as outliers (such as a charging pile charging 100 times on a certain day, far exceeding the mean of 50 times and the standard deviation of 15 times), and corrected by the median (48 times) of this feature or the mean of the nearby normal samples.

[0074] Specifically, the real-time usage data may include data such as "current utilization rate," "number of vehicles in queue," and "remaining battery power" collected in real time for each charging pile. Due to network fluctuations, instantaneous outliers may occur (e.g., the utilization rate displays as 120%), and some newly connected charging piles may have initial data gaps. For missing data from newly connected charging piles, the average real-time data of adjacent charging piles in the same group is used to fill in the gaps. For example, if a charging pile belongs to the "commercial-intensive" group, and the current utilization rates of its three neighboring charging piles in the same group are 75%, 80%, and 70%, the average utilization rate is filled to 75%. For features with time series characteristics (such as "number of vehicles in queue"), Kalman filtering is used to predict and correct missing values ​​while smoothing out noise. For example, if the real-time number of vehicles in queue at a charging pile is 3 at t=1, and data is missing at t=2, Kalman filtering is used to predict the number of vehicles in queue at t=2 to be 4 (considering the growth trend of the previous moment). For features such as "current usage rate," a Z-score absolute value >3 is defined as an anomaly (e.g., if the usage rate is 120%, Z-score = (1.2 - 0.8) / 0.1 = 4 > 3), indicating a sensor malfunction. Combined with the charging pile's rated power constraint, if a charging pile's rated power is 60kW but the real-time displayed power usage is 70kW, exceeding the rated value, it is considered an anomaly and corrected to 60kW. For numerical features such as "current usage rate" and "number of vehicles in queue," Z-score standardization is performed based on the mean and standard deviation of the real-time data to ensure consistency with historical data feature scales.

[0075] Specifically, the power resource supply data may include data such as "real-time available power," "time-of-use price," and "renewable energy ratio" of the power grid within the target area. Time-of-use prices may be incomplete due to policy adjustments, and real-time available power may experience brief fluctuations due to grid volatility. If the price for a certain period is missing (e.g., the price for "18:00-24:00" is not updated), interpolation is performed based on historical prices for the same period (e.g., the price for the same period last week was 0.8 yuan / kWh) and policy documents (e.g., the latest price adjustment notice). If the policy has been adjusted, the new policy will be used (e.g., adjusted to 0.9 yuan / kWh). Alternatively, a sliding window mean interpolation (e.g., the average of the previous 5 minutes) can be used. For example, if available power is missing at a certain moment, the average of the previous 5 minutes, 1750kW, is used to fill the gap. A 5-minute sliding window is set for "real-time available power." If the current value deviates from the window mean by more than 20% (e.g., the average is 1800kW, the current value is 2200kW), it is considered an abnormal grid fluctuation, and the window median of 1820kW is used for correction. At the same time, ensure that the available power does not exceed the total power supply capacity in the target area (e.g., 2000kW). If the abnormal value exceeds this limit, correct it to 2000kW.

[0076] S120. Based on the historical usage data and the real-time usage data, divide each charging pile into at least two different charging pile groups.

[0077] In some embodiments, dividing each charging pile into at least two different charging pile groups based on the historical usage data and the real-time usage data includes the following specific implementation process:

[0078] Based on the historical usage data and the real-time usage data, the charging pile group and the corresponding group label are obtained through a first preset neural network model.

[0079] Specifically, such as Figure 2 The first preset neural network model includes an input layer, an encoding layer, and a clustering layer.

[0080] In some embodiments, the step of obtaining the charging pile group and the corresponding group label of the charging pile group through a first preset neural network model based on the historical usage data and the real-time usage data includes the following steps S121 to S123:

[0081] S121. Based on the historical usage data and the real-time usage data, the input feature vector is obtained through the input layer.

