Electric vehicle charging resource planning and optimizing method and system

Abstract

The invention provides an electric vehicle charging resource planning and optimizing method, and aims to solve the problems of static performance, single target optimization, one-sided attraction evaluation, disjunction with traffic jam management and the like in existing planning. The method comprises the following steps: firstly, acquiring regional basic data, and establishing a charging station composite attraction evaluation model which comprehensively considers POI value, plot cost, estimated available charging pile quantity proportion and reputation parameters determined based on historical service data, so as to predict the probability of selecting each charging station by a user; and optimizing charging pile capacity configuration according to the predicted user selection probability and the total charging demand so as to minimize the combined cost of the construction cost and the average queuing waiting time of the users. The optimal charging site is screened and recommended through weighted aggregation of multiple indexes such as composite attraction, construction cost, queuing waiting time and the like. Scientific site selection and reasonable capacity configuration of the charging stations are achieved, urban traffic congestion is effectively relieved, balanced distribution of the charging flow is promoted, and the charging experience of users is improved.

Description

Technical Field

[0001] The present invention belongs to the fields of intelligent transportation, traffic engineering, urban planning and new energy vehicle infrastructure planning, and in particular relates to a method and system for planning and optimizing electric vehicle charging resources. Background Art

[0002] With the rapid popularization of electric vehicles, the construction and layout of charging infrastructure are becoming increasingly important. However, existing charging pile planning schemes have the following problems:

[0003] First, most existing planning schemes are static, which makes it difficult to adapt to the dynamic changes in urban traffic flow, user travel patterns, and energy market prices. They focus on single-objective optimization and lack consideration of the complex trade-offs and coordinated optimization among multiple objectives.

[0004] Secondly, the traditional charging pile site selection method fails to fully consider the potential impact of the layout of charging stations on urban traffic flow and congestion. The unreasonable distribution of charging stations may lead to excessive concentration of traffic flow and aggravate local traffic pressure.

[0005] Third, existing planning models often evaluate the attractiveness of charging stations to users based solely on factors such as the number of POIs (points of interest) or geographical distance, failing to fully consider the real impact of land costs, the real-time service capabilities of the station, and the accuracy and reliability of charging station service information on user choices. This may result in the planned station sites performing poorly in actual operations or users distrusting the service information.

[0006] Fourth, due to the above-mentioned defects, existing plans often lead to uneven distribution of charging resources in space and time. In some areas, charging facilities are over-concentrated, causing congestion, while in other areas, resources are idle, reducing the overall utilization efficiency of the infrastructure and affecting the charging experience of electric vehicle users.

[0007] Therefore, a method for planning and optimizing electric vehicle charging resources that can overcome the above-mentioned defects is needed to achieve scientific site selection of charging stations and reasonable configuration of charging pile capacity, effectively alleviate urban traffic congestion, promote balanced distribution of charging flow, and enhance user charging experience.

[0008] It should be noted that the information disclosed in the above background technology section is only used to enhance the understanding of the background of this application, and therefore may include information that does not constitute prior art known to ordinary technicians in this field. Summary of the Invention

[0009] In view of this, the present invention provides an electric vehicle charging resource planning and optimization method and system to overcome the problems of static charging pile planning in the prior art, such as single-objective optimization, one-sided attractiveness evaluation, and disconnection between planning and traffic congestion management. The method achieves scientific site selection for charging stations and reasonable configuration of charging pile capacity, effectively alleviates urban traffic congestion, promotes balanced distribution of charging traffic, and enhances the user charging experience.

[0010] An embodiment of the present invention provides a method for planning and optimizing electric vehicle charging resources, comprising the following steps:

[0011] Obtain geographic block information within the target planning area, road network connectivity information between blocks, POI distribution data within each block, and the geographic location and number of existing charging stations;

[0012] For each existing charging station and potential candidate, the composite attractiveness is quantified and calculated. The composite attractiveness is calculated using a preset weighting function based on a charging station attractiveness evaluation model based on the POI value of the geographical area where the charging station is located, the land cost, the estimated proportion of available charging piles at the charging station, and a reputation parameter determined based on the charging station's historical service quality data.

[0013] For users within the planning area, the probability of users selecting each charging station is predicted based on the geographical distance from the user to each charging station, the combined attractiveness of the charging station, and a preset distance attenuation factor.

[0014] The charging demand allocated to each charging station is calculated based on the predicted probability of users selecting each charging station and the total charging demand within the planning period. Based on the allocated charging demand, and with the goal of minimizing the combined cost of charging station construction and average user waiting time, the incremental number of charging piles to be configured at each existing charging station and the initial number of charging piles to be configured at each potential candidate charging station are determined.

[0015] For all potential candidate charging stations, their composite attractiveness, construction cost, and estimated average user waiting time are extracted as screening indicators. After normalizing the screening indicators, they are weighted and aggregated according to preset weights to obtain a comprehensive evaluation value for each candidate charging station. All candidate charging stations are ranked according to the comprehensive evaluation value, and a predetermined number of charging stations with the highest rankings are selected as recommended construction sites.

[0016] In some optional embodiments, the POI value is configured to be calculated based on the number of different types of POIs in the block and their respective preset weight values.

[0017] In some optional embodiments, the reputation parameter is configured to be obtained by iterative updating based on a historical record of deviations between the unit charging prices announced by the charging station in past periods and the unit charging prices actually paid by the user.

