A charging pile operation management method and system

Abstract

The present application relates to the technical field of charging pile management and power dispatching, and specifically discloses a charging pile operation management method and system. The method comprises: obtaining charging request data and power grid load data, generating and cleaning task vector data set; predicting charging demand and generating electricity price gradient information based on the data set; constructing a multi-objective optimization dispatching function and solving, generating a task allocation and dispatching scheme; executing power control according to the dispatching scheme and collecting charging state data; generating dynamic contract billing and incentive according to the state data; finally, performing hash packaging and privacy processing on the execution result, and generating a trusted record. The present application can improve the resource allocation efficiency and intelligent level of billing and incentive in the multi-user and multi-period concurrent charging scene, and realize efficient cooperation and trusted execution of the whole process of charging dispatching and settlement.

Description

Technical Field

[0001] This invention relates to the field of charging pile management and power dispatching technology, and in particular to a charging pile operation management method and system. Background Technology

[0002] With the rapid popularization of new energy vehicles, the number and usage frequency of electric vehicle charging stations in cities continue to rise. Charging requests are showing trends of high concurrency, concentrated time periods, and diversified power demands. Traditional static scheduling strategies and fixed-price billing models have revealed significant shortcomings in multi-user, multi-time-period concurrent charging environments. On the one hand, existing scheduling systems generally rely on single-objective or static parameters for power allocation, making it difficult to simultaneously consider multiple objectives such as user charging efficiency, grid load balance, and site equipment utilization. On the other hand, current mainstream billing systems lack integrated processing of user behavior differences, real-time electricity price changes, and charging feedback data, and cannot achieve smart contract execution linked to user incentive strategies, resulting in inflexible billing, unreasonable resource allocation, and limited user experience.

[0003] Furthermore, existing charging management platforms generally lack support for data credibility and privacy protection mechanisms in their system architecture, leading to difficulties in tracing transaction records, easy leakage of user data, and weak trust systems, thus limiting the security and scalability of the system. Currently, there is a lack of a charging pile operation and management technology solution that integrates dynamic resource scheduling optimization, smart contract automatic billing, and privacy protection mechanisms. There is an urgent need to introduce multi-objective scheduling models and distributed contract mechanisms to achieve a synergistic improvement in resource allocation efficiency and user billing fairness. Summary of the Invention

[0004] This invention provides a charging pile operation management method and system to solve the problem of how to construct a multi-objective optimization scheduling function based on task vector data and electricity price gradient and implement a differentiated dynamic contract billing mechanism, thereby improving the efficiency of resource allocation and the automation and intelligence level of billing incentives in multi-user, multi-time-period concurrent charging scenarios.

[0005] To address the aforementioned technical problems, this invention provides a charging pile operation and management method, comprising the following steps:

[0006] Based on the charging request data of charging users and the power grid load data, a cleaned task vector dataset is obtained.

[0007] Based on the cleaned task vector dataset and power grid load data, charging demand is predicted and electricity price gradient information is generated.

[0008] Obtain the electricity price gradient information and the task vector dataset, construct a set of multi-objective optimization scheduling functions, and generate a priority-ranked task allocation and scheduling scheme. The task allocation and scheduling scheme is as follows:

[0009]

[0010] in, The final output scheduling scheme; (i, j) represents the task-device allocation pair; To represent selecting the top K highest-scoring assignment pairs from all the rating results; Π i，j Scoring is assigned to scheduling priorities;

[0011] The task is assigned and dispatched according to the task allocation and scheduling scheme, power control is executed and charging execution status data is collected;

[0012] Based on the charging execution status data, dynamic contract billing and incentive generation are performed to form a smart contract execution result set, which represents:

[0013]

[0014] in, For user m, the set of smart contract execution results; ID m A unique identifier for the user task; This is the bill amount; To incentivize ratings; For the actual task start timestamp and the actual task completion timestamp; λ t Electricity price level; To determine whether reverse power supply behavior exists;

[0015] The results of smart contract execution are hashed and processed for privacy, generating trusted data records on the blockchain.

[0016] Furthermore, the step of generating a task vector set based on the charging user's charging request data and the power grid load data, and then performing the cleaning process, specifically includes:

[0017] The system acquires charging request data from charging users, grid load data, and site equipment status information, and performs structural analysis and data time alignment processing.

[0018] Power demand, duration, and task urgency are extracted from charging request data and grid load data to generate a set of task vectors;

[0019] The task vector set is formatted and outlier removed to obtain a cleaned task vector dataset.

[0020] Furthermore, the prediction of charging demand and generation of electricity price gradient information specifically includes:

[0021] Obtain the cleaned task vector dataset and power grid load data;

[0022] The time series model is trained and inferred to generate short-term charging demand forecast data.

[0023] Based on the charging demand forecast data and the cleaned task vector dataset, the electricity price gradient information for the corresponding time period is generated.

[0024] Furthermore, the construction and solution of the multi-objective optimization scheduling function set specifically includes:

[0025] Obtain electricity price gradient information and a cleaned task vector dataset, and construct a set of multi-objective optimization scheduling functions;

[0026] The initial task scheduling results are obtained by solving the set of multi-objective optimization scheduling functions.

[0027] Furthermore, the task allocation and scheduling scheme for generating priority order specifically includes:

[0028] Based on the initial task scheduling results and the cleaned task vector dataset, a priority-ranked task allocation and scheduling scheme is generated.

[0029] Furthermore, the collection of charging execution status data specifically includes:

[0030] Obtain the task allocation and scheduling plan, generate task issuance instructions, and allocate them to charging station equipment;

[0031] Based on the task-issued instructions and the charging station equipment status information, power control and bidirectional power supply adjustment operations are performed.

[0032] Collect output data from charging station equipment during task execution to generate charging execution status data.

[0033] Furthermore, the execution of dynamic contract billing specifically includes:

[0034] Obtain charging execution status data and electricity price gradient information, perform dynamic contract billing processing, and generate billing data.

[0035] Furthermore, the set of smart contract execution results specifically includes:

[0036] Based on the billing data and the cleaned task vector dataset, user incentive strategy data is generated;

[0037] The billing data and user incentive strategy data are structured and encapsulated to generate a set of smart contract execution results.