[0082] In some embodiments, a raw feature matrix can be formed by fusing historical usage data (such as daily peak-to-valley ratio, weekly usage fluctuation, and peak-hour concentration) and real-time usage data (such as current load rate, number of vehicles in queue, and response speed) of each charging pile within the target area. This raw feature matrix is ​​then input into the input layer. The input layer is a fully connected linear layer with the same number of neurons as the feature dimension of the raw feature matrix. After processing with a weighted bias term in the input layer, the input feature vector is obtained.

[0083] S122. Based on the input feature vector, a low-dimensional latent space vector is obtained through the coding layer.

[0084] In some embodiments, the encoding layer may employ a multi-layer fully connected neural network (such as two ReLU-activated hidden layers) or an autoencoder structure to achieve feature compression through nonlinear transformation. The encoding layer extracts the core features of the data through dimensionality reduction, mapping the high-dimensional input feature vector to a low-dimensional latent space, reducing computational complexity while retaining key information, ultimately obtaining the low-dimensional latent space vector.

[0085] S123. Based on the low-dimensional potential space vector, the charging pile group and the group label are obtained through the clustering layer.

[0086] In some embodiments, obtaining the charging pile group and the group label through the clustering layer based on the low-dimensional latent space vector includes the following steps S1231 to S1238:

[0087] S1231. Through the clustering layer, the low-dimensional potential space vector is transformed into a set of data points, wherein each data point in the set of data points corresponds one-to-one with each charging pile in the target area.

[0088] In some embodiments, each feature vector of a charging pile in the latent space vector can be transformed into a data point in Euclidean space to form a set of data points, with each data point uniquely corresponding to a charging pile.

[0089] S1232. Randomly determine a preset number of data points in the data point set as the initial data point center, wherein the preset number is used to characterize the preset number of the initial charging pile group.

[0090] In some embodiments, the initial number of charging pile groups K is determined based on prior knowledge or the elbow rule (e.g., K=5 in the embodiment), and 5 data points are randomly selected as the initial data centers.

[0091] S1233. Calculate the first Euclidean distance between each data point and the center of each initial data point.

[0092] S1234. Assign each data point to the initial charging pile group corresponding to the smallest Euclidean distance in the first Euclidean distance.

[0093] S1235. The mean of the initial charging pile group is calculated using the preset mean calculation formula set in the clustering layer.

[0094] In some embodiments, the preset mean calculation formula can be a K-means confidence formula, and the mean can represent the center of each initial charging pile group.

[0095] S1236. Repeat the above steps until the change threshold of the mean remains within the preset change threshold range, and obtain the center of the target data point based on the current mean.

[0096] In some embodiments, the distance calculation and assignment process can be repeated until the change threshold of the centers between two adjacent iterations (e.g., the maximum Euclidean distance) is less than a preset change threshold interval (e.g., <10). -3 This yields a stable target data point center.

[0097] S1237. Calculate the second Euclidean distance between each data point and the center of each target data point.

[0098] S1238. Assign each data point to the charging pile group corresponding to the smallest Euclidean distance in the second Euclidean distance, and determine the group label according to the data characteristics corresponding to the charging pile group.

[0099] In some embodiments, semantic labels (such as "transportation hub type" or "commercial intensive type") can be automatically generated based on the statistical characteristics of data features within the group (such as average load rate or traffic correlation), or the group labels can be determined through preset rule mapping (such as the label generation function in the patent embodiment).

[0100] S130. Based on real-time traffic information within the target area and the charging pile group, construct a correlation function to characterize the relationship between traffic conditions within the target area and the grouping status of the charging pile group.

[0101] In some embodiments, the real-time traffic information includes real-time traffic flow information within the target area and traffic road information within the target area.

[0102] Specifically, the real-time traffic flow information may include the real-time traffic flow (vehicles / minute), average vehicle speed (km / h), congestion index (0-1, 1 for severe congestion, 0 for no congestion), location of traffic accidents and scope of impact (e.g., causing a 50% decrease in traffic efficiency on roads within a 300-meter radius of the target area).

