Charging pile metering and billing optimization control method based on intelligent city management platform

Abstract

The invention discloses a charging pile metering and billing optimization control method based on an intelligent city management platform, and relates to the technical field of billing optimization, and the method comprises the steps: recognizing the delayable charging capability of each charging request, and dividing the requests into high-elasticity schedulable requests and low-elasticity supportability requests according to the capability; calculating the total schedulable power which can be provided by all the high-elasticity schedulable requests in the target adjustment time period; the total schedulable power is compared with the adjustment demand power of the power grid, the technical problem that the deviation between the declaration duration and the actual duration is too large in charging request elastic classification is solved, and the classification precision is improved.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of charging optimization, and specifically relates to a charging pile metering and charging optimization control method based on an intelligent urban management platform BACKGROUND

[0002] The rapid growth of the number of new energy vehicles has driven the continuous expansion of the deployment scale of charging piles. The load impact and metering and charging rationality after the charging piles are connected to the power grid have become key problems to be solved in the field of urban energy management.

[0003] The current charging pile scheduling mostly adopts a rough mode of first-come-first-serve, without fully considering the time flexibility difference of different charging requests, which causes the overload of the power grid during the peak period and the idling of resources during the trough period, directly affecting the stability of power supply.

[0004] The traditional charging method is mostly based on fixed electricity prices or simple time-of-use electricity prices, without effectively associating with the scheduling response capability, which cannot encourage users to charge off-peak and is difficult to balance the power grid regulation cost and user charging demand.

[0005] In the prior art, some scheduling schemes attempt to introduce load forecasting and priority sorting, but lack precise quantification of the delayable charging capability of the charging request, and the scheduling flexibility is insufficient. The charging optimization mostly focuses on electricity price adjustment, without building a dynamic linkage mechanism of scheduling response and charging rules, and it is difficult to realize the win-win of power grid regulation and user benefits.

[0006] In addition, most schemes do not consider the dynamic changes of real-time regulation demand of the power grid, and only allocate charging resources through static rules, which is difficult to adapt to complex power grid conditions, leading to imbalance between metering and charging fairness and power grid operation economy, and cannot meet the comprehensive needs of precise scheduling and efficient charging of the intelligent urban management platform. SUMMARY

[0007] The purpose of the present application is to provide a charging pile metering and charging optimization control method based on an intelligent urban management platform to solve the problems mentioned in the background.

[0008] The charging pile metering and charging optimization control method based on the intelligent urban management platform is realized through the following collaborative scheduling process: The delayable charging capability of each charging request is identified, and the requests are divided into high-flexibility schedulable requests and low-flexibility guaranteed requests according to the capability; It should be noted that this step is the basis of the entire collaborative scheduling, and the core is to realize the precise quantification of the delayable charging capability of the charging request through standardized technical means, providing a reliable classification basis for subsequent scheduling.

[0009] The total schedulable power that all high-flexibility schedulable requests can provide within the target regulation period is calculated; Adjustable power: refers to the adjustment power that the charging request can provide for the power grid within the target adjustment period, and the calculation formula is: adjustable power = rated power of charging pile x adjustable duration ratio x power adjustment coefficient; Comparing the total adjustable power with the adjustment demand power of the power grid; It should be emphasized that the core of the comparison step is to establish the matching relationship between the flexible resource and the demand of the power grid, and to provide a clear basis for the selection of subsequent scheduling strategies.

[0010] Based on the comparison result, perform scheduling: when the total adjustable power is greater than or equal to the adjustment demand power, schedule the selected high-elasticity adjustable request; when the total adjustable power is less than the adjustment demand power, on the basis of performing scheduling on the high-elasticity adjustable request, supplement part of the low-elasticity guarantee request according to the pre-set rule; The core of this step is to perform differentiated scheduling strategies according to different power matching results; According to the scheduling execution result, dynamic charging is carried out.

[0011] In some possible embodiments, the delayable charging capacity of each charging request is identified, and the request is divided according to the capacity, including: Obtain the remaining battery capacity, target charging capacity and planned stay duration of the charging request; According to the remaining battery capacity and the target charging capacity, calculate the total rigid demand required to complete the charging; According to the total rigid demand and the planned stay duration, calculate the maximum flexible duration that the charging start time can be allowed to delay; By comparing the relationship between the maximum flexible duration and the planned stay duration, the charging request is divided into high-elasticity adjustable request and low-elasticity guarantee request.

[0012] It should be noted that the core of the method for obtaining the planned stay duration is to introduce a statistical analysis mechanism of user historical charging behavior, to solve the problem that the planned stay duration is greatly deviated from the actual situation in the traditional method which only relies on the user's this time report, and to further improve the accuracy of the planned stay duration, and to provide more reliable basic data for the subsequent calculation of the maximum flexible duration and the classification of the charging request.

[0013] In some possible embodiments, after calculating the maximum flexible duration that the charging start time can be allowed to delay, it further includes: Based on the planned stay duration and the maximum flexible duration, calculate the latest must start charging time of the charging request within the planned stay duration; Cross-comparing the latest must start charging time with the current time and the congestion period predicted by the power grid; If the latest start time of charging falls into the congested period predicted by the power grid, the maximum flexible duration of the request is adjusted.

[0014] In some possible embodiments, the scheduling is performed based on the comparison result, and specifically includes: When the total schedulable power is greater than or equal to the regulation demand power, the maximum flexible duration of each high-flexibility schedulable request is sorted, and the request with a longer maximum flexible duration is preferentially scheduled; When the total schedulable power is less than the regulation demand power, hierarchical supplementary scheduling is performed: all the high-flexibility schedulable requests that can be responded to are included in the scheduling, the power gap is calculated, and the low-flexibility guaranteed requests are sorted according to the charging urgency determined by the remaining battery power of the low-flexibility guaranteed requests, and the supplementary scheduling is performed from the request with the lowest urgency to fill the power gap.

[0015] In some possible embodiments, the scheduling of the low-flexibility guaranteed requests in the hierarchical supplementary scheduling is implemented through a progressive power control including state verification: An initial charging power threshold lower than the rated demand power of the selected low-flexibility guaranteed request is set; The initial charging power threshold is used to perform charging, and the vehicle connection state is continuously monitored; After the vehicle connection state is monitored to be continuously stable for more than a preset verification duration, the charging power is increased to a target power level required by the scheduling scheme; If the vehicle connection is monitored to be disconnected before the verification duration is reached, the request is removed from the current scheduling task, and a scheduling object is reselected from the standby scheduling resource based on the power gap.

[0016] The design of the progressive power control strategy is to solve the scheduling failure problem of the low-flexibility guaranteed request in the supplementary scheduling process due to unstable vehicle connection, while avoiding the impact of instantaneous high-power charging on the power grid.