[0038] Furthermore, the generation of trusted data on-chain records specifically includes:

[0039] Obtain the set of smart contract execution results and encapsulate them using field hashing and encrypted parameters;

[0040] Based on the field hash and encryption parameter encapsulation results, a privacy-protected identifier structure for verification is generated;

[0041] The privacy protection identifier structure is packaged with the smart contract execution result set to generate a trusted data record on the blockchain.

[0042] Furthermore, a charging pile operation and management system includes:

[0043] The task generation module is used to generate a set of task vectors based on the charging request data of charging users and the grid load data, and to complete the cleaning process.

[0044] The prediction module is used to generate charging demand prediction data and electricity price gradient information based on the cleaned task vector dataset and grid load data;

[0045] The scheduling module is used to construct and solve a set of multi-objective optimization scheduling functions to generate priority-ordered task allocation and scheduling schemes.

[0046] The execution module is used to issue tasks, perform power control, and collect charging execution status data.

[0047] The contract module is used to perform dynamic contract billing based on charging execution status data and generate a set of smart contract execution results.

[0048] The on-chain module is used to perform hash encapsulation and privacy processing on the smart contract execution result set, generating trusted data on-chain records.

[0049] The key innovations of this invention include:

[0050] (1) Integration of task scheduling optimization algorithms: Construct a set of multi-objective scheduling functions that integrate electricity price gradient response, task timeliness and power load elasticity, and introduce a combination algorithm of Bellman equation and complex variable partial derivative to form an optimal solution mechanism for scheduling paths.

[0051] (2) Dynamic expression capability of smart contracts: A differentiated billing model based on the change of charging execution state and the result of multi-indicator normalization was designed, and the real-time incentive adjustment of user behavior was realized through game mechanism.

[0052] (3) Full-process integration of on-chain privacy protection mechanism: A smart contract result encapsulation method based on hash index is proposed, which can ensure that user transaction data is synchronously stored, verified and protected across different platforms, and protect users' willingness to participate and the system's trustworthy collaboration.

[0053] The following are its main beneficial effects:

[0054] (1) Achieve multi-objective collaborative scheduling optimization and improve the utilization efficiency of charging resources. This invention constructs a multi-objective scheduling optimization function that integrates electricity price gradient, task urgency and supply-demand deviation factors, and adopts a topology scheduling coupling model and functional variational inference algorithm to dynamically solve the optimal allocation path of charging tasks. Compared with traditional first-come-first-served or static polling strategies, it can significantly reduce queuing time and improve the overall charging pile utilization rate during peak charging periods.

[0055] (2) Introducing a differentiated dynamic contract billing mechanism to enhance the intelligent matching capability of billing and incentive strategies. In the dynamic contract module, this invention introduces a billing factor weighting calculation mechanism based on joint modeling of task execution status and electricity price gradient. Combined with a weighted curve integral model and Nash equilibrium strategy, it supports dynamic calculation of user bills and incentive amounts according to task type, completion efficiency, and load response, thereby more responsively guiding users to charge during off-peak hours or participate in V2G regulation, and promoting supply and demand balance.

[0056] (3) Construct a trusted data link throughout the entire process to ensure verifiable and privacy-protected multi-party interactions. This invention uses field hashing and encryption encapsulation strategies to structure the execution results of smart contracts and generate a privacy-protected identifier structure for cross-system verification. Finally, it packages the results into a trusted on-chain record. Compared with the traditional centralized database reconciliation method, this improves the tamper-proof and traceability capabilities of the data interaction process and enhances the credibility of data collaboration between systems. Attached Figure Description

[0057] Figure 1 A flowchart illustrating a charging pile operation and management method provided in an embodiment of this application;

[0058] Figure 2 This is a structural block diagram of a charging pile operation and management system provided in an embodiment of this application. Detailed Implementation

[0059] Example 1: Refer to Figure 1 This is a flowchart illustrating a charging pile operation and management method provided in an embodiment of the present invention. The process may include at least steps S100-S600:

[0060] S100. Based on the charging request data of charging users and the grid load data, generate a task vector set and complete the cleaning process to obtain the cleaned task vector dataset.

[0061] S200. Based on the task vector dataset and power grid load data, predict charging demand and generate electricity price gradient information.

[0062] S300. Construct and solve a set of multi-objective optimization scheduling functions to generate a priority-ordered task allocation and scheduling scheme.

[0063] S400: Based on the task allocation and scheduling scheme, complete the task issuance, execute power control, and collect charging execution status data.

[0064] S500 generates dynamic contract billing and incentives based on charging execution status data, forming a set of smart contract execution results.

[0065] S600 performs hash encapsulation and privacy processing on the smart contract execution result set to generate trusted data records on the blockchain.

[0066] Step S100 includes at least steps S110-S130:

[0067] S110: Obtain charging request data from charging users, grid load data, and site equipment status information, and perform structural analysis and data time alignment processing.

[0068] Specifically, the system acquires charging request data recorded from the charging station operation platform, user mobile terminal, or vehicle internal control terminal. The charging request data includes user identification, reservation time period, expected charging power, vehicle type identification, etc. Simultaneously, it acquires grid load data from the power dispatch platform. The grid load data includes fields such as the total power carrying capacity of the current power supply area, zoned load curves, and instantaneous voltage stability level. At the same time, it acquires site equipment status information output by the charging station's internal management system. The status information includes charging pile number, power level, port availability, current operating status, and fault records, etc.

[0069] After the data access is completed, a unified structure parsing operation is performed to match fields, label types, and correct structures for the three types of data, ensuring that the data structure can be received and used by subsequent processing modules. Based on the parsing, further data time alignment processing is performed to unify the timestamps of charging request data, grid load data, and site equipment status information, forming a data set with temporal consistency. This time alignment operation employs a unified time base alignment strategy, where all data is sliced ​​at the smallest time granularity to eliminate synchronization deviations caused by different data acquisition frequencies.

[0070] The structural analysis and data alignment processes completed in this step provide a standardized data format for generating S120 task vectors, while ensuring that the time consistency condition is met in the subsequent modeling process.