[0103] The traffic and road information may include: road network topology (e.g., nodes are intersections, edges are roads, including attributes such as number of lanes and speed limits), geographical location of charging piles (latitude and longitude coordinates), and surrounding POI information (e.g., shopping malls and residential areas affecting pedestrian flow).

[0104] In some embodiments, the construction of a correlation function based on real-time traffic information within the target area and the charging pile group to characterize the relationship between traffic conditions within the target area and the grouping status of the charging pile group includes the steps S131 to S132 as shown below:

[0105] S131. Construct a spatiotemporal feature matrix based on the real-time traffic flow information and the traffic road information.

[0106] In some embodiments, an undirected graph can be constructed using the locations of charging piles and intersections within a target area. The undirected graph includes a first node and a second node, where the first node represents the location of the charging pile and the second node represents the location of the intersection. For each charging pile node in the undirected graph, road network features within a 5km radius (such as average congestion index and traffic efficiency of the three nearest intersections) are extracted to form a node spatial feature vector.

[0107] In some embodiments, real-time traffic flow information (e.g., a real-time congestion index sequence) from the past eight real-time time steps (e.g., one step every 15 minutes) is embedded to form a real-time time feature matrix.

[0108] In some embodiments, the spatiotemporal feature matrix is ​​obtained by fusing the node spatial feature vector with the real-time feature matrix.

[0109] S132. Based on the group label and the spatiotemporal feature matrix, the correlation function is obtained through a second preset neural network model.

[0110] In some embodiments, such as Figure 3 The second preset neural network model includes an encoder self-attention layer, a feedforward network layer, a decoder attention layer, and a fully connected output layer.

[0111] In some embodiments, obtaining the correlation function based on the group label and the spatiotemporal feature matrix using a second preset neural network model includes the following steps S1321 to S1324:

[0112] S1321. Based on the group label and the spatiotemporal feature matrix, obtain the attention feature matrix through the encoder's self-attention layer.

[0113] In some embodiments, the group labels (such as "transportation hub type" and "commercial intensive type") are converted into group label encoding vectors by the encoder self-attention layer, and the spatiotemporal feature matrix is ​​linearly transformed to obtain the query Q, key K and value V corresponding to the spatiotemporal feature matrix.

[0114] In some embodiments, multi-head attention is calculated based on the query Q, key K, and value V, and the calculated multi-head attention is concatenated with the group label encoding vector to obtain the attention feature matrix.

[0115] S1322. Based on the attention feature matrix, an enhanced feature matrix is ​​obtained through the feedforward network layer.

[0116] In some embodiments, the feedforward network layer may employ a two-layer fully connected neural network structure, activated by the ReLU function in between. The enhanced feature matrix is ​​obtained by capturing the nonlinear spatiotemporal correlations in the attention feature matrix through the feedforward network layer.

[0117] S1323. Based on the enhanced feature matrix and the group label, the decoded feature matrix is ​​obtained through the decoder attention layer.

[0118] In some embodiments, the group label is used as the query Q′, the enhanced feature matrix is ​​used as the key K′ and value V′, and the attention weight of each group label to the spatiotemporal feature is calculated through the decoder attention layer, thereby obtaining the decoded feature matrix, which represents the correlation strength between each charging pile group and the spatiotemporal feature matrix.

[0119] S1324. Based on the decoded feature matrix, the correlation function is obtained through the fully connected output layer.

[0120] In some embodiments, the decoded feature matrix is ​​converted into parameters of a correlation function through a fully connected output layer, and the correlation function is output. In some embodiments, the correlation function can be represented by the following expression:

[0121] Charging pile power demand = f(congestion index, group label) = 0.6 × congestion index + 0.4 × group priority

[0122] The congestion index can be included in the real-time traffic flow information. For example, the congestion index ranges from 0 to 1, where 1 indicates severe congestion and 0 indicates no congestion.