[0018] In some optional embodiments, the composite attractiveness is further corrected during calculation based on the traffic flow to road capacity ratio of the geographical area where the charging station is located or the road connected to the charging station. When the traffic flow to road capacity ratio increases, the composite attractiveness value decreases.

[0019] In some optional embodiments, the distance decay factor is configured to be dynamically adjusted based on the average traffic congestion level of the road network; as the average traffic congestion level increases, the impact of the distance decay factor increases.

[0020] In some optional embodiments, the average waiting time of users in queue is configured to adopt the M / M / c queuing theory model and is calculated according to the number of charging piles configured in the charging station, the user arrival rate and the unit charging service rate.

[0021] In some optional embodiments, the construction cost of the charging station includes the land acquisition cost and the charging pile procurement and installation cost.

[0022] In some optional embodiments, determining the target of the incremental configuration quantity and the initial configuration quantity of charging piles further includes satisfying a preset power grid power supply capacity constraint.

[0023] In some optional embodiments, the weight parameters of the indicators are determined by an expert scoring method or an entropy weight method.

[0024] In some optional embodiments, the method further includes a rolling planning step: updating the recommended construction site information to the geographical location and number of charging piles of the existing charging station, and using this as new input data, repeatedly executing the method to perform dynamic optimization planning.

[0025] In some optional embodiments, the road network connection information further includes travel time estimates for connections between geographic blocks, and the travel time estimates are obtained by weighted calculation based on basic weights determined by historical traffic data, adjustment factors for real-time traffic conditions, and event-driven adjustment factors for special event information.

[0026] In some optional embodiments, the goal of minimizing the combined cost further includes external traffic congestion costs that may be caused by the operation of the charging station, where the external traffic congestion costs are quantified based on the traffic congestion level in the area where the charging station is located and the estimated user flow.

[0027] An embodiment of the present invention provides an electric vehicle charging resource planning and optimization system for implementing the above-mentioned electric vehicle charging resource planning and optimization method, including:

[0028] A data acquisition module is configured to obtain geographic block information within the target planning area, road network connectivity information between blocks, POI distribution data within each block, and the geographic location and number of existing charging stations;

[0029] An attractiveness assessment module is configured to quantify the composite attractiveness of each existing charging station and potential candidate charging station. The composite attractiveness is calculated using a preset weighting function based on the POI value of the geographical block where the charging station is located, the land cost, the estimated proportion of available charging piles at the charging station, and a reputation parameter determined based on the charging station's historical service quality data;

[0030] A user selection prediction module is configured to predict the probability of users selecting each charging station within the planning area based on the geographical distance between the user and each charging station and the composite attractiveness of the charging station, combined with a preset distance attenuation factor;

[0031] The resource allocation optimization module is configured to calculate the charging demand allocated to each charging station based on the predicted probability of users selecting each charging station and the total charging demand within the planning period. Based on the allocated charging demand, and with the goal of minimizing the combined cost of charging station construction and the average user waiting time, the module determines the incremental number of charging piles to be configured at each existing charging station and the initial number of charging piles to be configured at each potential candidate charging station.

[0032] The decision-making screening module is configured to extract indicators such as the composite attractiveness, construction cost, and estimated average waiting time of users for all potential candidate charging stations; after normalizing the indicators, the normalized indicators are weighted and aggregated according to preset weights to obtain a comprehensive evaluation value for each candidate charging station; all candidate charging stations are ranked according to the comprehensive evaluation value, and a predetermined number of charging stations with the highest ranking are selected as recommended construction sites.

[0033] In some optional embodiments, the attractiveness evaluation module is configured to make corrections based on the traffic flow to road capacity ratio of the geographical area where the charging station is located or the road connected to the charging station when calculating the composite attractiveness, so that when the traffic flow to road capacity ratio value increases, the composite attractiveness value decreases.

[0034] In some optional embodiments, the reputation parameter is configured to be obtained by iterative updating based on a historical record of deviations between the unit charging prices announced by the charging station in past periods and the unit charging prices actually paid by the user.

[0035] In some optional embodiments, the resource configuration optimization module is further configured to meet a preset grid power supply capacity constraint when determining the incremental configuration quantity and the initial configuration quantity of charging piles.

[0036] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the disclosure.

[0037] The electric vehicle charging resource planning and optimization method and system of the present invention have the following beneficial effects:

[0038] The electric vehicle charging resource planning and optimization method and system proposed in the present invention can comprehensively and dynamically evaluate the attractiveness of charging stations by comprehensively considering multi-dimensional factors such as POI value, land cost, estimated number of available piles, traffic congestion damping coefficient and information reputation parameters, so that the planning scheme is closer to actual operation and user choice. This method has a built-in traffic congestion damping coefficient in the attractiveness evaluation model and takes into account the traffic congestion cost in the multi-index decision-making, which helps to guide the planning scheme to avoid excessive concentration in already congested areas, thereby assisting urban traffic management. In addition, the present invention incorporates the average waiting time of users in the queue into the optimization target and takes into account the impact of site service reputation on user choice, which helps to improve the overall charging convenience and user satisfaction. By minimizing the combined cost of construction cost and user expected queuing loss, and optimizing under constraints such as grid power supply capacity, the economy and rationality of the number of charging piles configured are ensured. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Other features, objects and advantages of the present application will become more apparent from the detailed description of non-limiting embodiments made with reference to the following drawings.