[0071] S120. Extract power demand, duration and task urgency from the charging request data and the grid load data to generate a task vector set.

[0072] Based on a unified and time-aligned dataset, three core scheduling factors are extracted: the power demand of charging users, the expected duration, and the urgency of the task, to construct a task vector set.

[0073] The power requirement is derived from the reservation parameters submitted by the user or the power setting value fed back by the vehicle management system, and is combined with the site equipment status information for boundary correction so that the extracted value conforms to the current site's executable capacity range; the duration is derived from the user's reservation time period and the estimated vehicle filling time, forming a scheduling interval in the time dimension; the task urgency is calculated by combining multiple factors such as the relative interval between the current time and the reservation time period, the user's identity type (such as emergency vehicle, shared fleet, private car, etc.) and the user's historical credit rating, forming a scheduling priority level index.

[0074] The three types of parameters are uniformly abstracted into a task scheduling parameter set, which is then transformed into task vectors through field structuring. The task vector set is stored using a unified data structure, and each task vector serves as one of the basic input items for subsequent multi-objective scheduling functions and prediction models.

[0075] This step is strongly related to the task scheduling modeling process of S300, where the task urgency will be directly referenced in the priority ranking stage and will affect the final task allocation order.

[0076] S130. Perform format conversion and outlier removal on the task vector set to obtain a cleaned task vector dataset.

[0077] Furthermore, the task vector set is subjected to structure format standardization and outlier removal operations to form a data structure suitable for use by the S200 load prediction model and the S300 scheduling modeling module, generating a cleaned task vector dataset.

[0078] First, a format conversion process is performed to normalize the power requirement, duration, and task urgency fields in the task vector to the system's set standard range, unifying the data type format (such as converting integers to floating points, time periods to duration variables, etc.). After the format is unified, the cleansing rule base is called to perform an outlier removal operation, including cases where the power value exceeds the equipment's maximum load, the duration exceeds the site's operating cycle, or the urgency is negative. The outlier vector will be recorded and removed from the set.

[0079] The cleaned task vector dataset retains the same timestamps and structure fields as the original charging request data, grid load data, and site status information. It can be directly used as input data for building a time series prediction model in S210, and also provides cleaned and accurate parameters for the task constraints in the subsequent S310 scheduling function.

[0080] Through the unified structure parsing, field extraction, and cleaning mechanism set in step S100 of this invention, the fusion and standardization of multi-source data at the time, structure, and semantic levels are achieved, constructing a task vector dataset with timeliness and modeling adaptability. This dataset not only significantly improves the accuracy of the charging demand prediction model but also establishes a stable data input foundation for scheduling optimization and contract execution, thereby forming a controllable and iterative data closed-loop structure in the overall system, providing reliable input assurance for subsequent modules.

[0081] Step S200 includes at least steps S210-S230:

[0082] S210. Obtain the task vector dataset and the power grid load data, and construct a time series model for short-term regional load forecasting.

[0083] Specifically, based on the cleaned task vector dataset obtained in step S130, it is combined with grid load data provided by charging stations or regional power dispatch centers for input processing to construct a short-term regional load prediction model. The model construction includes the following:

[0084] First, read the data segments in the task vector dataset that are associated with the current prediction window, including the power demand field, duration field, and task urgency field; second, use the above fields as feature inputs and perform joint modeling processing with the historical load power value, peak and valley load identifier, load growth rate, and other numerical features in the power grid load data under the same time window.

[0085] Understandably, the time series model constructs a training sample set using a fixed sliding window, supporting short-term predictions for multiple consecutive time periods, and is suitable for the load management needs of city-level or regional-level charging station networks. Furthermore, the modeling process supports a multi-task training structure, used to simultaneously predict the changing trends of different user demands within multiple time periods.

[0086] The input data of the model depends on the completeness and standardization of the task vector field in step S130, and the model’s time granularity and task time alignment rules are inherited from the structure parsing and time alignment mechanism in step S110.

[0087] S220. The time series model is trained and inferenced to generate short-term charging demand prediction data.

[0088] After the time series model is built, training and inference operations are performed on the model. Specifically, this includes: first, constructing a training sample set based on historical task vector data and known load data, calling the model training module to perform parameter optimization, and forming a regional load prediction model with generalization ability; then, inputting the current real-time task vector dataset and grid load data, calling the trained model to perform inference processing, and generating short-term charging demand prediction data.

[0089] Furthermore, the short-term charging demand forecast data is the interval demand response value within the time segmentation period set by the current system, which is used to describe the growth or decline trend of the charging power required per unit time in multiple upcoming time windows for a certain site or region.

[0090] Understandably, the predicted data is indexed by timestamps, and a sequence of predicted values ​​is output along with a confidence estimation field, which is used to set adjustment factors for multi-objective functions during subsequent dynamic scheduling.

[0091] The charging demand forecast data will serve as the basis for determining peak and valley loads and prioritizing time periods during the subsequent S230 electricity price gradient generation process.

[0092] S230. Based on the charging demand prediction data and the task vector dataset, generate electricity price gradient information for the corresponding time period.

[0093] After acquiring the charging demand prediction data, based on the predicted values ​​and the scheduling time and power demand fields in the task vector dataset, electricity price gradient information for the corresponding time period is generated. Specifically, this includes:

[0094] First, the prediction period is divided into multiple continuous windows according to the time slice granularity defined by the system. Second, the predicted power value in each time window is weighted and combined with the task request quantity in the task vector for that window to generate a comprehensive load pressure index. Third, based on the comprehensive load pressure index, the corresponding electricity price level value is generated according to the tiered rules to finally form the electricity price gradient information.

[0095] Furthermore, the electricity price gradient information is output in units of time windows, and each record includes the start and end points of the time period, the corresponding electricity price level identifier, and the load index source field. The electricity price level is used as a cost constraint factor to be embedded in the multi-objective optimization function set in S310 during subsequent task scheduling, and the load index source field is used to assist in constructing incentive strategies.

[0096] The electricity price gradient information will be directly used in the dynamic contract billing processing step of S510 to match the user task execution period with the current electricity price level, thereby determining the billing formula parameters, and generating incentive weights through S520.