[0017] In some possible embodiments, the dynamic charging based on the scheduling execution result is executed in linkage with the following operations for maintaining stable scheduling effect: The actual charging power curve is monitored and compared with the scheduling scheme in real time; When the actual charging behavior deviates from the scheduling scheme and generates a regulation power gap, one or more standby requests are selected and started for charging from the standby request queue based on the regulation power gap and the pre-calculated schedulable power of the standby request; The charging parameters of the related parties are updated according to the deviation behavior and the starting execution of the standby request.

[0018] In some possible embodiments, the method further includes: extracting a historical planned stay duration and a corresponding historical actual stay duration in a historical charging record of the user; calculating a deviation value of the historical planned stay duration and the historical actual stay duration; when the deviation value exceeds a preset threshold, replacing the planned stay duration of the current charging request with a statistical value based on the historical actual stay duration, and then performing the step of identifying the delayable charging capability.

[0019] In some possible implementation manners, the requested planned stay duration is obtained, specifically: when the number of historical charging records of the user reaches a statistical requirement, reading an average value or a median of the historical actual stay duration; performing weighted fusion of the average value or the median of the historical actual stay duration and the planned stay duration declared by the user this time to generate a corrected planned stay duration; using the corrected planned stay duration, performing the step of calculating the maximum flexible duration.

[0020] In some possible implementation manners, before the step of ranking according to the charging urgency, the method further includes: obtaining historical charging records of the low-flexibility guarantee request user under similar initial power and similar time period; when the historical records show that the actual stay duration of the user is generally shorter than the planned stay duration declared by the user, increasing the charging urgency of the current request of the user by one level in ranking.

[0021] In some possible implementation manners, the step of selecting a backup request from the backup request queue includes: calculating a number of active scheduling participations and a number of passive backup activations of each backup request user in historical records; lowering the ranking position of the backup request of a user in the queue when a ratio of the number of active scheduling participations to the number of passive backup activations of the user is lower than a set value; performing the operation of selecting and starting the backup request according to the lowered ranking.

[0022] Specifically, the core of the design of the selection step is to introduce an evaluation mechanism of historical scheduling behaviors of users, to realize dynamic optimization of ranking of the backup request queue, to preferentially select a user actively cooperating with scheduling as a backup request, to further encourage the active scheduling participation behavior of the user, and to improve the overall efficiency and sustainability of scheduling.

[0023] The technical solutions provided by the embodiments of the present disclosure at least bring the following beneficial effects: The technical problem of too large deviation between the declared length and the actual length in the charging request elastic classification is solved by fusing the historical data of the planned length of stay with the declared value, and the classification accuracy is improved; and the scheduling failure problem caused by the disconnection of the low elasticity request is effectively avoided by relying on the progressive power control strategy of the low elasticity supplementary scheduling, and the reliable scheduling of the low elasticity request is realized. BRIEF DESCRIPTION OF DRAWINGS

[0024] Figure 1 The method structure of the application is shown in the figure. DETAILED DESCRIPTION

[0025] The technical solutions of the application will be described clearly and completely in combination with the embodiments. Obviously, the described embodiments are only some of the embodiments of the application, not all. Based on the embodiments in the application, all other embodiments obtained by those skilled in the art without creative labor are within the protection scope of the application.

[0026] The core principle of the application is to correct the single declaration information of the user by fusing the historical charging behavior data of the user, so as to more accurately quantify the time elasticity of the charging request; and then dynamically match the elasticity classification with the grid regulation demand, and use the billing mechanism to encourage the user behavior to cooperate with the scheduling target. The specific parameters (such as weight, threshold, length, etc.) described below are all example values adopted in a specific experiment or application scenario to realize the above principle.

[0027] Those skilled in the art should understand that, without departing from the above core principle of the application, these parameters can be adjusted through conventional experiments or optimization according to the specific system performance, data quality, grid requirements and operation strategy, and these adjustments should all fall within the protection scope of the application.

[0028] The intelligent urban management platform based on the application adopts a micro-service architecture, including a charging request receiving module, an elasticity evaluation module, a scheduling engine module, a billing module and a real-time monitoring module. The modules communicate with each other through RESTful API, and are connected to the grid scheduling center, the charging pile terminal and the user APP through HTTPS / GB / T27930 protocol.

[0029] Please refer to Figure 1 The application provides a charging pile metering and billing optimization control method based on an intelligent urban management platform, which comprises the following steps: Step 1: identifying the delayable charging capacity of each charging request, and dividing the requests according to the capacity, the specific process of the step being: obtaining the remaining battery capacity, the target charging capacity and the planned length of stay of the charging request; According to the battery remaining power and the target charging power, the total rigid demand power required for completing charging is calculated; According to the total rigid demand power and the planned stay duration, the maximum flexible duration that the charging start time can be delayed is calculated; According to the total rigid demand power and the rated power of the charging pile, the theoretical required charging duration is calculated. Then, according to the planned stay duration and the theoretical required charging duration, the maximum flexible duration is calculated. The calculation formula is: Theoretical required charging duration = total rigid demand power / rated power of charging pile Maximum flexible duration = planned stay duration - theoretical required charging duration If the calculation result is negative, the maximum flexible duration is 0.

[0030] By comparing the ratio of the maximum flexible duration to the planned stay duration (flexible ratio) with a preset threshold, requests with a flexible ratio less than the preset threshold are classified as low-flexibility guaranteed requests, and requests with a flexible ratio greater than or equal to the preset threshold are classified as high-flexibility dispatchable requests; As an implementation, the preset threshold can be set to 0.5 (i.e. 50%), which can be dynamically adjusted according to historical dispatching results; Among them, the planned stay duration of the request is obtained, specifically, the execution process is: when the number of historical charging records of the request user reaches the statistical requirement, the average value or median of the historical actual stay duration is read; The average value or median of the historical actual stay duration is weighted and fused with the planned stay duration reported by the user this time to generate a corrected planned stay duration; The corrected planned stay duration is used to perform the step of calculating the maximum flexible duration; In this embodiment, the platform ensures the accuracy and timeliness of the operation through data synchronization.

[0031] To obtain a more reliable planned stay duration, the historical charging behavior data of the user is used to correct the reported value this time.

[0032] When the number of historical records of the user reaches the requirement (such as no less than 5), the statistical value (such as the average value or the median) of the historical actual stay duration is weighted and fused with the reported value this time to obtain the corrected planned stay duration. For example, the weight of the historical statistical value can be between 0.6 and 0.8, and the weight of the reported value this time can be between 0.2 and 0.4. The weight ratio can be optimized according to user behavior data analysis.

[0033] If the number of historical records of the user is insufficient, the planned stay duration reported by the user this time is directly used.

[0034] In practical applications, intelligent urban management platforms obtain core parameters through multi-source data interaction: Remaining battery power: The charging pile reads this information in real time through its communication interface with the electric vehicle (following standard protocols such as GB / T27930) and uploads it to the platform.