[0040] Figure 1 This is a flow chart of a method for planning and optimizing electric vehicle charging resources according to an embodiment of the present application;

[0041] Figure 2 It is a structural diagram of an electric vehicle charging resource planning and optimization system according to an embodiment of the present application. DETAILED DESCRIPTION

[0042] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be embodied in many forms and should not be construed as limited to the examples set forth herein; rather, these embodiments are provided so that this disclosure will be thorough and complete and will fully convey the concepts of the example embodiments to those skilled in the art. The described features, structures, or characteristics may be combined in any suitable manner in one or more embodiments.

[0043] In addition, the accompanying drawings are merely schematic illustrations of the present disclosure and are not necessarily drawn to scale. Identical reference numerals in the figures denote identical or similar parts, and thus repetitive descriptions thereof will be omitted. Some of the block diagrams shown in the accompanying drawings are functional entities that do not necessarily correspond to physically or logically separate entities. These functional entities may be implemented in software, in one or more hardware modules or integrated circuits, or in different networks and / or processor devices and / or microcontroller devices.

[0044] The flowcharts shown in the accompanying drawings are merely exemplary and do not necessarily include all steps. For example, some steps may be decomposed, while some steps may be combined or partially combined. Therefore, the actual execution order may change according to actual circumstances.

[0045] The present invention relates to charging resource planning and optimization. By quantifying the factors influencing user selection of charging stations, user behavior is predicted and used as a basis for optimizing charging resource allocation. User selection of charging stations is influenced by a variety of factors, including geographic location, surrounding environment, service capabilities, and price reputation. Geographic location influences user selection through a distance decay function: the farther the distance, the lower the probability of selection. The surrounding environment reflects demand potential through POIs (points of interest); the higher the POI value, the greater the appeal. Service capabilities are reflected by the ratio of available charging piles; the higher the ratio, the more users are inclined to choose. Price reputation is assessed through historical price deviations: the greater the deviation, the lower the trust, and the lower the user's willingness to choose. By quantifying these factors and constructing a reasonable selection model, it is possible to predict the probability of users choosing different charging stations. Using an optimization algorithm, charging demand is allocated based on the predicted user selection probability, and the number of charging piles configured is determined with the goal of minimizing cost, thereby achieving reasonable planning and optimization of charging resources.

[0046] like Figure 1 As shown, an embodiment of the present invention provides a method for planning and optimizing electric vehicle charging resources, which realizes the scientific site selection and capacity configuration of charging stations through the coordinated action of multiple steps, including:

[0047] S100: Obtain geographic block information within the target planning area, road network connectivity information between blocks, POI distribution data within each block, and the geographic locations and number of existing charging stations. This aims to obtain basic data for the planning area and lay a data foundation for subsequent analysis. Geographic block information represents a gridded division of the planning area to facilitate data management and calculation. The block size can be adjusted based on actual needs, for example, using a 300m×300m or 1km×1km square grid. Road network connectivity information reflects the accessibility between blocks. Travel time estimates can be dynamically weighted based on historical traffic data, real-time traffic conditions, and special event information. For example, weights can be adjusted using floating vehicle data or real-time road condition information released by traffic management authorities. Point of interest (POI) distribution data reflects the distribution of different types of places (such as residences, workplaces, shopping centers, etc.) within each block. It can be obtained using the API provided by the map service provider and classified and weighted according to the actual scenario. The geographic location and number of existing charging stations serve as the starting point for planning and provide a basis for subsequent expansion or new construction.

[0048] S200. For each existing charging station and potential candidate charging station, quantitatively calculate its composite attractiveness. Composite attractiveness is calculated using a preset weighting function to establish a charging station attractiveness evaluation model based on the POI value of the geographical block where the charging station is located, the land cost, the estimated ratio of available charging piles at the charging station, and a reputation parameter determined based on the charging station's historical service quality data. A charging station attractiveness evaluation model is established to quantify the attractiveness of each charging station to users, providing a basis for user selection predictions. Composite attractiveness is a quantitative indicator that comprehensively considers multiple factors, including POI value, which reflects the demand potential of the block where the station is located; land cost is a negative constraint; the estimated ratio of available charging piles reflects the station's real-time service capabilities; and reputation parameters, which reflect the reliability of the station's service quality. Reputation parameters can be referenced, for example, by user evaluation scores or platform complaint rates.

[0049] S300: For users within the planning area, the probability of the user selecting each charging station is predicted based on the user's geographical distance from each charging station, the charging station's composite attractiveness, and a preset distance decay factor. Predicting user selection probability is a prerequisite for rational resource allocation. A Huff model or a Logit model can be used to predict the user's geographical distance from each charging station, the composite attractiveness, and the distance decay factor. The distance decay factor can be dynamically adjusted based on the average traffic congestion level of the road network. The higher the traffic congestion, the more sensitive the user is to distance.