[0097] This step achieves the transformation from demand forecasting to cost control at the task level, and forms a closed loop with the S300 scheduling function and S500 billing function structure.

[0098] By jointly modeling the task vector dataset and the power grid load data, this step constructs a short-term regional load forecasting model and completes the inference of demand forecasting results. Based on this, electricity price gradient information is generated in combination with task scheduling requirements, providing a parameter foundation with consistent time accuracy and clear structural semantics for the subsequent construction of multi-objective scheduling functions and matching of contract billing rules. This realizes the data connection between task-level scheduling control and operation-level cost management, and ensures the coordination and consistency of the system execution link.

[0099] Step S300 includes at least steps S310-S330:

[0100] S310. Obtain the electricity price gradient information and the task vector dataset, and construct a set of multi-objective optimization scheduling functions.

[0101] Specifically, the cleaned task vector dataset generated in step S130 and the electricity price gradient information generated in step S230 are obtained. Based on this, a function system integrating multiple scheduling objectives such as charging efficiency, electricity price cost, grid load balance, and task urgency is constructed.

[0102] To achieve optimal scheduling decisions, the following multi-objective scheduling optimization model is designed:

[0103] ① Multi-objective functional optimization objective function

[0104] Construct the following multi-objective functional:

[0105]

[0106] in:

[0107] A multi-objective optimization scheduling functional is used to express the comprehensive objective function of the charging task allocation scheme;

[0108] x: A set of task-device scheduling variables, representing the allocation path decisions for multiple tasks on multiple devices;

[0109] α1, α2, α3: These represent the weight parameters of the three optimization objectives, and their values ​​are set by the system strategy.

[0110] Ω1, Ω2, Ω3: represent the electricity price cost integral interval, the grid load integral interval, and the emergency dispatch interval, respectively;

[0111] Φ1(x, t): represents the electricity price dispatch cost function per unit time under task-time allocation x;

[0112] Φ2(x,t): represents the power grid load disturbance term caused by dispatching per unit time;

[0113] Φ3(x,t): Represents the weighted scheduling index of the task scheduling path in terms of task urgency.

[0114] ② Construction of electricity price dispatch items (combined with electricity price gradient)

[0115] Based on the electricity price gradient information generated by S230, the following electricity price cost scheduling term is constructed:

[0116] Φ1(x,t)=P t ·D i，t

[0117] in:

[0118] P t The unit electricity price at time point t is derived from the electricity price gradient information generated in S230;

[0119] D i，t : The duration allocated to task i at time t, used to represent the time cost of the scheduled task within a certain time slice.

[0120] This metric measures the price cost of each task during the execution scheduling process.

[0121] ③ Construction of power grid load disturbance terms (topology and load)

[0122]

[0123] in:

[0124] The actual load value of the h-th charging station at time t;

[0125] The average load of all system sites at time t;

[0126] Spatial gradient operator, used here to measure the spatial offset of load between sites;

[0127] |·|: Represents the vector magnitude, i.e., the measure function of load deviation;

[0128] This function represents the squared difference of load disturbance at each site, used to measure load balance. This function guides the scheduling results to approach the overall load balance of the system.

[0129] ④ Construction of urgency scheduling items (task priority)

[0130]

[0131] in:

[0132] N: The total number of tasks that can be scheduled in the current scheduling window;

[0133] δ i : Whether task i is scheduled (a scheduling variable, with a value of 0 or 1);

[0134] U i The urgency value of task i is derived from the task vector field in S120;

[0135] This function is a weighted sum of the urgency of all selected tasks, used to favor high-priority tasks. This implements the priority control mechanism for task scheduling.

[0136] S320. Solve the set of multi-objective optimization scheduling functions to obtain the initial scheduling result of the task.

[0137] After constructing the multi-objective optimization functional, a multi-objective shortest path mapping function is used for the search. This search process incorporates a variational extreme point determination criterion and performs path graph structure compression.

[0138] ⑤ Path optimization objective function

[0139]

[0140] Where: min: represents the functional Find the minimum value, that is, find the optimal scheduling path;

[0141] x: Set of task-equipment allocation decision paths;

[0142] The set of all possible task allocation schemes is called the scheduling feasible space;

[0143] δ i，j : Scheduling decision variable, indicating whether to assign task i to device j;

[0144] ∑ j δ i，j =1: Constraint condition, ensuring that each task is assigned to only one device;

[0145] δ i，j ∈{0,1}: A binary variable definition representing the state of the scheduling decision.

[0146] ⑥ Definition of multi-objective optimization result indicators

[0147] Output the initial scheduling result set:

[0148]

[0149] in, This represents the initial scheduling result set for task i and device j;

[0150] δ i，j : Whether to allocate;

[0151] P t Electricity price level for the corresponding time period;

[0152] D i，t Task duration;

[0153] U i : Task urgency;

[0154] φ(x, t): The objective function value of the task allocation scheme at time t.

[0155] The scheduling result will be used in S330 to construct the priority sorting logic.

[0156] S330. Based on the initial task scheduling result and the task vector dataset, generate a priority-ranked task allocation and scheduling scheme.

[0157] Obtain the initial task scheduling result set Then, a sorting function is constructed to score and sort all task device matching pairs, and finally an executable scheduling list is output.

[0158] ⑦ Priority scoring function

[0159]

[0160] in:

[0161] Π i，j The scheduling priority score of task i and device j;

[0162] Ψ(): Multi-factor priority scoring function;

[0163] ω1, ω2, ω3: Priority weight coefficients, set by the user or platform;

[0164] P t Electricity price per unit of time t;

[0165] Load gradient of site j;

[0166] The maximum value of the load gradient within the scheduling time interval is used for normalization;

[0167] U i The urgency of task i;

[0168] The maximum urgency level within the current scheduling cycle, used for normalization.

[0169] This scoring function is used to select task scheduling combinations according to priority. The three factors correspond to priority of low electricity price, priority of load balancing, and priority of task urgency, respectively.