[0035] Target charging capacity and planned stay duration: These are set by the user when submitting a charging request through the charging station's human-machine interface or a dedicated mobile application (APP). The APP sends the request data (including user ID, target charging capacity, planned stay duration, etc.) to the platform server via network communication methods such as HTTPS / API.

[0036] Data fusion: The platform associates and merges information from the vehicle (battery data) and the user (request data) based on the user ID and the charging pile ID to form a complete charging request data packet.

[0037] After obtaining the above parameters, the platform performs calculations and classifications. The total rigid demand for electricity is the basis for distinguishing elastic boundaries. The maximum elastic duration reflects the request's adaptability to grid regulation. By comparing the maximum elastic duration with the planned dwell time, accurate classification of charging requests is achieved.

[0038] In this embodiment, the intelligent city management platform is equipped with edge computing nodes, each with a processing capacity of no less than 8 cores and 16GB of memory. It supports real-time analysis of more than 1,000 concurrent charging requests, and the calculation latency of a single request does not exceed 100 milliseconds, ensuring the real-time performance and accuracy of the classification results.

[0039] Step 2: Calculate the total schedulable power that all highly elastic schedulable requests can provide during the target adjustment period; The target adjustment period is dynamically set by the power grid dispatch center based on the regional load forecast results and synchronized in real time to the intelligent urban management platform via a dedicated line. It is usually divided into peak adjustment period, off-peak adjustment period and flat adjustment period. Each period lasts 1-4 hours and the adjustment interval is no less than 30 minutes to avoid frequent changes that could cause dispatch chaos.

[0040] After receiving the time period information, the platform automatically marks the adjustment direction for that time period, including negative adjustment and positive adjustment. Negative adjustment means that the power grid needs to reduce the load, while positive adjustment means that the power grid needs to increase the load.

[0041] For each highly flexible and schedulable request, the platform calculates its contributing power within the target adjustment period. The calculation logic is that the contributing power equals the charging pile's rated power multiplied by the proportion of schedulable time multiplied by the power adjustment coefficient.

[0042] The schedulable time percentage is the overlap between the target adjustment period and the delayable time range of the request, divided by the total duration of the target adjustment period, and is automatically calculated by the platform's time axis algorithm. The calculation formula is: Scheduleable duration percentage = MAX(0, MIN(Request delayable end time, Adjustment period end time) - MAX(Request delayable start time, Adjustment period start time)) / (Adjustment period end time - Adjustment period start time) The request delay start time is the request arrival time, and the request delay end time is the request arrival time plus its maximum elastic duration.

[0043] The power regulation coefficient characterizes the ratio of the actual safe output power of a charging pile to its rated power under specific grid conditions. This coefficient can be dynamically set by the grid dispatch center based on real-time stability margins of the regional grid (such as frequency deviation, voltage deviation, line load rate, and other comprehensive indicators). For example, when the grid issues a load reduction instruction, the coefficient can be set to 0.7-0.9; otherwise, the default coefficient is 1.0. This setting aims to ensure that charging dispatching activities do not jeopardize the safe and stable operation of the grid.

[0044] When calculating the total schedulable power, the historical scheduling response data of users can be considered to improve the reliability of the scheduling plan. For example, users with high historical response rates can be assigned higher reliability weights, and vice versa.

[0045] Those skilled in the art can establish or adjust the mapping relationship between specific response rates and weights based on user behavior data obtained in actual operations.

[0046] From a practical perspective, this precise accounting process clarifies the available flexible resources for power grid regulation, which helps reduce the risk of mismatch between subsequent dispatch and actual demand.

[0047] Step 3: Compare the total dispatchable power with the power demand for grid regulation; Furthermore, the regulated power demand is calculated by the power grid dispatch center using a load forecasting system combined with real-time load data. The forecasting model employs an LSTM neural network, whose inputs include historical load, weather, date type, and charging pile connection status, and whose output is the regulated power demand for the target period. The model is trained on historical data, and the prediction error does not exceed ±5%.

[0048] Before making comparisons, the platform standardizes the total dispatchable power and the power required for regulation (e.g., by unifying units and time precision). To avoid frequent scheduling switches due to minor fluctuations or prediction errors, a power margin or a judgment threshold can be introduced. For example, it can be set that when the total dispatchable power reaches more than 90% of the power required for regulation, the demand is considered met.

[0049] It should be understood that if data is abnormal or communication is interrupted, the platform will enable data caching or degradation strategies to ensure the scheduling process continues to run.

[0050] Those skilled in the art can determine or adjust the specific size of the margin or threshold based on the fluctuation characteristics of the specific power grid and the response accuracy of the dispatching system.

[0051] Specifically, the intelligent city management platform and the power grid dispatch center use dedicated communication lines to ensure the real-time performance and stability of data transmission. In fact, this precise comparison process clarifies the matching status between flexible resources and power grid demand, providing a clear decision-making basis for subsequent differentiated dispatching and improving the targeting and effectiveness of dispatching measures.

[0052] Step 4: Perform scheduling based on the comparison results; The above can be divided into the following two parts: When the total schedulable power is greater than or equal to the adjustment demand power, the requests with longer maximum elasticity durations are sorted according to their maximum elasticity durations, and requests with longer maximum elasticity durations are scheduled first. In this embodiment, the platform scheduling engine uses an improved bubble sort algorithm to quickly sort the data and add them to the scheduling queue in sequence until the cumulative scheduling power meets the requirements.

[0053] During execution, the platform monitors the charging pile response status in real time and removes requests that do not respond within 2 seconds from the scheduling queue to ensure scheduling effectiveness.

[0054] When the total dispatchable power is less than the power required for regulation, a tiered supplementary scheduling is performed: first, all responsive, highly elastic dispatchable requests are included in the scheduling; then, the power gap is calculated; and the charging urgency is determined by the remaining battery capacity of each low-elasticity guaranteed request. Supplementary scheduling is then performed starting with the request with the lowest urgency to fill the power gap. If the total available power is insufficient, a tiered supplementary scheduling is performed: first, all responsive high elasticity requests are scheduled, the power gap is calculated, and then the low elasticity requests are sorted by their remaining battery power, with supplementary scheduling starting from the requests with the lowest urgency. In this embodiment, the design of the hierarchical supplementary scheduling strategy fully reflects the priority principle of power grid regulation, while maximizing the protection of users' rigid charging needs.

[0055] The first phase of full-scale, highly elastic request scheduling does not simply involve including all requests; rather, it undergoes rigorous responsiveness screening. The responsiveness criteria include three core dimensions: the real-time operating status of the charging piles, the validity of the user's charging request, and the stability of network communication.