[0050] S400. Calculate the charging demand allocated to each charging station based on the predicted probability of the user selecting each charging station and the total charging demand within the planning period. Based on the allocated charging demand, and with the goal of minimizing the combined cost of charging station construction and the average user waiting time, determine the incremental number of charging piles to be configured at each existing charging station and the initial number of charging piles to be configured at each potential candidate charging station. Optimize charging station resource allocation. The goal of charging resource allocation is to minimize construction and service costs while meeting user needs. Based on the predicted probability of the user selecting each charging station and the total charging demand within the planning period, a proportional allocation method or other demand allocation algorithm may be used to calculate the charging demand allocated to each charging station. Based on the allocated charging demand, with the goal of minimizing the combined cost of charging station construction costs and users' average waiting time in queues, linear programming, integer programming, or heuristic algorithms can be used to determine the incremental number of charging piles for each existing charging station and the initial number of charging piles for each potential candidate charging station. The construction cost of the charging station includes land costs, equipment procurement costs, and installation costs, etc. The average waiting time of users in queues can be estimated using a queuing theory model (such as the M / M / c model), and constraints such as grid power supply capacity constraints and land area constraints can be added.

[0051] S500: For all potential candidate charging stations, extract their composite attractiveness, construction cost, and estimated average user waiting time as screening indicators. After normalizing the screening indicators, perform weighted aggregation of the normalized screening indicators based on preset weights to obtain a comprehensive evaluation value for each candidate charging station. Rank all candidate charging stations based on the comprehensive evaluation values, and select a predetermined number of charging stations with the highest ranking as recommended construction sites. Perform multi-indicator decision-making screening, which aims to select the optimal solution from among the candidate sites. Extract indicators such as composite attractiveness, construction cost, and average user waiting time, normalize them, and perform weighted aggregation based on preset weights to obtain a comprehensive evaluation value for each candidate charging station. The weight parameters can be determined using an expert scoring method or an entropy weighting method. Rank all candidate charging stations based on the comprehensive evaluation values, and select a predetermined number of charging stations with the highest ranking as recommended construction sites. Through the above steps, the embodiments of the present invention can comprehensively consider multiple factors to achieve scientific planning and optimization of charging resources, thereby improving user charging experience, alleviating traffic congestion, and increasing resource utilization efficiency.

[0052] In some embodiments, the POI value is calculated using a weighted method based on the number of different POI types within a block and their respective preset weights. This embodiment is a specific implementation of composite attractiveness calculation. The POI value is intended to reflect the demand potential of the block in which the station is located. Different types of POI contribute differently to charging demand, necessitating a weighted approach. POI types can be categorized based on actual application scenarios, such as residences, workplaces, shopping centers, and entertainment venues. Each type can be assigned a different weight. Weights can be determined using various methods, such as expert scoring, analytic hierarchy process (AHP), and big data analysis. For example, a residence might have a weight of 0.8, a workplace 0.7, a shopping center 0.5, and an entertainment venue 0.3. Specific weights can be adjusted based on actual circumstances. The number of POIs can be retrieved through the map API, and the POI value for the block is calculated by multiplying the number by the weight and then summing the results. Normalization can also be performed to ensure the value is within a reasonable range, such as [0, 1]. This embodiment enables a more refined assessment of the demand potential of charging stations, improving planning accuracy.

[0053] In some embodiments, a reputation parameter is configured to be iteratively updated based on the historical deviation between the unit charging price publicly displayed by the charging station and the unit charging price actually paid by users over past periods. This embodiment is a specific implementation of composite attractiveness calculation, intended to quantify the reliability of charging station service information and thus influence user selection. This reputation parameter can be understood as a trust indicator, reflecting the accuracy of the charging station operator's price forecast. The more accurate the price forecast, the higher the user's trust in the charging station and the more likely they are to choose it. The historical period can be adjusted according to actual needs, such as the most recent week, the most recent month, or the most recent quarter. The historical deviation can be the absolute value, relative deviation, or weighted average deviation of the difference between the actual price and the predicted price. Iterative updates can be performed using various methods, such as exponential smoothing and Kalman filtering, and upper and lower limits can also be set to prevent excessive parameter fluctuations. For example, the trust factor can be set to a range of [0, 1], with an initial value of 0.5. If the actual price is higher than the predicted price, the trust factor is reduced, and vice versa. This embodiment can encourage charging station operators to improve the accuracy of price predictions, enhance user trust, and guide users to choose charging stations with more reliable service quality.

[0054] In some embodiments, the calculation of composite attractiveness is further modified based on the traffic flow to road capacity ratio (V / C ratio) of the geographic area where the charging station is located or the roads connecting to the charging station. As the V / C ratio increases, the composite attractiveness decreases. This embodiment improves the calculation of composite attractiveness and aims to address the prior art's neglect of the impact of charging station layout on traffic congestion. The V / C ratio is a quantitative indicator of traffic congestion; a higher V / C ratio indicates more severe congestion. By incorporating the V / C ratio into the composite attractiveness model and establishing a negative correlation, this embodiment encourages planning schemes to favor areas with good traffic conditions and avoid over-concentration in already congested areas. This helps balance the distribution of charging traffic and alleviate traffic pressure. The V / C ratio can be obtained in a variety of ways, such as obtaining real-time data from traffic management departments, predicting it based on historical traffic data, or even estimating it using a map API. Modification methods can also be varied, such as directly using the V / C ratio as a damping coefficient, or employing a piecewise or exponential function. A threshold for the V / C ratio can be pre-set; when it exceeds the threshold, the composite attractiveness is significantly reduced. This embodiment helps guide the construction of charging stations to avoid congested areas, thereby assisting urban traffic management and improving overall urban operation efficiency.