[0170] ⑧ Sorting and Scheduling Output

[0171] Combine all tasks – devices according to ∏ i，j Sort the values ​​from high to low, and select the top few scheduling instructions to generate a scheduling list:

[0172]

[0173] in:

[0174] The final output scheduling scheme;

[0175] (i, j): Task-device allocation pair;

[0176] This means selecting the top K highest-scoring assignments from all the rating results;

[0177] ∏ i，j The scheduling priority score of task i and device j;

[0178] Output scheduling scheme structure:

[0179] Includes scheduling sequence, task number, equipment number, execution time period, and electricity price level.

[0180] The scheduling scheme will serve as input for subsequent S410 task issuance and drive V2G control execution.

[0181] By introducing a multi-objective optimization scheduling model and combining the task vector data and electricity price gradient input provided by this invention, the optimal task-equipment allocation structure can be output under multiple constraints such as grid load balance, user cost control and response time. This establishes an interpretable and executable charging scheduling system at the task level, effectively supporting the execution of subsequent power regulation and contract settlement links, demonstrating the technological advancement of this invention in terms of intelligence, mathematical modeling and system engineering integration.

[0182] Step S400 includes at least steps S410-S430:

[0183] S410. Obtain the task allocation and scheduling scheme, generate task issuance instructions, and allocate them to charging station equipment.

[0184] Specifically, the task allocation and scheduling scheme output in step S330 is obtained. This scheme includes multiple matching pairs of task numbers and station device numbers, as well as the charging start time, target power value, estimated duration, and priority ranking index corresponding to each task allocation. The task allocation and scheduling scheme is then structured and parsed, and task issuance instructions are generated according to the current system's established issuance protocol and device interface standards.

[0185] The task issuance instruction includes the following fields:

[0186] Task number;

[0187] Target device number;

[0188] Command issuance time;

[0189] Scheduling time period (start and end time);

[0190] Assign power levels;

[0191] Electricity price level codes (referenced from S230 output);

[0192] Supports V2G bidirectional operation flag (automatically configured by S300 scheduling policy);

[0193] Priority label.

[0194] During the instruction generation process, consistency verification is also required based on the device status information and task urgency information provided in steps S100 to S130 to ensure that the target device for instruction scheduling is currently in a receptive state and that its charging capacity meets the power and time requirements of the assigned task. If the verification fails, the process reverts to the alternative scheme list in the task scheduling scheme in S330 for replacement.

[0195] Once verification is complete, the task issuance command will be distributed to the specific site equipment control module via the local area communication bus or the platform edge node gateway interface.

[0196] S420. Based on the task-issued instructions and the charging station equipment status information, perform power control and V2G bidirectional power supply adjustment operations.

[0197] Furthermore, based on the task issuance instruction generated and issued in step S410, and the charging station equipment status information obtained in step S110, the power regulation control process and bidirectional energy management operation are executed in each specific station equipment controller.

[0198] Specifically, the following operations are included:

[0199] Dynamic power setting: Based on the power level field in the task-issued instruction and combined with the current available port information of the device, the power output value is set in real time, and the output power is dynamically tracked using proportional-integral-derivative control logic.

[0200] Time period control verification: The system controls the on / off status of equipment based on the scheduling start and end time fields, and records the deviation between the start time and the actual response time as execution response data in the subsequent billing strategy.

[0201] V2G Response Judgment: If the task instruction indicates support for V2G reverse power supply operation, and the power grid is currently in a peak load state, the system calls the device's bidirectional converter module to enable the device to perform reverse power supply operation and feed battery power back to the power grid; if the power grid state does not allow it, the reverse power supply action is skipped and the system enters the normal charging state.

[0202] Power disturbance monitoring and feedback coordination: During power control execution, the system continuously acquires the output power change curve of the equipment and compares it with the scheduled target power curve in real time. If the deviation exceeds the tolerance threshold, fine-tuning compensation is performed to ensure that the task power target is achieved.

[0203] Understandably, the power control and power supply regulation operations need to meet multi-objective constraints, and their parameter configurations are derived from the function constraint results of the S300 scheduling model. At the same time, the output state data will be used for S430 state generation and S500 billing model invocation.

[0204] S430: Collect output data of charging station equipment during task execution and generate charging execution status data.

[0205] Once the task is actually executed by the device, the system collects key output data from the entire task execution process in real time and constructs a "charging execution status data" structure. This data structure includes, but is not limited to:

[0206] Actual start time and end time;

[0207] Actual charging duration;

[0208] Actual output power curve;

[0209] Total power output value;

[0210] Voltage and current variation curves;

[0211] Interrupt count and fault flags;

[0212] Whether to perform reverse power supply (V2G);

[0213] The current task number is bound to the device number in the record;

[0214] Electricity price level and actual metering status.

[0215] The above execution status data are all input variables for the subsequent S500 dynamic contract billing module. The system needs to complete the following operations:

[0216] Sampling during execution: Data is acquired during the charging process according to the minimum sampling period, and structural encoding is completed;

[0217] Abnormal status identification: Mark abnormal behaviors such as sudden power drop, abnormal power failure, and interruption recovery with abnormal labels;

[0218] State structure generation: Finally, it is encapsulated into charging execution state data and uploaded to the central platform or edge node for real-time access by smart contract modules.

[0219] This step receives the task allocation and scheduling scheme, generates task issuance instructions with scheduling precision control and V2G capability indicators, executes site-level dynamic power adjustment operations, and collects charging process status information in real time, completing a system-level closed loop from scheduling logic to physical execution. Through standardized encapsulation of charging execution status data, a highly reliable and consistent data input path is established for subsequent smart contract execution and refined billing strategies, realizing an efficient "model-execution-feedback" closed-loop architecture in the system of this invention.

[0220] Step 500 includes at least steps S510-S530:

[0221] S510. Obtain the charging execution status data and the electricity price gradient information, perform dynamic contract billing processing, and generate bill data.

[0222] In this step, the system constructs a dynamic contract billing model based on the charging execution status data collected in S430 (including actual charging duration, average power, and whether V2G power supply is used), and the electricity price gradient information output in S230. We adopt a functional mapping billing structure and introduce higher-order function integrals to construct a differentiated settlement method:

[0223] ①Billing contract functional definition:

[0224]

[0225] Total contract bill value for user m;

[0226] t1, t2: Actual charging start and end times (from execution status data);

[0227] The average power output value of task i at time t;

[0228] λ tElectricity price gradient value corresponding to time t (from S230);

[0229] The rate of change of electricity consumption per unit time for user m;

[0230] V2G power supply indicator (1 if reverse power supply is performed, 0 otherwise);

[0231] β1 and β2 are the charging consumption pricing coefficient and the reverse supply revenue coefficient, respectively, which can be adjusted according to the strategy.