[0056] The platform uses a real-time status monitoring module to filter each highly elastic request one by one, with the filtering time controlled within 5 milliseconds. After filtering, all responsive highly elastic requests are included in the scheduling queue, fully exploring the adjustment potential of flexible resources; The second phase of power gap calculation uses precise difference calculation logic, that is, the power gap equals the adjustment demand power minus the cumulative scheduling power of the full volume of high elasticity requests.

[0057] The calculation results are rounded to one decimal place to provide a clear target value for supplementary scheduling. Determining the urgency of charging is one of the core technical aspects of this strategy, as it is directly related to the remaining battery capacity for low-elasticity guaranteed requests.

[0058] The platform uses a built-in battery parameter model to automatically calculate the ratio of remaining battery power to total battery capacity, i.e., the remaining power percentage. A higher remaining power percentage indicates a lower urgency for the user to charge; conversely, a lower remaining power percentage indicates a higher urgency for the user to charge. By selecting to supplement the dispatch from the request with the lowest urgency, it is possible to maximize the protection of the rigid demand of users with high charging urgency while meeting the grid regulation requirements.

[0059] For the supplementary scheduling execution process of low-elasticity guarantee requests, this solution is implemented through progressive power control that includes state verification, specifically as follows: Set an initial charging power threshold lower than the rated power requirement for the selected low-elasticity guarantee request; perform charging with the initial charging power threshold and continuously monitor the vehicle's connectivity status; after the vehicle's connectivity status is continuously and stably monitored for more than a preset verification time, increase its charging power to the target power level required by the scheduling scheme. If a vehicle connection is detected to be disconnected before the verification time is reached, the request will be removed from the current scheduling task, and a new scheduling object will be selected from the backup scheduling resources based on the power gap.

[0060] In this embodiment, the initial charging power threshold is set to 30%-50% of the rated power, for example, 2.1kW for a 7kW AC charging pile and 30kW for a 60kW DC charging pile, which is dynamically adjusted according to the grid load. After charging is performed at the initial charging power threshold, the platform continuously collects the vehicle's connection status data through the real-time status monitoring module of the charging pile terminal.

[0061] The core parameters monitored include three dimensions: the locking status of the charging gun, the communication connection status between the vehicle and the charging pile, and the stability of the charging current and voltage. The monitoring frequency is once per second to ensure timely detection of any abnormalities in the vehicle connection.

[0062] Set a status verification duration to confirm that the vehicle connection has entered a stable charging state. This duration should be longer than the typical time required for the charging station and vehicle to complete the start-up handshake and initial parameter exchange. For example, 5 seconds can be set for DC charging stations and 10 seconds for AC charging stations to avoid impacting the power grid.

[0063] After detecting that the vehicle's connection status has remained stable for a period of time exceeding the preset verification time, the platform uses the power control module to gradually increase its charging power from the initial charging power threshold to the target power level required by the scheduling scheme.

[0064] The power increase process adopts a linear incremental approach, with an increase rate of 10% of the rated demand power per second, ensuring a smooth power increase and avoiding the impact of instantaneous high-power charging on the power grid. The target power level is set as the power value required by the dispatch scheme, that is, the power that this low-elasticity guarantee request needs to provide during the supplementary dispatch process, ensuring that the power gap can be accurately filled.

[0065] If a vehicle connection is detected to be lost before the verification time is reached, the platform will remove the request from the current scheduling task through the scheduling task management module and update the power gap data.

[0066] The updated power gap data is equal to the original power gap data minus the target power level requested, ensuring the accuracy of the power gap data.

[0067] Subsequently, based on the updated power gap data, the platform reselects scheduling targets from the backup scheduling resources. The backup scheduling resources consist of low-elasticity guaranteed requests that were not included in the initial supplementary scheduling queue, as well as newly added high-elasticity schedulable requests.

[0068] The rules for reselecting scheduling objects are consistent with the rules for initial supplementary scheduling, that is, according to the charging urgency of each standby scheduling resource, supplementary scheduling is carried out starting from the request with the lowest urgency, ensuring that the supplementary scheduling process complies with the priority principle of grid regulation.

[0069] During the supplementary scheduling process, the platform will monitor the cumulative supplementary power in real time, and will immediately stop supplementary scheduling when the power gap reaches 100%.

[0070] Meanwhile, to cope with sudden emergency charging demands, the platform reserves 10% of low-elasticity guaranteed requests as an emergency reserve. Even if the power gap is not fully filled, these requests will not be dispatched. This strategy helps improve the dispatch efficiency of high-elasticity requests and accurately meet the grid regulation needs.

[0071] All dispatch instructions are sent to the corresponding charging pile terminals via the platform's encrypted communication module. The instructions employ end-to-end encrypted transmission, ensuring full data security during the transmission process.

[0072] After receiving a dispatch command, the charging pile terminal must respond within 2 seconds and report the response status to the platform in real time. The platform's dispatch monitoring module will monitor the status of all charging piles that have received commands at 10-second intervals.

[0073] For charging piles that fail to follow instructions, the monitoring module will automatically send a second instruction; if the second instruction still does not receive a response, the monitoring module will include the charging pile in the fault early warning system and notify the maintenance personnel to handle it.

[0074] By implementing this differentiated scheduling strategy, the adjustment potential of highly elastic requests is fully utilized, while resource gaps are made up through low-elasticity supplementary scheduling. This helps to smooth the power grid load curve and reduce the risk of overload during peak hours.

[0075] Step 5: Perform dynamic billing based on the scheduling execution results; Specifically, dynamic billing is a core means of incentivizing users to cooperate with scheduling. To further ensure the stability of scheduling results, dynamic billing is performed based on the scheduling execution results and is executed in conjunction with the following operations to maintain the stability of scheduling results: Real-time monitoring and comparison of actual charging power curves with scheduling schemes; When it is detected that the actual charging behavior deviates from the scheduling plan and generates a power adjustment gap, based on the power adjustment gap and the pre-calculated power available for scheduling from the backup request queue, one or more backup requests are automatically selected and started for charging. Based on the deviation behavior and the initiation and execution of the backup request, the billing parameters of the relevant parties are updated.

[0076] It should be noted that the linkage execution strategy can solve the problem of adjustment effect failure caused by deviation of actual charging behavior in traditional scheduling schemes. At the same time, through the dynamic updating of billing parameters, positive guidance and constraints on user behavior can be achieved, further improving the stability and sustainability of scheduling.

[0077] It should be understood that the current joint execution strategy is implemented by the platform's scheduling effect stability control module. This module synchronizes data with the billing module, scheduling module, and real-time monitoring module in real time to ensure the real-time performance and accuracy of the joint operation.

[0078] Specifically, real-time monitoring and comparison of the actual charging power curve with the scheduling scheme is the foundation of this coordinated execution strategy. The platform collects actual charging power data of all charging piles included in the scheduling queue through a real-time monitoring module, at a frequency of once per second, to ensure accurate capture of dynamic changes in actual charging power.