[0055] In some embodiments, the distance attenuation factor is dynamically adjusted based on the average traffic congestion level of the road network; when the average traffic congestion level increases, the impact of the distance attenuation factor increases. This embodiment is an improvement to the prediction of user selection probability, aiming to more accurately simulate the impact of traffic congestion on user selection behavior. The distance attenuation factor reflects the user's sensitivity to distance. The farther the distance, the lower the probability that the user will choose the charging station. In the case of traffic congestion, users are more inclined to choose charging stations that are closer to save time and avoid congestion. Therefore, it is necessary to dynamically adjust the distance attenuation factor based on the average traffic congestion level of the road network. When the average traffic congestion level increases, the impact of the distance attenuation factor increases, which means that the user's willingness to choose a long-distance charging station is further reduced. The average traffic congestion level can be measured by a variety of indicators, such as the average V / C ratio, average travel time, congestion index, etc. These data can be obtained through the API of the traffic management department or map service provider. The distance decay factor can be adjusted linearly, nonlinearly, or piecewise. For example, the following formula can be used: β(t) = β0 + k*V / C(t), where β(t) is the distance decay factor at time t, β0 is the baseline value, k is the adjustment coefficient, and V / C(t) is the average V / C ratio at time t. This embodiment can more realistically reflect the impact of traffic congestion on user charging choices, thereby improving the accuracy of user choice predictions.

[0056] In some embodiments, the average queuing waiting time of users is configured to adopt the M / M / c queuing theory model, which is calculated based on the number of charging piles configured at the charging station, the user arrival rate, and the unit charging service rate. This embodiment limits the specific calculation method of the average queuing waiting time of users in the optimization target. The M / M / c queuing theory model is a commonly used queuing system analysis model, which assumes that user arrivals obey the Poisson distribution (M), the service time obeys the exponential distribution (M), and the number of service stations is c. In an embodiment of the present invention, the model is used to estimate the average queuing waiting time of the charging station, which directly affects the user experience and is also an important factor to be considered in resource allocation optimization. The number of charging piles configured c is the input parameter of the model and is determined by the resource allocation optimization module. The user arrival rate λ represents the number of users arriving at the charging station per unit time, which can be obtained through historical data or a prediction model. The unit charging service rate μ represents the number of users that each charging pile can serve per unit time, which depends on the charging power and service efficiency of the charging pile. Based on the above parameters, the M / M / c model can calculate indicators such as the average queue length and the average queue waiting time. For example, the formula: E[T queue The average queuing time is calculated using the formula: ] = P(>0)*(E[service time] / (c-ρ)), where P(>0) is the queuing probability, ρ is the service intensity, and ρ = λ / (c*μ). Other queuing models can also be used, such as the M / G / c model or the G / G / c model. The M / M / c model can simplify the calculation to a certain extent and more accurately reflect the relationship between queue waiting time and the number of charging piles configured, user arrival rate, and service rate.

[0057] In some embodiments, the construction cost of a charging station includes the land acquisition cost and the charging pile procurement and installation cost. This embodiment specifically defines the cost components considered when optimizing resource allocation. The land acquisition cost is the cost of the land required to build a charging station. It can be the cost of directly purchasing the land or the cost of leasing the land. The cost depends on factors such as the location and area. The charging pile procurement and installation cost is the cost required to purchase and install the charging pile. The cost of a single pile depends on factors such as the type of charging pile (such as AC slow charging, DC fast charging), power size, brand, etc. The installation cost includes labor costs, material costs, etc. When optimizing the resource allocation of charging stations, it is necessary to comprehensively consider these cost factors, as well as service quality costs such as user queuing losses, to achieve the optimization of the overall benefits. Construction costs can also include the purchase cost of power equipment such as transformers. By considering construction costs more comprehensively, resource allocation plans can be made more economical and feasible.

[0058] In some embodiments, determining the goals of the incremental configuration number and the initial configuration number of charging piles further includes satisfying a preset grid power supply capacity constraint. This embodiment adds constraints to the optimization target. The grid power supply capacity constraint refers to the maximum power that the grid in the area where the charging station is located can provide. The total power demand of the charging piles cannot exceed this limit, otherwise it may cause grid overload and affect power supply stability. The grid power supply capacity can be obtained by consulting the local power company or consulting relevant information. When optimizing resource allocation, it is necessary to comply with the grid power supply capacity constraints while meeting the user's charging needs to ensure the safe and stable operation of the charging facilities. This embodiment can avoid grid safety issues caused by excessive construction of charging facilities and improve the sustainability of charging infrastructure.

[0059] In some embodiments, the weight parameters of the indicators are determined using an expert scoring method or an entropy weighting method. This embodiment defines the method for determining weight parameters in multi-indicator decision-making screening. Multi-indicator decision-making screening aims to comprehensively evaluate the advantages and disadvantages of each candidate charging station and select the optimal option. Different indicators (such as composite attractiveness, operating costs, and user queuing delay costs) have different importance in the decision-making process and require corresponding weight parameters. The expert scoring method involves inviting experts in relevant fields to score the importance of each indicator based on their experience and knowledge. The scores are then averaged or weighted to determine the final weight parameters. The entropy weighting method is an objective weighting method. Its basic concept is that the greater the variability of an indicator, the greater the amount of information it can provide, and therefore the greater its weight should be. The specific steps include: data standardization, calculating the information entropy of the indicator, and calculating the weight of the indicator. The expert scoring method can incorporate the expert's experience and knowledge, but may be affected by subjective factors. The entropy weighting method is based on the data itself and is more objective, but may ignore the expert's experience and knowledge. Using the expert scoring method or the entropy weighting method to determine the weight parameters can ensure the scientific and rationality of the decision-making.