[0232] This function integrates energy consumption during charging, electricity price changes, and reverse power supply into a unified billing expression.

[0233] S520. Generate user incentive strategy data based on the billing data and the task vector dataset.

[0234] After the billing data is generated, the system constructs a user behavior incentive scoring function by combining indicators such as task urgency, appointment response timeliness, and off-peak degree from the task vector dataset generated in S130. The improved Nash equilibrium response function is designed as follows:

[0235] ② User incentive response function:

[0236]

[0237] The incentive score of user m is used to calculate the amount of incentives that can be obtained (such as points, tokens, etc.);

[0238] The timestamp of when user m actually started charging;

[0239] User m's appointment timestamp;

[0240] The duration of time that user m is actually charging during periods of low electricity prices;

[0241] μ total Total charging time for user m;

[0242] U m : The urgency value of user m's task;

[0243] U max The maximum urgency value among all current user tasks, used for normalizing the incentive factor calculation;

[0244] γ1: Response timeliness incentive weighting coefficient;

[0245] γ2: Incentive weighting coefficient for off-peak load regulation behavior;

[0246] γ3: Incentive weighting coefficient for high-priority tasks;

[0247] The incentive score output by this function will be used as the basis for strategy execution, such as incentive token configuration and points accumulation.

[0248] S530. The billing data and the user incentive strategy data are structured and encapsulated to generate a set of smart contract execution results.

[0249] Finally, the billing data generated in S510 and the user incentive strategy data generated in S520 are structurally encapsulated to form a smart contract execution result set. The structure includes:

[0250] ③ Representation of the set of smart contract execution results:

[0251]

[0252] The set of smart contract execution results for user m;

[0253] ID m User task unique identifier;

[0254] Bill amount;

[0255] Incentive rating;

[0256] The actual task start timestamp and the actual task completion timestamp;

[0257] λ t Electricity price level;

[0258] Does reverse power supply behavior exist?

[0259] The encapsulated result will be input into the S610 module for subsequent generation of on-chain record and verification structures.

[0260] By introducing a function and structure contract encapsulation mechanism, this module achieves three functions in dynamic and nonlinear billing scenarios: accurate settlement, behavior guidance, and data structure solidification. Each formula includes execution status data, electricity price gradients, and task vector parameters, ensuring tight coupling between the model and system functions, and possessing practical engineering application value. Within its closed-loop structural logic, the entire S500 module provides reliable core data support for blockchain contract execution and distributed ledger, demonstrating the mathematical modeling innovation and institutional incentive mechanism of this invention in the intelligent billing system for charging piles.

[0261] Step S600 includes at least steps S610-S630:

[0262] S610. Obtain the set of execution results of the smart contract and perform field hashing and encryption parameter encapsulation processing.

[0263] In this step, the system obtains the set of smart contract execution results generated by S530, and performs hash digest generation and encrypted parameter encapsulation on key fields. The specific implementation details are as follows:

[0264] First, extract structured field content from the smart contract execution result set, including bill amount, incentive score, task identifier, execution period, electricity price gradient and V2G status;

[0265] Secondly, a deterministic hash function is used to generate field-level digest values ​​for the above fields, and a corresponding field mapping index is generated;

[0266] Then, the preset key parameter set is called to perform symmetric encryption on the bill amount and task identifier fields, and the encrypted ciphertext and hash digest are encapsulated together to form an intermediate encapsulation structure;

[0267] Finally, the intermediate encapsulation structure is cached in the contract execution log chain as the input data structure for subsequent verification and on-chain operations.

[0268] This process ensures that the billing data and incentive policies are verifiable under the premise of privacy protection, and establishes a two-way mapping channel between on-chain records and off-chain data.

[0269] S620. Based on the field hash and encryption parameter encapsulation result, generate a privacy protection identifier structure for verification.

[0270] This step constructs a verifiable privacy identifier structure based on the intermediate encapsulation structure generated in S610, used for integrity verification of on-chain records and triggering of the zero-knowledge proof mechanism. Specific implementation includes:

[0271] First, extract the hash digest value of each field in the intermediate encapsulation structure;

[0272] Next, the desensitization identifier template library is called to perform template mapping and irreversible masking on the fields containing task execution information and billing information, generating an anonymous identifier mapping table;

[0273] Then, a field-level combined signature structure is generated, and a digest signature operation is performed on the field structure based on the private key module in the smart contract execution environment.

[0274] Finally, the anonymous identifier mapping table and the field signature structure are merged to generate a privacy-protected identifier structure, which will serve as the identifier benchmark for trusted data on the blockchain of smart contracts.

[0275] This identifier structure can be used for on-chain smart contracts to authorize and verify off-chain behavior, trigger states, and assert contract logic, ensuring data security and the legality of operations.

[0276] S630. Package the privacy protection identifier structure with the smart contract execution result set to generate a trusted data on-chain record.

[0277] This step, as the final stage of the S600 process, encapsulates the aforementioned processing results on-chain, ensuring the data is reliably uploaded to the blockchain. The specific operations are as follows:

[0278] First, the privacy protection identifier structure and the smart contract execution result set are aligned by field and packaged into a structure to form a unified contract execution block;

[0279] Next, add on-chain metadata such as on-chain timestamp, node identifier, and consensus digest to the contract execution block;

[0280] Then, the lightweight blockchain node interface bound to the smart contract is invoked to broadcast the contract execution block to the blockchain network;

[0281] Finally, after the on-chain record is formed, its on-chain record number and block hash value are fed back to the contract execution log chain, forming a closed-loop authentication system that links on-chain and off-chain operations.

[0282] This process ensures the reliable storage of smart contract execution data, providing a solid foundation for subsequent auditing, traceability, and compliance verification.