[0079] The actual charging power curve is generated using a sliding window averaging algorithm with a window size of 5 seconds to ensure the smoothness and stability of the curve.

[0080] The scheduling scheme is obtained by retrieving the corresponding charging power curve data from the pre-generated scheduling scheme through the scheme storage unit of the scheduling module.

[0081] The comparison process uses a dynamic threshold comparison algorithm. The comparison threshold is set to ±5% of the charging power of the scheduling scheme. When the actual charging power curve exceeds the comparison threshold for 3 consecutive seconds, it is determined that the actual charging behavior deviates from the scheduling scheme.

[0082] When it is detected that the actual charging behavior deviates from the scheduling plan and generates a power adjustment gap, the system automatically selects and starts one or more backup requests from the backup request queue for charging based on the power adjustment gap and the pre-calculated backup request schedulable power. This is the core correction link of the linkage execution strategy.

[0083] Specifically, the calculation of the power gap is carried out using precise difference calculation, that is, the power gap is equal to the charging power of the scheduling plan minus the actual charging power. The calculation result is retained to one decimal place, providing a clear target value for the selection of standby requests.

[0084] The construction of the backup request queue is completed synchronously when the scheduling scheme is generated. The queue consists of highly elastic schedulable requests and low-elasticity guaranteed requests that were not included in the initial scheduling queue, and its number is 20% of the number of requests in the initial scheduling queue, ensuring the sufficiency of backup requests. The pre-calculation of the schedulable power of backup requests adopts the same calculation logic as in step 2 above, ensuring data consistency and accuracy.

[0085] The selection rule for backup requests is based on the matching degree between the adjustment power gap and the schedulable power of the backup requests, and is implemented using a greedy algorithm. The steps for selecting backup requests from the backup request queue include: calculating the number of times each backup request user has actively participated in scheduling and the number of times passive backup has been activated in historical records; users whose ratio of active scheduling participation to passive backup activation is lower than a set value have their backup requests moved down in the queue's ranking position. Based on the downgraded sorting, perform the operation of selecting and initiating the standby request.

[0086] In this embodiment, the entity executing the selection step is the platform's backup request selection module. This module synchronizes data in real time with the user history behavior recording module and the backup request queue management module to ensure the accuracy and fairness of the selection operation.

[0087] Calculating the number of times each standby requesting user actively participated in scheduling and the number of times passive standby was activated in the historical records is the fundamental step in this selection process.

[0088] The number of times active scheduling is participated in is defined as the number of times a user's request is included in the initial scheduling queue and successfully executed. This data is obtained through the active scheduling record unit of the user's historical behavior record module. The number of times passive standby is activated is defined as the number of times a user's request is included in the standby request queue and activated due to power adjustment gaps. This data is obtained through the passive standby record unit of the user's historical behavior record module. The calculation process employs a precise counting and statistical algorithm, with a statistical time range of the past 30 days to ensure the timeliness and representativeness of the data.

[0089] For users whose ratio of active scheduling participation to passive backup activation is lower than a set value, their backup requests will be relegated to a lower position in the queue. This is the core optimization step in this selection process. The ratio of active scheduling participation to passive backup activation is calculated using precise division logic, that is, the ratio equals the number of active scheduling participations divided by the number of passive backup activations, and the result is rounded to two decimal places.

[0090] The standard setting for the value is 1.5. This value is determined based on statistical analysis of users' historical scheduling behavior and is used to distinguish users' scheduling cooperation tendencies.

[0091] When a user's ratio is below 1.5, it indicates that the user has a relatively high number of passive backup activations and a relatively low number of active scheduling participations, and the user's backup request will be downgraded in the queue.

[0092] The rule for downgrading the ranking position adopts a fixed-level downgrading method, with each level decreasing by 5 positions to ensure the fairness and operability of the downgrading operation. For example, if a user's backup request is initially ranked 10th in the queue, and its ratio is lower than a set value, its ranking position will be downgraded to 15th. For users ranked at the end of the queue, if their ranking position is insufficient to be downgraded by 5 positions, their ranking position will be adjusted to the end of the queue to ensure the effectiveness of the downgrading operation.

[0093] The selection step involves selecting and initiating backup requests based on the adjusted ranking. The platform prioritizes backup requests with higher ranking positions to ensure that users who actively cooperate with scheduling are given priority.

[0094] The selection process is based on the matching degree between the regulated power gap and the schedulable power of the standby request. It adopts a greedy algorithm to prioritize the standby request whose schedulable power is closest to the regulated power gap. If the schedulable power of a single standby request cannot meet the needs of the regulated power gap, multiple standby requests are selected in turn until the cumulative schedulable power reaches 100%-120% of the regulated power gap, ensuring that the regulated power gap is fully compensated.

[0095] The initiation and execution of backup requests employs the same scheduling logic as in step 4. For highly elastic schedulable requests, requests with longer maximum elastic durations are prioritized for scheduling. For low-elasticity, guaranteed requests, a progressive power control strategy incorporating status verification is used to ensure that the initiation and execution of backup requests comply with scheduling priority principles and stability requirements. This selection step introduces an evaluation mechanism based on users' historical scheduling behavior, enabling dynamic optimization of the backup request queue. Prioritizing users who actively cooperate with scheduling as backup requests further incentivizes users' proactive scheduling participation, improving overall scheduling efficiency and sustainability.

[0096] Updating the billing parameters of relevant parties based on the deviation behavior and the initiation and execution of backup requests is the billing linkage link of this linkage execution strategy.

[0097] The definition of relevant parties includes users who exhibited the deviant behavior and users who initiated the backup request, ensuring the relevance and fairness of billing parameter updates.

[0098] The billing parameters are updated using differentiated update strategies based on the type of deviation behavior and the initiation and execution of standby requests.

[0099] For users who exhibit deviation behavior, if their actual charging power is lower than the charging power of the scheduling plan and causes a power adjustment gap, the base electricity price in their billing parameters will be increased by 5%-10%, with the specific increase ratio dynamically adjusted according to the degree of deviation. If its actual charging power is higher than the charging power of the dispatch plan and causes a power adjustment gap, the base electricity price in its billing parameters will be increased by 10%-15%, and the specific increase ratio will be dynamically adjusted according to the degree of deviation.

[0100] For users who initiate standby requests, the dispatch reward price in their billing parameters will be increased by one level. If they are already at the highest level, their dispatch reward price will be increased by 0.05 yuan / kWh based on the original level, ensuring positive incentives for standby request users.

[0101] Billing parameters are updated through the parameter update unit of the billing module, with real-time updates to ensure the accuracy and timeliness of billing. The updated billing parameters will serve as the basis for subsequent billing and will be incorporated into the calculation of total electricity costs, ensuring the effectiveness and sustainability of the coordinated execution strategy.