[0060] In some embodiments, the method further includes a rolling planning step: updating the construction site information recommended in step S500 to the geographical location and number of charging piles included in the existing charging station distribution and attribute information in S100, and using this as new input data, repeating steps S100 to S500 to perform dynamic optimization planning. This embodiment adds a rolling planning step to solve the static problem of charging pile planning schemes in the prior art, so that the planning scheme can adapt to changes in the urban environment and the evolution of user needs. Rolling planning is a dynamic iterative planning method. Its core idea is to decompose the planning process into multiple continuous planning cycles. After each cycle, the planning results are fed back to the planning of the next cycle, thereby achieving continuous learning and optimization of the planning. In the present invention, the specific implementation method of the rolling planning step is: first, execute steps S100 to S500 to obtain the construction site information (including geographical location and planned capacity) recommended in this planning cycle, and then update this information to the existing charging station distribution and attribute information in S100, and use this as the new initial data. At the beginning of the next planning cycle, repeat steps S100 to S500 to achieve dynamic adjustment and adaptation of the planning scheme. The planning cycle can be set according to actual needs, such as updating the plan once a quarter, once a year, or every two years. Through the rolling planning step, the planning scheme can be self-adjusted and optimized according to actual operating data and market changes, maintaining the long-term effectiveness and foresight of the planning scheme.

[0061] In some embodiments, road network connectivity information further includes travel time estimates for connections between geographic blocks. These travel time estimates are calculated by weighting a base weight determined by historical traffic data, an adjustment factor based on real-time traffic conditions, and an event-driven adjustment factor based on special event information. This embodiment limits the specific implementation of road network connectivity information and aims to improve the accuracy of travel time estimates. Travel time is a key factor influencing user selection of charging stations, and accurate travel time estimates can improve the accuracy of user selection predictions. Historical traffic data reflects normal traffic conditions, such as average vehicle speeds or V / C ratios at different times of day and days of the week, and can serve as the base weight for travel time estimation. Real-time traffic conditions reflect current traffic conditions, such as a city's real-time congestion index and road closure information, and can serve as a dynamic adjustment factor to adjust the base weight. Special event information reflects unexpected events that may impact traffic, such as large-scale events and road construction, and can serve as an event-driven adjustment factor to correct travel time. This weighting calculation can be performed using various methods, such as linear weighting, nonlinear weighting, or piecewise weighting. By combining historical data, real-time traffic conditions, and special event information, travel time can be estimated more accurately, improving the accuracy of user selection predictions and thus increasing the rationality of charging resource planning.

[0062] In some embodiments, the objective of minimizing combined costs further includes the external congestion costs that may be caused by charging station operations. These costs are quantified based on the level of congestion in the area where the charging station is located and the estimated user traffic volume. This embodiment expands upon the optimization objective to more comprehensively consider the impact of charging station operations on urban traffic. External congestion costs refer to the social costs incurred by additional traffic congestion caused by users traveling to charging stations, such as increased delays for other vehicles, increased fuel consumption, and increased exhaust emissions. Quantifying these costs requires considering the level of congestion in the area where the charging station is located and the estimated user traffic volume. Congestion levels can be measured using metrics such as the V / C ratio and average travel time. Estimated user traffic volume can be calculated using a user-selected prediction model and total charging demand. External congestion costs can be quantified in various ways, such as by multiplying the congestion level by user traffic volume to obtain congested mileage and then calculating the congestion cost per mile. Alternatively, additional queue time caused by the charging station can be calculated using a queuing theory model and then multiplied by the time value. By incorporating external traffic congestion costs into the optimization objectives, planning schemes can be guided to avoid building charging stations in traffic-sensitive areas, thereby reducing the impact on urban traffic and improving urban operation efficiency.

[0063] like Figure 2 As shown, an embodiment of the present invention provides an electric vehicle charging resource planning and optimization system, which aims to achieve scientific site selection and capacity configuration of electric vehicle charging resources, effectively alleviate urban traffic congestion, and promote balanced distribution of charging flow. The system includes a data acquisition module M100, an attractiveness evaluation module M200, a user selection prediction module M300, a resource allocation optimization module M400 and a decision screening module M500. These modules work together to achieve comprehensive planning and optimization of charging resources.

[0064] Data acquisition module M100 is used to obtain geographic block information, road network connectivity information, POI distribution data, and the location and number of existing charging stations within the target planning area. Geographic block information is the basic unit for dividing the planning area, road network connectivity information is used to estimate the travel time for users to reach charging stations, and POI distribution data reflects the distribution of different types of venues. These data are the basis for attractiveness assessment.

[0065] The attractiveness assessment module M200 quantifies the composite attractiveness of each existing charging station and potential candidate. This composite attractiveness is calculated using a preset weighting function based on the point of interest (POI) value of the charging station's geographic area, the land cost, the estimated proportion of available charging piles at the charging station, and a reputation parameter determined based on the charging station's historical service quality data. By comprehensively considering multiple factors, the attractiveness assessment module M200 provides a comprehensive assessment of a charging station's attractiveness to users, providing a basis for predicting user choices.