[0283] This module introduces field-level hash digest calculation, symmetric encryption, template de-identification mapping, and a zero-knowledge structure signature mechanism to construct a highly verifiable data encapsulation and on-chain solution that balances privacy protection. Through field alignment and contract structure packaging mechanisms, it ensures consistency and data integrity both on-chain and off-chain, enhancing the credibility, security, and compliance of smart contract execution records in multi-party collaborative operation scenarios. This comprehensively guarantees the commercial viability and regulatory compatibility of this invention patent.

[0284] Example 2: Figure 2 A structural block diagram of a charging pile operation and management system according to an embodiment of the present invention is shown. Figure 2 As shown, the structure may include:

[0285] Task generation module 10 is used to generate a set of task vectors based on charging user request data and grid load data, and to perform cleaning processing. This module mainly includes:

[0286] The data parsing submodule is used to perform structured parsing of user requests, power grid status, and site equipment information from charging terminals.

[0287] Feature extraction submodule: Extracts key scheduling features such as charging power, reservation duration, and task urgency;

[0288] The data cleaning submodule performs format standardization, anomaly removal, and timestamp alignment on the task vector set to generate a schedulable task vector dataset.

[0289] This module forms the basic input source for subsequent load forecasting, price modeling, and resource scheduling.

[0290] Prediction module 20, which generates charging demand prediction data and electricity price gradient information based on the cleaned task vector dataset and grid load data, includes:

[0291] The load forecasting model construction submodule calls time series algorithms to build a short-term regional load forecasting model;

[0292] Model inference submodule: Performs inference based on cleaned task vector data and historical load data to predict dynamic grid load in future time periods;

[0293] Electricity price gradient generation submodule: Combines prediction results with task requirements to generate real-time electricity price gradient curves for different time periods and regions.

[0294] This module is used in the system to adjust the pricing mechanism in advance and guide users to distribute their load.

[0295] Scheduling module 30 is used to construct and solve a set of multi-objective optimization scheduling functions to generate priority-ordered task allocation and scheduling schemes. It includes:

[0296] The optimization function construction submodule constructs a set of scheduling targets that simultaneously consider factors such as user costs, grid peak and valley loads, and equipment utilization.

[0297] The multi-objective solution submodule calls the improved non-dominated sorting algorithm to solve the optimization function and obtain the initial task scheduling results;

[0298] Priority Adjustment Submodule: Combining task urgency and predicted load, it sorts and adjusts task priorities based on the initial scheduling results to form the final scheduling plan.

[0299] This module enables dynamic matching of resources and objectives and is the core optimization module of the system.

[0300] Execution module 40, which is used to complete task issuance, power control, and collection of charging execution status data, mainly includes:

[0301] Instruction generation submodule: Converts task scheduling schemes into task issuance instructions that can be recognized by the device;

[0302] Dynamic power regulation submodule: Based on real-time device status and V2G capability, it performs power regulation and bidirectional power supply switching control;

[0303] Status Acquisition Submodule: Collects status data such as output power, duration, voltage and current fluctuations of the charging station during task execution.

[0304] This module enables precise control and feedback acquisition of the physical charging process, connecting virtual scheduling and physical execution.

[0305] Contract module 50 is used to perform dynamic contract billing based on charging execution status data and generate a set of smart contract execution results. It mainly includes:

[0306] Dynamic billing submodule: Matches the execution status data with the electricity price gradient to achieve billing based on power and time period;

[0307] Incentive generation submodule: Combines billing data with factors such as task urgency to generate differentiated user incentive strategies;

[0308] Contract encapsulation submodule: Structures and packages billing data and incentive strategy data to generate a set of smart contract execution results.

[0309] This module implements a closed-loop economic incentive mechanism both on-chain and off-chain, and is the core of the contract-driven mechanism of this invention.

[0310] On-chain module 60 is used to perform hash encapsulation and privacy processing on the smart contract execution result set, generating trusted data on-chain records. It mainly includes:

[0311] Hash Encryption Encapsulation Submodule: Encrypts and hashes sensitive fields such as invoices and task identifiers, and generates data digests.

[0312] Privacy identifier generation submodule: Constructs a data structure that can be used for verification, supporting subsequent auditing and cross-entity verification;

[0313] On-chain encapsulation sub-module: Packages the privacy identifier structure and the smart contract execution result set onto the blockchain to form an immutable data record.

[0314] This module ensures trustworthy collaboration among all parties and supports legitimate and compliant interactions between multiple entities.

[0315] The system described in this invention constructs a collaborative mechanism for the entire operation process of charging piles, centered on task-driven mechanisms, employing multi-objective optimization, utilizing smart contracts as rules, and secured by blockchain technology. It possesses the following characteristics:

[0316] Beneficial effects:

[0317] (1) Achieve precise matching and efficient allocation of resources: By constructing a linkage mechanism between task vectors and real-time load models, the system effectively alleviates resource conflicts during peak charging periods and improves equipment utilization and grid operation stability.

[0318] (2) Support for multi-objective collaborative scheduling optimization: By introducing NSGA-II, weighted priority queue and scheduling function system, the system achieves the optimal balance between grid pressure relief, user waiting time control and cost control, thereby improving the overall operating economy.

[0319] (3) Ensuring data trustworthiness and privacy security: The system constructs a data on-chain structure based on lightweight blockchain and zero-knowledge encryption algorithm to ensure that all execution records are verifiable, immutable and privacy protected, meeting the collaborative compliance requirements of multiple participants in power trading and charging services.

[0320] (4) Automatic contract-driven operation closed loop: Through dynamic electricity price incentive mechanism and smart contract structure, an automatic response mechanism for charging behavior, supply and demand adjustment and economic revenue sharing is realized, which improves the automation level of platform operation and user response flexibility.

[0321] (5) It can be extended to city-level energy management systems: The system structure has good modularity, service-oriented and chain-encapsulation capabilities, and can be used as an important component of regional smart energy platforms or vehicle-to-grid collaborative systems, with broad commercial application prospects.