[0123] S140. Based on the correlation function, the power resource supply data, and the real-time usage data, a target power resource allocation plan is obtained.

[0124] In some embodiments, obtaining the target power resource allocation scheme based on the correlation function, the power resource supply data, and the real-time usage data includes the following steps S141 to S143:

[0125] S141. Construct an objective function and corresponding constraints based on the correlation function, the power resource supply data, and the real-time usage data.

[0126] S142. Solve the objective function to obtain the objective decision variables, wherein the objective decision variables include the real-time power allocation of each charging pile in the target area and the total power allocation of each charging pile group.

[0127] In some embodiments, the target decision variables include a first target decision variable and a second target decision variable, and the objective function and the constraints are represented by the following expressions;

[0128]

[0129] Where min[] represents the objective of the objective function, and P = [P1, P2, ..., PN]. N G represents the number of charging stations in the target area, P represents the real-time power allocation for each charging station, and G represents the number of charging stations in the target area. k This represents the k-th charging pile group. The dynamic weight is calculated based on the correlation function and is used to characterize the degree of influence of traffic conditions on the priority of the charging pile group. This represents the total power allocation for the k-th charging pile group. λ represents the electricity cost for the k-th charging pile group, where K represents the number of charging pile groups, and λ represents a preset penalty coefficient, which characterizes the degree of balance between supply and demand in electricity allocation. This represents the real-time electricity demand forecast value for the k-th charging pile group, which is calculated using the correlation function. This represents the supply-demand imbalance penalty term in the objective function. If the total power allocation of the k-th charging pile group is less than its real-time power demand forecast, a positive difference is generated. The square of this difference is then added to the objective function as a penalty term. i Smax represents the real-time power allocation of the i-th charging pile, and Smax represents the upper limit of power supply in the target area.

[0130] S143. The target decision variable is determined as the target power resource allocation scheme.

[0131] In some embodiments, P and The target decision variables, namely the target power resource allocation scheme, include the real-time power allocation of each charging pile in the target area and the total power allocation of each charging pile group.

[0132] In some embodiments, in conjunction with the specific application scenarios described above, the groups can be categorized from high to low priority as follows:

[0133] The “transportation hub type” charging pile group includes 10 charging piles, which are located near highway service areas and have the highest priority during peak charging times. The total power allocation is 600kW.

[0134] The “fast charging station cluster in commercial areas” includes 15 charging stations located around the shopping center, which have fast charging needs and a total power allocation of 500kW.

[0135] It also includes a "slow-speed" charging pile group for residential areas (containing 20 charging piles) and a "dedicated" charging pile group for industrial areas (containing 5 charging piles).

[0136] In summary, this application provides a method for allocating power resources for charging piles within a target area. Through dynamic priority weighting and a supply-demand balance penalty mechanism, it implements precise and efficient management of power resource allocation for charging piles within the target area, thereby minimizing the overall cost of power resource allocation for charging equipment within the target area and ensuring the balance of power supply and demand.

[0137] To better implement the above methods, this application also provides a power resource allocation device for charging piles within a region. This device can be integrated into an electronic system, which can be a system, a server, or other similar system. The system can be a mobile phone, tablet computer, smart Bluetooth system, laptop computer, personal computer, or other similar system; the server can be a single server or a server cluster composed of multiple servers.

[0138] For example, in this embodiment, the method of this application embodiment will be described in detail by taking the power resource allocation device of the charging piles in the region as specifically integrated into the system.

[0139] For example, such as Figure 4 As shown, the power resource allocation device 400 for charging piles within this domain may include a first unit 401, a second unit 402, a third unit 403, and a fourth unit 404. The device includes:

[0140] The first unit 401 is used to acquire historical usage data, real-time usage data, and power resource supply data of multiple charging piles in the target area;

[0141] The second unit 402 is used to divide each charging pile into at least two different charging pile groups based on the historical usage data and the real-time usage data.