[0066] The User Selection Prediction Module (M300) predicts the probability of users selecting each charging station within the planning area based on their geographic distance to each station and the station's combined attractiveness, combined with a preset distance attenuation factor. This provides a basis for resource allocation. The User Selection Prediction Module (M300) bridges the gap between attractiveness assessment and resource allocation, simulating user behavior to allocate charging demand to various charging stations.

[0067] Resource Allocation Optimization Module M400 calculates the charging demand allocated to each charging station based on the predicted probability of user selection and the total charging demand within the planning period. Based on the allocated charging demand, and with the goal of minimizing the combined cost of charging station construction and average user waiting time, it determines the incremental number of charging piles to be allocated to each existing charging station and the initial number of charging piles to be allocated to each potential candidate charging station. This module optimizes the number of charging piles based on the allocated demand, taking into account both economic benefits and user experience, effectively avoiding resource waste and user queues.

[0068] The decision-screening module, M500, is used to extract metrics such as the composite attractiveness, construction cost, and estimated average user waiting time for all potential candidate charging stations. After normalizing these metrics, the normalized metrics are weighted and aggregated according to preset weights to obtain a comprehensive evaluation value for each candidate charging station. All candidate charging stations are ranked based on the comprehensive evaluation value, and a predetermined number of charging stations ranked high are selected as recommended construction sites. This module uses multi-metric decision-making to evaluate candidate sites from multiple perspectives and select the optimal option. Working together, these modules enable the scientific planning and optimization of electric vehicle charging resources, thereby improving the utilization efficiency of charging facilities, enhancing the user charging experience, and alleviating urban traffic congestion.

[0069] In some embodiments, the attractiveness evaluation module M200 is configured to, when calculating the composite attractiveness, modify the traffic flow to road capacity ratio (V / C ratio) of the geographic area where the charging station is located or the roads connected to the charging station. As the V / C ratio increases, the composite attractiveness decreases. This embodiment limits the functionality of the attractiveness evaluation module M200 and aims to address the problem that existing charging station planning schemes are prone to causing traffic congestion. The attractiveness evaluation module M200 is specifically configured to: obtain the traffic flow to road capacity ratio (V / C ratio) of the geographic area where the charging station is located or the roads connected to the charging station from the data acquisition module M100. A higher V / C ratio indicates greater traffic congestion in the area or road; modify the V / C ratio when calculating the composite attractiveness; as the V / C ratio increases, the composite attractiveness decreases. The reduction can be linear, nonlinear, or stepwise; and the modified composite attractiveness is used as the user's attractiveness evaluation value for the charging station, which is then used by the user selection prediction module M300. This embodiment can effectively avoid the excessive concentration of charging stations in traffic congestion areas, thereby alleviating traffic pressure and improving urban operation efficiency.

[0070] In some embodiments, a reputation parameter is iteratively updated based on the historical deviation between the unit charging prices published by the charging station over past periods and the unit charging prices actually paid by users. This embodiment defines the reputation parameter used by the attractiveness evaluation module M200 when calculating the composite attractiveness. The attractiveness evaluation module M200 quantifies the attractiveness of a charging station by comprehensively considering multiple factors, with the authenticity and reliability of service information being key factors influencing user selection. This reputation parameter is used to quantify the reliability of service information. Specifically, the attractiveness evaluation module M200 first obtains the unit charging prices published by the charging station over past periods and the unit charging prices actually paid by users from the data acquisition module M100. It then calculates the deviation between the two and dynamically adjusts the reputation parameter based on the historical deviation using a pre-set iterative update mechanism. Various algorithms can be used to iteratively update the reputation parameter, such as exponential smoothing and weighted average of historical data. This dynamic reputation parameter update mechanism encourages charging station operators to provide authentic and reliable service information, increasing user trust and thus enhancing the user experience.

[0071] In some embodiments, the resource allocation optimization module M400 is further configured to satisfy preset grid power capacity constraints when determining the incremental and initial number of charging piles. This embodiment limits the functionality of the resource allocation optimization module M400 to ensure the safety and feasibility of charging infrastructure construction. The resource allocation optimization module M400 calculates the charging demand allocated to each charging station based on the user selection probability and total charging demand output by the user selection prediction module M300. It also determines the incremental number of charging piles for each existing charging station and the initial number of charging piles for each potential candidate charging station. In addition to meeting user charging demand and minimizing the combined cost of construction costs and average user waiting time, the resource allocation optimization module M400 also considers grid power capacity constraints. Specifically, this module is configured to: obtain grid power capacity data from the data acquisition module M100; ensure that the total power demand of the charging piles at all charging stations does not exceed the grid power capacity limit when determining the number of charging piles; and adjust the charging pile configuration plan if these constraints are not met, such as reducing the number of charging piles at some stations or adjusting the site selection. By considering the power supply capacity constraints of the power grid, grid security issues caused by excessive construction of charging facilities can be avoided and the sustainability of charging infrastructure can be improved.

[0072] The above content is a further detailed description of the present application in conjunction with specific preferred embodiments, and the specific implementation of the present application cannot be considered to be limited to these descriptions. For ordinary technicians in the technical field to which the present application belongs, several simple deductions or substitutions can be made without departing from the concept of the present application, and all of them should be considered to fall within the scope of protection of the present application.