[0322] Obviously, the embodiments described above are only some embodiments of this application, not all embodiments. The accompanying drawings show preferred embodiments of this application, but do not limit the patent scope of this application. This application can be implemented in many different forms; rather, the purpose of providing these embodiments is to provide a more thorough and comprehensive understanding of the disclosure of this application. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing specific embodiments, or make equivalent substitutions for some of the technical features. Any equivalent structures made using the content of this application's specification and drawings, directly or indirectly applied to other related technical fields, are similarly within the scope of patent protection of this application.

Claims

1. A method for operating and managing charging piles, characterized in that, Includes the following steps: Based on the charging request data of charging users and the power grid load data, a cleaned task vector dataset is obtained. Based on the cleaned task vector dataset and power grid load data, charging demand is predicted and electricity price gradient information is generated. Obtain the electricity price gradient information and the task vector dataset, construct a set of multi-objective optimization scheduling functions, and generate a priority-ranked task allocation and scheduling scheme. The construction of a multi-objective optimization scheduling function set includes: Define the functional: ； in, Optimize the scheduling functional for multiple objectives; A set of scheduling variables for task-device configuration; These represent the weight parameters for the three optimization objectives; These represent the electricity price cost integral interval, the grid load integral interval, and the emergency dispatch interval, respectively. Represents the electricity price dispatch cost function; This represents the power grid load disturbance term; This represents the weighted scheduling index; among which, For at a certain point in time The unit electricity price; For the task In time The duration allocation; This is the actual load value; Average load; For spatial gradient operators; Indicates the magnitude of the vector; This represents the total number of schedulable tasks in the current scheduling window. For the task The scheduling variable; For the task The urgency value; A multi-objective shortest path mapping function is used for searching, and task-device combinations are sorted from high to low based on a priority scoring function; the expression for the priority scoring function is: ； in, For the task With equipment Scheduling priority score; This is a multi-factor priority scoring function; This refers to the priority weighting coefficient; For time The unit electricity price; For the site The load gradient; This represents the maximum value of the load gradient within the scheduling time interval; For the task The degree of urgency; The maximum urgency level within the current scheduling cycle; The task allocation and scheduling scheme is as follows: ； in, The final output scheduling scheme; Assign task-device pairs; To indicate that the top scores are selected from all the rating results The highest-rated pair of assignments; Scoring is assigned to scheduling priorities; The task is assigned and dispatched according to the task allocation and scheduling scheme, power control is executed and charging execution status data is collected; Based on the charging execution status data, dynamic contract billing and incentive generation are performed to form a set of smart contract execution results; The execution of dynamic contract billing includes using a functional mapping billing structure to generate billing data, the expression of which is: ； in, For users Total value of the contract bill; This refers to the actual start and end times of charging. For the task In time The average power output value; For time The corresponding electricity price gradient value; For users The rate of change of electrical charge per unit time; V2G power supply indicator; These are the charging consumption pricing coefficient and the reverse supply revenue coefficient, respectively. After the billing data is generated, a user behavior incentive scoring function is constructed by combining the task urgency, appointment response timeliness, and off-peak period degree from the task vector dataset: ； in, The incentive rating result for user m; For users The actual timestamp when charging started; The appointment timestamp for user m; For users The duration of the actual charging process during periods of low electricity prices; For users Total charging time; The task urgency value for user m; This represents the maximum urgency value among all current user tasks. The incentive weighting coefficient is based on the timeliness of response; The incentive weighting coefficient for off-peak load regulation behavior; The incentive weight coefficient is assigned to high-priority tasks; The set of smart contract execution results represents: ； in, Let m be the set of smart contract execution results for user m. A unique identifier for the user task; This is the bill amount; To incentivize ratings; These are the actual task start timestamp and the actual task completion timestamp; Electricity price level; To determine whether reverse power supply behavior exists; The results of smart contract execution are hashed and processed for privacy, generating trusted data records on the blockchain.

2. The method according to claim 1, characterized in that, Based on charging user request data and grid load data, a cleaned task vector dataset is obtained, which specifically includes: The system acquires charging request data from charging users, grid load data, and site equipment status information, and performs structural analysis and data time alignment processing. Power demand, duration, and task urgency are extracted from charging request data and grid load data to generate a set of task vectors; The task vector set is formatted and outlier removed to obtain a cleaned task vector dataset.

3. The method according to claim 1, characterized in that, The process of predicting charging demand and generating electricity price gradient information specifically includes: Obtain the cleaned task vector dataset and power grid load data; The time series model is trained and inferred to generate short-term charging demand forecast data. Based on the charging demand forecast data and the cleaned task vector dataset, the electricity price gradient information for the corresponding time period is generated.

4. The method according to claim 1, characterized in that, Construct a set of multi-objective optimization scheduling functions, specifically including: Obtain electricity price gradient information and a cleaned task vector dataset, and construct a set of multi-objective optimization scheduling functions; The initial task scheduling results are obtained by solving the set of multi-objective optimization scheduling functions.

5. The method according to claim 1, characterized in that, The priority-based task allocation and scheduling scheme specifically includes: Based on the initial task scheduling results and the cleaned task vector dataset, a priority-ranked task allocation and scheduling scheme is generated.

6. The method according to claim 1, characterized in that, The generation of trusted data on-chain records specifically includes: Obtain the set of smart contract execution results and encapsulate them using field hashing and encrypted parameters; Based on the field hash and encryption parameter encapsulation results, a privacy-protected identifier structure for verification is generated; The privacy protection identifier structure is packaged with the smart contract execution result set to generate a trusted data record on the blockchain.

7. A charging pile operation and management system, applied to the charging pile operation and management method according to any one of claims 1-6, characterized in that, include: The task generation module is used to generate a set of task vectors based on the charging request data of charging users and the grid load data, and to complete the cleaning process. The prediction module is used to generate charging demand prediction data and electricity price gradient information based on the cleaned task vector dataset and grid load data; The scheduling module is used to construct and solve a set of multi-objective optimization scheduling functions to generate priority-ordered task allocation and scheduling schemes. The execution module is used to issue tasks, perform power control, and collect charging execution status data. The contract module is used to perform dynamic contract billing based on charging execution status data and generate a set of smart contract execution results. The on-chain module is used to perform hash encapsulation and privacy processing on the smart contract execution result set, generating trusted data on-chain records.

Images