[0142] The third unit 403 is used to construct an association function based on real-time traffic information within the target area and the charging pile group to characterize the relationship between traffic conditions within the target area and the grouping status of the charging pile group.

[0143] The fourth unit 404 is used to obtain a target power resource allocation scheme based on the correlation function, the power resource supply data, and the real-time usage data.

[0144] In practice, each of the above units can be implemented as an independent entity or can be arbitrarily combined to be implemented as the same or several entities. For the specific implementation of each of the above units, please refer to the previous method embodiments, which will not be repeated here.

[0145] As can be seen from the above, the embodiments of this application can minimize the overall cost of power resource allocation for charging equipment within the target area and ensure the balance between power supply and demand.

[0146] This application also provides an electronic system, which can be a system, a server, or other similar system. The system can be a mobile phone, tablet computer, smart Bluetooth system, laptop computer, personal computer, etc.; the server can be a single server or a server cluster composed of multiple servers, etc.

[0147] In some embodiments, the product processing device may also be integrated into multiple electronic systems. For example, the product processing device may be integrated into multiple servers, and the power resource allocation method for charging piles within the domain of this application may be implemented by multiple servers.

[0148] In this embodiment, the electronic system of this embodiment will be used as an example for detailed description, such as... Figure 5 As shown, it illustrates a structural diagram of the system 500 involved in the embodiments of this application. Specifically:

[0149] The system 500 may include components such as a processor 501 with one or more processing cores, a memory 502 with one or more media, a power supply 503, an input module 504, and a communication module 505. Those skilled in the art will understand that... Figure 5 The structure of system 500 shown does not constitute a limitation on system 500, and may include more or fewer components than shown, or combine certain components, or have different component arrangements. Wherein:

[0150] The processor 501 is the control center of the system 500. It connects various parts of the system 500 via various interfaces and lines. By running or executing software programs and / or modules stored in the memory 502, and by calling data stored in the memory 502, it performs various functions of the system 500 and processes data, thereby providing overall monitoring of the system 500. In some embodiments, the processor 501 may include one or more processing cores; in some embodiments, the processor 501 may integrate an application processor and a modem processor, wherein the application processor mainly handles the operating system, user interface, and applications, and the modem processor mainly handles wireless communication. It is understood that the modem processor may also not be integrated into the processor 501.

[0151] The memory 502 can be used to store software programs and modules. The processor 501 executes various functional applications and data processing by running the software programs and modules stored in the memory 502. The memory 502 may mainly include a program storage area and a data storage area. The program storage area may store the operating system, application programs required for at least one function (such as sound playback function, image playback function, etc.), etc.; the data storage area may store data created according to the use of the system 500, etc. In addition, the memory 502 may include high-speed random access memory, and may also include non-volatile memory, such as at least one disk storage device, flash memory device, or other volatile solid-state storage device. Accordingly, the memory 502 may also include a memory controller to provide the processor 501 with access to the memory 502.

[0152] System 500 also includes a power supply 503 that supplies power to the various components. In some embodiments, power supply 503 can be logically connected to processor 501 through a power management system, thereby enabling functions such as charging, discharging, and power consumption management through the power management system. Power supply 503 may also include one or more DC or AC power supplies, recharging systems, power fault detection circuits, power converters or inverters, power status indicators, and other arbitrary components.

[0153] The system 500 may also include an input module 504, which can be used to receive input digital or character information and generate keyboard, mouse, joystick, optical or trackball signal inputs related to user settings and function control.

[0154] The system 500 may also include a communication module 505. In some embodiments, the communication module 505 may include a wireless module, through which the system 500 can perform short-range wireless transmission, thereby providing users with wireless broadband internet access. For example, the communication module 505 can be used to help users send and receive emails, browse web pages, and access streaming media.