[0102] In this embodiment, the basic framework of the dynamic billing rule system remains unchanged, still including three parts: basic electricity price, dispatch reward electricity price, and emergency dispatch penalty electricity price. The basic electricity price is implemented with reference to the local power grid sales electricity price standard, which is 0.8 yuan / kWh during peak hours, 0.5 yuan / kWh during flat hours, and 0.3 yuan / kWh during off-peak hours. The dispatch incentive price applies only to highly flexible dispatchable requests and is calculated based on the dispatched electricity contribution. The incentive standard is divided into three tiers according to the priority of the dispatch response: Tier 1 is 0.15 yuan / kWh, Tier 2 is 0.10 yuan / kWh, and Tier 3 is 0.05 yuan / kWh. The emergency dispatch penalty price applies only to low-elasticity, guaranteed requests that are supplemented by dispatch during peak hours. The penalty standard is 0.1 yuan / kWh, which is used to balance the adjustment costs of the power grid during peak hours. No penalty fee is charged for low-elasticity requests that are supplemented by dispatch during off-peak hours.

[0103] The billing module automatically extracts key data from the scheduling execution list and the scheduling effect stability control module, including the charging duration of the scheduled request, the actual charging power, the scheduling type, the adjustment period, the deviation behavior type, and the status of the standby request activation. The total electricity cost is calculated using a formula, which equals the basic electricity cost plus the bonus electricity cost plus the penalty electricity cost plus the cost of updating and adjusting the billing parameters.

[0104] Among them, the basic electricity fee is equal to the actual charging amount multiplied by the basic electricity price of the corresponding time period; the bonus electricity fee is equal to the dispatch contribution amount multiplied by the bonus electricity price of the corresponding level. The dispatch contribution amount is the part of the actual charging amount that participates in grid regulation, that is, the charging amount within the dispatch period. The penalty electricity fee is equal to the emergency dispatch electricity volume multiplied by the penalty electricity price. The emergency dispatch electricity volume is the charging electricity volume during the period when low elasticity requests are supplemented by dispatch during peak hours. The cost of updating and adjusting billing parameters equals the cost of adjusting deviation behavior plus the cost of standby request incentives, ensuring the completeness and accuracy of billing.

[0105] The platform's billing module supports real-time rate calculation and bill generation, with a bill generation delay of no more than 5 minutes. The bill includes detailed information such as user ID, charging station number, charging time, actual charging volume, basic electricity cost, reward amount, penalty amount, billing parameter update and adjustment costs, and total cost, ensuring users clearly understand the breakdown of the charges.

[0106] The bill is pushed to the user via mobile app and simultaneously synchronized to the platform backend for the user to query and calculate.

[0107] Once a user completes payment, the bill status will be updated to "settled" in real time. For outstanding bills, the platform will send two reminders, one 24 hours and one hour before the due date, to ensure that fees are settled in a timely manner.

[0108] The billing module synchronizes data in real time with the scheduling module, charging pile terminals, and scheduling effect stability control module, ensuring that billing data is completely consistent with the actual scheduling execution and avoiding billing deviations. This coordinated execution strategy constructs a closed-loop scheduling control system, further improving the stability and accuracy of scheduling. Simultaneously, the dynamic updating of billing parameters provides positive guidance and constraints on user behavior, which helps improve the scheduling response rate for highly elastic requests and creates continuous incentives for flexible adjustment of the power grid load.

[0109] The beneficial effects of this invention are as follows: by integrating historical data of planned dwell time with declared values, it accurately solves the technical problem of excessive deviation between declared and actual dwell time in the elastic classification of charging requests, thereby improving classification accuracy; relying on the progressive power control strategy of low-elasticity supplementary scheduling, it effectively avoids scheduling failure caused by connection disconnection of low-elasticity requests, achieving reliable scheduling of low-elasticity requests; at the same time, through the synergy of the screening and sorting rules of the backup request queue and the differentiated dynamic billing scheme, it improves the execution efficiency of charging requests, reduces the idle rate of charging piles, effectively incentivizes users to participate in the flexible scheduling of the power grid, and alleviates the peak load pressure on the power grid.

[0110] As one embodiment of the present invention, after calculating the maximum flexible duration for which the charging start time can be delayed, the method further includes: calculating the latest time that the charging request must start charging within the planned stay duration based on the planned stay duration and the maximum flexible duration; Cross-compare the latest time when charging must begin with the current time and the congestion period predicted by the power grid. In this embodiment, the calculation logic for the latest time that charging must begin is the end time of the planned stay minus the theoretical charging time. The planned end time is the sum of the planned start time of the charging request and the planned end time. The theoretical charging time is the required electricity divided by the rated output power of the charging pile, and the result is rounded up to the minute level to adapt to the time granularity of power grid dispatch. The planned duration of stay is determined by a weighted fusion of historical charging records and the user's current declared value. The number of historical records is no less than 5, with a weight of 0.6 to 0.8 for historical records and a weight of 0.2 to 0.4 for the user's declared value.

[0111] Specifically, the weights are determined based on the charging behavior statistics of no less than 1,000 users in the region over one year. The deviation rate between the user-reported value and the actual dwell time is 35% to 45%, while the deviation rate of historical records is only 10% to 15%. Therefore, historical records are given higher weights to ensure accuracy. Retaining the weight of user-reported values ​​takes into account the special characteristics of the scenario. Simulation tests show that this range can keep the calculation deviation rate within 8%, which meets the scheduling accuracy requirements.

[0112] Furthermore, the congestion periods predicted by the power grid are generated by the LSTM load forecasting model, with a prediction cycle of 15 minutes. The criterion is that the actual load of the power grid is not less than 85% of the rated load, which is expressed as a continuous time interval on the minute level and synchronized to the local database of the dispatching system through the GB / T27930 protocol.

[0113] The current time is the system time when the scheduling system completes the maximum elastic duration calculation, accurate to the minute, in HHMM format, and synchronized using the NTP protocol.

[0114] If the latest time when charging must begin falls within the grid's predicted congestion period, the maximum elasticity time for the request will be compressed and adjusted. In this embodiment, the compression correction rule is as follows: determine the starting time of the congestion period, calculate the time difference between it and the latest time when charging must begin as the basic correction value, and subtract this correction value from the original maximum elastic duration to obtain the corrected duration.

[0115] The corrected duration must not be negative. If the calculation result is negative, it should be set to 0, and the request must start charging immediately.

[0116] After the tightening correction is completed, the latest time to start charging must be recalculated based on the corrected maximum elasticity duration, and then cross-checked with the congestion period again until it does not fall into the congestion period or the duration is corrected to 0.

[0117] The cross-comparison judgment logic is as follows: if the latest time that must start charging falls between the start and end times of the congestion period (including the boundary value), it is judged as falling into the category.