Claims

1. A method for planning and optimizing electric vehicle charging resources, characterized in that: The following steps are involved: Obtaining geographic block information within the target planning area, road network connectivity information between the blocks, POI distribution data within each block, and the geographic location and number of existing charging stations; For each existing charging station and potential candidate charging station, the composite attractiveness is quantified and calculated using a preset weighting function based on a charging station attractiveness evaluation model based on the POI value of the geographical area where the charging station is located, the land cost, the estimated proportion of available charging piles at the charging station, and a reputation parameter determined based on the charging station's historical service quality data. For users within the planning area, the probability of the user selecting each charging station is predicted based on the geographical distance from the user to each charging station, the composite attractiveness of the charging station, and a preset distance attenuation factor; Calculating the charging demand allocated to each charging station based on the predicted probability of the user selecting each charging station and the total charging demand within the planning period; determining the incremental number of charging piles to be configured at each existing charging station and the initial number of charging piles to be configured at each potential candidate charging station based on the allocated charging demand and with the goal of minimizing the combined cost of charging station construction cost and average user waiting time; For all potential candidate charging stations, extract the composite attractiveness, construction cost, and estimated average user waiting time as screening indicators; after normalizing the screening indicators, perform weighted aggregation on the normalized screening indicators according to preset weights to obtain a comprehensive evaluation value for each candidate charging station; All candidate charging stations are ranked according to the comprehensive evaluation values, and a predetermined number of charging stations ranked high are selected as recommended construction sites.

2. The electric vehicle charging resource planning and optimization method according to claim 1, characterized in that: The composite attractiveness is further corrected during calculation based on the traffic flow to road capacity ratio of the geographical area where the charging station is located or the road connected to the charging station. When the traffic flow to road capacity ratio increases, the composite attractiveness decreases.

3. The electric vehicle charging resource planning and optimization method according to claim 1, characterized in that: The distance decay factor is configured to be dynamically adjusted according to an average traffic congestion level of the road network; as the average traffic congestion level increases, the impact of the distance decay factor increases.

4. The electric vehicle charging resource planning and optimization method according to claim 1, characterized in that: The average waiting time of users in queues is configured to adopt the M / M / c queuing theory model and is calculated according to the number of charging piles configured at the charging station, the user arrival rate, and the unit charging service rate.

5. The electric vehicle charging resource planning and optimization method according to claim 1, characterized in that: The goal of determining the incremental configuration quantity and the initial configuration quantity of charging piles further includes meeting a preset power grid power supply capacity constraint.

6. The electric vehicle charging resource planning and optimization method according to claim 1, characterized in that: The weight parameters of the indicators are determined by expert scoring method or entropy weight method.

7. The electric vehicle charging resource planning and optimization method according to claim 1, characterized in that: The method further includes a rolling planning step: updating the recommended construction site information to the geographical location and number of charging piles of the existing charging station, and using this as new input data, repeatedly executing the method to perform dynamic optimization planning.

8. The electric vehicle charging resource planning and optimization method according to claim 1, characterized in that: The road network connection information further includes estimated travel time values ​​for connections between geographic blocks, which are obtained by weighted calculation based on basic weights determined by historical traffic data, adjustment factors for real-time traffic conditions, and event-driven adjustment factors for special event information.

9. The electric vehicle charging resource planning and optimization method according to claim 1, characterized in that: The goal of minimizing the combined cost further includes external traffic congestion costs that may be caused by the operation of the charging station, and the external traffic congestion costs are quantified based on the traffic congestion level in the area where the charging station is located and the estimated user flow.

10. An electric vehicle charging resource planning and optimization system, characterized in that: A method for planning and optimizing electric vehicle charging resources according to any one of claims 1 to 9, comprising: a data acquisition module configured to acquire geographic block information within the target planning area, road network connectivity information between the blocks, POI distribution data within each block, and the geographic location and number of existing charging stations; An attractiveness evaluation module is configured to quantify the composite attractiveness of each existing charging station and potential candidate charging station, where the composite attractiveness is calculated using a preset weighting function based on the POI value of the geographical area where the charging station is located, the land cost, the estimated ratio of available charging piles at the charging station, and a reputation parameter determined based on the historical service quality data of the charging station; A user selection prediction module is configured to predict, for users within the planning area, the probability of the user selecting each charging station based on the geographical distance from the user to each charging station and the composite attractiveness of the charging station, combined with a preset distance attenuation factor; a resource allocation optimization module configured to calculate the charging demand allocated to each charging station based on the predicted probability of the user selecting each charging station and the total charging demand within the planning period; based on the allocated charging demand, and with the goal of minimizing the combined cost of charging station construction cost and user average waiting time in queues, determine the incremental number of charging piles to be configured at each existing charging station and the initial number of charging piles to be configured at each potential candidate charging station; The decision-making screening module is configured to extract the composite attractiveness, construction cost, and estimated average waiting time of users for all potential candidate charging stations; after normalizing the indicators, perform weighted aggregation on the normalized indicators according to preset weights to obtain a comprehensive evaluation value for each candidate charging station; rank all candidate charging stations according to the comprehensive evaluation value, and select a predetermined number of charging stations with the highest ranking as recommended construction sites.

Images