[0155] Although not shown, system 500 may also include a display unit, etc., which will not be described in detail here. Specifically, in this embodiment, the processor 501 in system 500 loads the executable files corresponding to the processes of one or more applications into memory 502 according to the following instructions, and the processor 501 runs the applications stored in memory 502 to realize various functions, as follows:

[0156] Acquire historical usage data, real-time usage data, and power supply data of multiple charging piles within the target area;

[0157] Based on the historical usage data and the real-time usage data, each charging pile is divided into at least two different charging pile groups;

[0158] Based on real-time traffic information within the target area and the charging pile group, a correlation function is constructed to characterize the relationship between traffic conditions within the target area and the grouping status of the charging pile group.

[0159] Based on the correlation function, the power supply data, and the real-time usage data, a target power resource allocation plan is obtained.

[0160] As can be seen from the above, the embodiments of this application can minimize the overall cost of power resource allocation for charging equipment within the target area and ensure the balance between power supply and demand.

[0161] Those skilled in the art will understand that all or part of the steps in the various methods of the above embodiments can be accomplished by instructions, or by instructions controlling related hardware. These instructions can be stored in a medium and loaded and executed by a processor.

[0162] To this end, embodiments of this application provide a medium storing multiple instructions that can be loaded by a processor to execute steps in any of the power resource allocation methods for charging piles within a domain provided in embodiments of this application. For example, the instructions can execute the following steps:

[0163] Acquire historical usage data, real-time usage data, and power supply data of multiple charging piles within the target area;

[0164] Based on the historical usage data and the real-time usage data, each charging pile is divided into at least two different charging pile groups;

[0165] Based on real-time traffic information within the target area and the charging pile group, a correlation function is constructed to characterize the relationship between traffic conditions within the target area and the grouping status of the charging pile group.

[0166] Based on the correlation function, the power supply data, and the real-time usage data, a target power resource allocation plan is obtained.

[0167] The medium may include: read-only memory (ROM), random access memory (RAM), disk or optical disk, etc.

[0168] According to one aspect of this application, a computer program product or computer program is provided, comprising computer instructions stored in a medium. A processor of a computer system reads the computer instructions from the medium and executes the computer instructions, causing the computer system to perform the methods provided in the various optional implementations of the above embodiments.

[0169] Since the instructions stored in the medium can execute the steps in any of the power resource allocation methods for charging piles within the domain provided in the embodiments of this application, the beneficial effects that any of the power resource allocation methods for charging piles within the domain provided in the embodiments of this application can achieve can be realized. For details, please refer to the previous embodiments, which will not be repeated here.

[0170] The above provides a detailed description of a power resource allocation method, device, system, and medium for charging piles within a domain, as provided in the embodiments of this application. Specific examples have been used to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the method and core ideas of this application. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of this application. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A method for allocating power resources for charging piles within a region, characterized in that, include: Acquire historical usage data, real-time usage data, and power supply data of multiple charging piles within the target area; Based on the historical usage data and the real-time usage data, each charging pile is divided into at least two different charging pile groups; Based on real-time traffic information within the target area and the charging pile group, a correlation function is constructed to characterize the relationship between traffic conditions within the target area and the grouping status of the charging pile group. Based on the correlation function, the power supply data, and the real-time usage data, a target power resource allocation plan is obtained.

2. The method as described in claim 1, characterized in that, The method of dividing each charging pile into at least two different charging pile groups based on the historical usage data and the real-time usage data includes: Based on the historical usage data and the real-time usage data, the charging pile group and the corresponding group label are obtained through a first preset neural network model.

3. The method as described in claim 2, characterized in that, The first preset neural network model includes an input layer, an encoding layer, and a clustering layer; The step of obtaining the charging pile group and its corresponding group label based on the historical usage data and the real-time usage data through a first preset neural network model includes: Based on the historical usage data and the real-time usage data, an input feature vector is obtained through the input layer; Based on the input feature vector, a low-dimensional latent space vector is obtained through the coding layer; Based on the low-dimensional latent space vector, the charging pile group and the group label are obtained through the clustering layer.