[0118] The entire correction process takes no more than 5 seconds to ensure efficient scheduling response.

[0119] As a specific example: A user requests 50kWh of electricity, uses a 10kW charging station, and has a theoretical charging time of 5 minutes. The user reports a 2-hour stay, starting at 6 PM. After integrating historical data, the planned end time is 8 PM, with the latest start time being 7:55 PM.

[0120] The power grid predicts congestion from 7:30 PM to 8:10 PM, and the current time is 6:00 PM. Since the latest time to start charging falls within the congestion period, the time difference between the congestion start time of 7:30 PM and 7:55 PM is calculated to be 25 minutes. The original maximum flexibility time of 30 minutes is reduced by 25 minutes, and the revised maximum flexibility time is 5 minutes.

[0121] The latest time to start charging has been recalculated to 20:00 minus 5 minutes, which is 19:55. This still falls within the congestion period. After the second correction, the maximum flexibility time is 0. The request must start charging immediately, and the correction process ends.

[0122] The beneficial effects of this invention are that it can avoid the risk of grid congestion, ensure the stability of grid operation, identify charging requests that may be initiated during congestion periods through cross-comparison, and achieve peak-shifting regulation through flexible time-tightening correction, thereby reducing the probability of grid operation exceeding rated load and reducing problems such as voltage fluctuations and frequency deviations.

[0123] By balancing the completion rate of charging requests with the needs of grid dispatch, and ensuring that the duration is not negative after clear correction, multiple comparisons are made to ensure that charging requests can be completed within the planned dwell time, thus forming a dynamic balance between grid dispatch and user demand and improving the reliability of the charging experience.

[0124] Optimize the efficiency of flexible time utilization, reserve sufficient flexibility during non-congested periods, and compress it in a targeted manner during congested periods, so as to make the utilization of flexible time more targeted and improve the flexibility and adaptability of the scheduling system.

[0125] As one embodiment of the present invention, the method further includes: Extract the historical planned dwell time and the corresponding historical actual dwell time from the requesting user's historical charging records; calculate the deviation value between the historical planned dwell time and the historical actual dwell time; When the deviation value exceeds the preset threshold, the planned dwell time of this charging request is replaced with a statistical value based on the historical actual dwell time, and then the step of identifying the ability to delay charging is executed.

[0126] In this embodiment, the scope of historical charging records extraction is the user's charging data over the past year, with a minimum of 5 records; if fewer than 5, only the actual valid records are counted. Historical planned dwell time refers to the dwell time declared by the user during past charging sessions, while historical actual dwell time refers to the actual time recorded by the charging pile from charging initiation to disconnection, both accurate to the minute.

[0127] The deviation score is calculated by dividing the absolute difference between the historical planned stay duration and the historical actual stay duration by the historical actual stay duration, with the result rounded to two decimal places. The deviation score is used to assess the credibility of a user's historical reporting behavior. A deviation score threshold can be preset; when the calculated deviation score exceeds this threshold, the credibility of the user's current report is considered low.

[0128] As an example, the deviation threshold can be set to 0.3. This value can be set based on the analysis of historical operational data. For example, when the threshold is set to 0.3, it is possible to filter out the user reports with the highest deviation and make corrections, thus striking a balance between correction accuracy and data coverage.

[0129] The statistical value based on the actual historical stay duration is calculated using the median. If there is an even number of actual historical stay durations, the average of the two middle values ​​is taken and rounded up to the minute level. Outliers are removed during the statistical process. The criteria for outliers is a value that deviates from the average of all actual historical stay durations by ±30%.

[0130] After the replacement is completed, the steps to identify the delayed charging capability are executed according to the original method and process to ensure the accuracy of parameters in subsequent steps such as maximum elasticity calculation, congestion period comparison and tightening correction. The entire data extraction and correction process takes no more than 3 seconds and does not affect the scheduling response efficiency.

[0131] This invention extracts planned and actual dwell time from users' historical charging records, calculates the deviation, and filters low-credibility reporting requests based on preset thresholds. It then replaces the planned dwell time with the median of historical actual dwell time after outlier removal. This effectively avoids parameter distortion caused by user misreporting or arbitrary reporting, providing reliable data support for subsequent scheduling. Furthermore, through quantified threshold settings, clear outlier removal rules, and statistical methods, it ensures the corrective logic is implementable. Simultaneously, it considers individual differences in user charging scenarios and the overall accuracy of the scheduling system, improving the adaptability of charging scheduling schemes to users' actual usage habits. This reduces charging interruptions, grid scheduling conflicts, and resource waste caused by parameter deviations, lowers the frequency of system maintenance adjustments and dispute rates, and ultimately improves charging pile utilization, grid load regulation accuracy, and the overall stability and efficiency of the scheduling system.

[0132] As one embodiment of the present invention, before the step of sorting according to the urgency of charging, the method further includes: Obtain historical charging records of users with low elasticity and guaranteed charging requests under similar initial battery levels and similar time periods; When historical records show that a user's actual stay duration is generally shorter than their corresponding declared planned stay duration, the urgency of the user's current charging request will be increased by one level in the ranking.

[0133] In this embodiment, the criteria for determining low-elasticity guarantee requests are the same as those mentioned above, namely, the user's charging needs have essential attributes, and charging interruption will affect the core usage scenario.

[0134] The range for determining similar initial power levels is the initial power level declared by the user plus or minus 10%. This range is set based on the common power metering accuracy of charging piles and the fluctuation pattern of user charging demand to ensure the matching degree of power levels in historical records.

[0135] The rule for dividing similar time periods is to divide them into 2-hour intervals, and to match both weekday and non-weekday attributes. That is, requests for weekdays will only match records of the same interval for historical weekdays, and the same applies to non-weekdays, to ensure consistency of time period scenarios.

[0136] The conditions for extracting historical charging records are that there are at least 3 valid records with similar initial charge levels and similar time periods; if there are fewer than 3, this supplementary step will not be performed.

[0137] The historical data extraction scope covers the user's charging data over the past 6 months, balancing data timeliness and sample representativeness.

[0138] The criteria for determining whether the actual stay duration is generally shorter than the corresponding declared planned stay duration are as follows: in historical records under similar conditions, the proportion of records with actual stay duration shorter than planned stay duration is not less than 70%, and the average duration difference of each corresponding record is not less than 15 minutes. If both conditions are met, it is determined to be generally shorter.

[0139] This standard is based on statistics of inelastic user charging behavior within the region and can accurately identify users who habitually report short stops.

[0140] The specific rule for increasing the urgency of charging by one level is to raise the priority of the user's current request by one level based on the original sorting rules. The adjusted priority shall not exceed the highest level to avoid priority overflow affecting the overall sorting logic.

[0141] After executing this supplementary step, the steps based on the urgency of charging will be executed according to the original method. The entire data extraction, judgment and priority adjustment process will take no more than 2 seconds and will not affect the overall scheduling efficiency.

[0142] This invention accurately matches users with similar initial battery levels and historical charging records for similar time periods for those making inelastic, guaranteed charging requests. Based on quantitative criteria, it identifies users who habitually submit short-stay reports and specifically upgrades their charging urgency ranking. This effectively avoids the risk of unmet charging needs due to user habitual reporting errors, protecting the core charging rights of inelastic users. Furthermore, through clear definition of similarity conditions, sample size requirements, and judgment rules, it ensures that the adjustment logic is implementable and free from subjective bias. It also considers the timeliness and representativeness of historical data, making the charging urgency ranking more closely reflect users' actual charging behavior and needs. This improves the accuracy of the ranking results and the rationality of the scheduling plan, further strengthening the service priority of inelastic, guaranteed charging requests, reducing user disputes caused by improper ranking, and enhancing user satisfaction and operational stability of the entire scheduling system.

[0143] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method for optimizing and controlling the metering and billing of charging piles based on an intelligent urban management platform, characterized in that, The optimization is achieved through the following collaborative scheduling process: Identify the deferred charging capability of each charging request and classify the requests into highly resilient schedulable requests and low resilient guaranteed requests based on this capability. Calculate the total schedulable power that all highly elastic schedulable requests can provide during the target adjustment period; Compare the total dispatchable power with the power demand for grid regulation; Scheduling is performed based on the comparison results: when the total schedulable power is greater than or equal to the regulation demand power, the selected high-elasticity schedulable requests are scheduled; when the total schedulable power is less than the regulation demand power, in addition to scheduling the high-elasticity schedulable requests, some low-elasticity guarantee requests are supplemented by scheduling according to preset rules. Dynamic billing is performed based on the scheduling execution results.

2. The charging pile metering and billing optimization control method based on an intelligent urban management platform according to claim 1, characterized in that, Identify the deferred charging capability of each charging request and classify the requests based on this capability, including: Obtain the remaining battery power, target charging level, and planned dwell time for the charging request; Calculate the total rigid power required to complete the charging process based on the remaining battery power and the target charging power. Based on the total rigid demand for electricity and the planned dwell time, calculate the maximum flexible duration for which the charging start time can be delayed. By comparing the relationship between the maximum elastic duration and the planned dwell time, charging requests are divided into highly elastic schedulable requests and low elasticity guaranteed requests.

3. The charging pile metering and billing optimization control method based on an intelligent urban management platform according to claim 2, characterized in that, After calculating the maximum allowable resilience duration for the charging start time delay, the following is also included: Based on the planned stay duration and the maximum flexible duration, calculate the latest time that the charging request must start charging within the planned stay duration; Cross-compare the latest time when charging must begin with the current time and the congestion period predicted by the power grid. If the latest time when charging must begin falls within the grid's predicted congestion period, the maximum elasticity time for the request will be shortened and adjusted.

4. The charging pile metering and billing optimization control method based on an intelligent urban management platform according to claim 1, characterized in that, Scheduling is performed based on the comparison results, specifically including: When the total schedulable power is greater than or equal to the adjustment demand power, the requests are sorted according to the maximum elastic duration of each highly elastic schedulable request, and the requests with the longer maximum elastic duration are scheduled first. When the total dispatchable power is less than the regulation demand power, a tiered supplementary scheduling is performed: first, all responsive, highly flexible dispatchable requests are included in the scheduling; then, the power gap is calculated; and the charging urgency is determined by the remaining battery power of each low-flexibility guarantee request. Supplementary scheduling is then performed starting with the request with the lowest urgency to fill the power gap.

5. The charging pile metering and billing optimization control method based on an intelligent urban management platform according to claim 4, characterized in that, In tiered supplemental scheduling, scheduling of low-elasticity guarantee requests is achieved through progressive power control that includes state verification: Set an initial charging power threshold below the rated power requirement for the selected low-resilience guarantee request. Charging is performed using the initial charging power threshold, and the vehicle's connectivity status is continuously monitored; After monitoring that the vehicle's connection status has remained stable for a period of time longer than the preset verification time, its charging power is increased to the target power level required by the scheduling plan. If a vehicle connection is detected to be disconnected before the verification time is reached, the request is removed from the current scheduling task, and a new scheduling object is selected from the backup scheduling resources based on the power gap.

6. The charging pile metering and billing optimization control method based on an intelligent urban management platform according to claim 1, characterized in that, The dynamic billing based on the scheduling execution results is executed in conjunction with the following operations to maintain stable scheduling performance: Real-time monitoring and comparison of actual charging power curves with scheduling schemes; When it is detected that the actual charging behavior deviates from the scheduling plan and generates a power adjustment gap, based on the power adjustment gap and the pre-calculated power available for scheduling from the backup request queue, one or more backup requests are automatically selected and started for charging. Based on the deviation behavior and the initiation and execution of the backup request, the billing parameters of the relevant parties are updated.

7. The charging pile metering and billing optimization control method based on an intelligent urban management platform according to claim 1, characterized in that, The method further includes: Extract the historical planned dwell time and the corresponding historical actual dwell time from the requesting user's historical charging records; Calculate the deviation between the historical planned duration of stay and the historical actual duration of stay; When the deviation value exceeds the preset threshold, the planned dwell time of this charging request is replaced with a statistical value based on the historical actual dwell time, and then the step of identifying the ability to delay charging is executed.

8. The charging pile metering and billing optimization control method based on an intelligent urban management platform according to claim 2, characterized in that, Get the planned dwell time for the request, specifically: When the number of a user’s historical charging records reaches the statistical requirement, read the average or median of their historical actual dwell time. The average or median of the historical actual stay duration is weighted and combined with the user's current declared planned stay duration to generate the corrected planned stay duration. Using the revised planned stay duration, perform the step of calculating the maximum flexible duration.

9. The charging pile metering and billing optimization control method based on an intelligent urban management platform according to claim 4, characterized in that, Before the steps that prioritize charging urgency, the following are also included: Obtain the historical charging records of the low-elasticity guarantee request user under similar initial power levels and similar time periods; When the historical records show that the user's actual stay time is generally shorter than the corresponding declared planned stay time, the urgency of the user's current charging request will be increased by one level in the ranking.

10. The charging pile metering and billing optimization control method based on an intelligent urban management platform according to claim 6, characterized in that, The steps for selecting a standby request from the standby request queue include: Calculate the number of times each standby requesting user actively participated in scheduling and the number of times passive standby was activated in the historical records; For users whose ratio of active scheduling participation to passive backup activation is lower than a set value, their backup requests will be ranked lower in the queue. Based on the downgraded sorting, perform the operation of selecting and initiating the standby request.

Images