4. The method as described in claim 3, characterized in that, The step of obtaining the charging pile group and the group label through the clustering layer based on the low-dimensional latent space vector includes: The clustering layer transforms the low-dimensional latent space vector into a set of data points, where each data point in the set corresponds one-to-one with each charging pile in the target area. A preset number of data points in the data point set are randomly selected as the initial data point centers, wherein the preset number is used to characterize the preset number of the initial charging pile group; The first Euclidean distance between each data point and the center of each initial data point is calculated. Each data point is assigned to the initial charging pile group corresponding to the smallest Euclidean distance in the first Euclidean distance; The mean of the initial charging pile group is calculated using the preset mean calculation formula set in the clustering layer. Repeat the above steps until the change threshold of the mean remains within the preset change threshold range, and obtain the center of the target data point based on the current mean. The second Euclidean distance between each data point and the center of each target data point is calculated. Each data point is assigned to a charging pile group corresponding to the smallest Euclidean distance in the second Euclidean distance, and the group label is determined according to the data characteristics corresponding to the charging pile group.

5. The method as described in claim 2, characterized in that, The real-time traffic information includes real-time traffic flow information within the target area and traffic road information within the target area; Based on real-time traffic information within the target area and the charging pile group, a correlation function is constructed to characterize the relationship between traffic conditions within the target area and the grouping status of the charging pile group, including: Based on the real-time traffic flow information and the traffic road information, a spatiotemporal feature matrix is ​​constructed; The correlation function is obtained through a second preset neural network model based on the group label and the spatiotemporal feature matrix.

6. The method as described in claim 5, characterized in that, The second preset neural network model includes an encoder self-attention layer, a feedforward network layer, a decoder attention layer, and a fully connected output layer; The step of obtaining the association function based on the group label and the spatiotemporal feature matrix through a second preset neural network model includes: Based on the group labels and the spatiotemporal feature matrix, an attention feature matrix is ​​obtained through the encoder's self-attention layer; Based on the attention feature matrix, an enhanced feature matrix is ​​obtained through the feedforward network layer; Based on the enhanced feature matrix and the group label, the decoded feature matrix is ​​obtained through the decoder attention layer; The correlation function is obtained through the fully connected output layer based on the decoded feature matrix.

7. The method as described in claim 1, characterized in that, The process of obtaining the target power resource allocation plan based on the correlation function, the power resource supply data, and the real-time usage data includes: Based on the correlation function, the power resource supply data, and the real-time usage data, construct the objective function and the corresponding constraints for the objective function; Solve the objective function to obtain the objective decision variables, wherein the objective decision variables include the target real-time power allocation of each charging pile in the target area and the power allocation of each charging pile group; The target decision variable is determined as the target power resource allocation scheme.

8. A power resource allocation device for charging piles within a region, characterized in that, The device includes: The first unit is used to acquire historical usage data, real-time usage data, and power supply data of multiple charging piles within the target area; The second unit is used to divide each charging pile into at least two different charging pile groups based on the historical usage data and the real-time usage data. The third unit is used to construct a correlation function based on real-time traffic information within the target area and the charging pile group to characterize the relationship between traffic conditions within the target area and the grouping status of the charging pile group. The fourth unit is used to obtain the target power resource allocation plan based on the correlation function, the power resource supply data, and the real-time usage data.

9. A system, characterized in that, The method includes a processor and a memory, the memory storing multiple instructions; the processor loads instructions from the memory to perform the steps of the method as described in any one of claims 1 to 7.

10. A medium, characterized in that, The medium stores a plurality of instructions adapted for loading by a processor to execute the steps of the method according to any one of claims 1 to 7.

Images