Intelligent scheduling and queuing optimization system for charging piles based on big data analysis

Abstract

This invention belongs to the technical field of resource scheduling and management, and relates to an intelligent scheduling and queuing optimization system for charging piles based on big data analysis. The system includes: a state parameter aggregation module for analyzing sensor and electrical signals to establish a dynamic state parameter set; a predictive index calculation module for performing time-series extrapolation to generate time index pairs containing expected arrival and predicted release times; a time-series constraint filtering module for establishing exclusive serviceable time windows and performing verification to filter target pair sets; a global scheduling optimization module for establishing the globally optimal allocation using a combinatorial optimization algorithm with the total predicted waiting time as the target; a resource locking control module for issuing pre-occupancy signals to achieve logical locking of physical resources in the controller; and a closed-loop execution monitoring module for converting into a navigation path and monitoring deviation magnitude to achieve dynamic adjustment. This invention solves the problem of improving matching efficiency and shortening waiting time by utilizing dynamic operating parameters and a global perspective to coordinate resources.

Description

Technical Field

[0001] This invention belongs to the technical field of resource scheduling and management, and relates to a smart scheduling and queuing optimization system for charging piles based on big data analysis. Background Technology

[0002] With the increasing number of electric vehicles, the utilization value and user experience of charging infrastructure have become key technological focuses for the industry. Currently, charging demand exhibits dynamic changes in both time and space, while the service capacity of fixed charging facilities is not well-matched with this dynamic demand. This phenomenon can lead to long waiting times for vehicles at some locations, while equipment at other locations remains idle, indicating that the overall utilization rate of various charging resources needs further improvement.

[0003] To address the above situation, existing industry solutions primarily rely on information dissemination and a first-come, first-served matching model. A common approach is to use an application to display the real-time occupancy status of nearby charging stations, indicating whether they are available or in use. Users can then make decisions based on these static indicators and navigate to their desired station independently. Furthermore, some technical solutions have added basic reservation control logic, allowing users to lock in available stations for a fixed period via commands, thus replacing the traditional station-finding process that lacks auxiliary information.

[0004] Because the existing technologies mentioned above are mainly based on static state display and fixed-duration reservation, they exhibit certain limitations when applied in variable traffic environments and scenarios with concurrent charging requests. On the one hand, solutions relying on the current real-time state are prone to state lag due to dynamic changes. When a user goes to the charging station based on the current availability, the station may already be occupied by a closer vehicle, increasing the vehicle's search cost. Furthermore, the lack of estimation and reference for the remaining service time of currently charging vehicles limits the dimensions of scheduling considerations. On the other hand, reservation functions using fixed-duration reservations often do not adaptively adjust to changes in the actual travel distance and real-time road network conditions, making them susceptible to traffic delays that could invalidate reservations, or excessively long reservation windows that could reduce the actual turnover rate of the charging stations. This type of discrete decision-making model based on the user's local information lacks a holistic resource coordination mechanism from a global perspective, and there is room for optimization in improving the overall collaboration and operational efficiency of large-scale networks. Summary of the Invention

[0005] To address the aforementioned problems, this invention provides a smart scheduling and queuing optimization system for charging piles based on big data analysis.

[0006] A charging pile intelligent scheduling and queuing optimization system based on big data analysis includes:

[0007] The state parameter aggregation module is used to analyze vehicle-mounted sensor signals and pile-end electrical signals to establish a dynamic state parameter set for the vehicle to be charged and the candidate charging piles.

[0008] The predictive index calculation module performs time-series extrapolation calculations based on a dynamic state parameter set to generate a predictive time index pair for each vehicle and charging pile pair. The predictive time index pair includes the expected arrival time and the predicted release time.

[0009] The timing constraint filtering module is used to establish an exclusive serviceable time window, perform timing logic verification based on the predicted time indicators, and filter the target pair set that meets the timing constraints.

[0010] The global scheduling optimization module takes the total predicted waiting time of vehicles waiting to be charged as the global optimization target. It uses a combinatorial optimization algorithm to perform global matching and optimization in the target pairing set, determines the target charging pile code corresponding to each vehicle waiting to be charged, and generates a global optimization scheduling instruction.

[0011] The resource locking control module sends a pre-occupancy signal to the target charging pile according to the global optimization scheduling instruction. It sets the status to the pre-occupancy flag in the local control register of the target charging pile controller and blocks all charging request attempts except for the identity of the vehicle to be charged during the duration of the pre-occupancy flag, thus completing the logical locking of the physical charging resources.

[0012] The closed-loop execution monitoring module is used to convert global optimization scheduling instructions into closed-loop directional navigation paths and monitor the execution status in real time. If the deviation of the vehicle to be charged on the closed-loop directional navigation path exceeds a preset threshold, a cancellation signal is sent to the target charging pile to release the pre-occupancy flag, thereby realizing the dynamic closed-loop adjustment of the scheduling scheme.

[0013] A further aspect of the present invention includes a state parameter aggregation module, which performs the following operations:

[0014] The system acquires the GPS coordinate sequence, battery state of charge percentage, and battery rated capacity of the vehicle to be charged, and converts the GPS coordinate sequence into a vehicle displacement vector.

[0015] Simultaneously acquire the real-time working status indicator bit, rated charging power level, and the accumulated charging amount pulse signal of the vehicle currently being charged for each candidate charging pile;

[0016] By aggregating vehicle displacement vector, battery state of charge percentage, real-time operating status indicator bit, and accumulated charge pulse signal, a dynamic state parameter set containing spatiotemporal characteristics and energy state characteristics is generated.

[0017] A further aspect of the present invention includes a predictive index calculation module, which performs the following operations:

[0018] Based on the vehicle displacement vector in the dynamic state parameter set and combined with the real-time road network congestion coefficient, the estimated arrival time of the vehicle to be charged to each candidate charging station is calculated.

[0019] For candidate charging piles that are currently occupied, the predicted release time of the candidate charging pile is determined by extrapolating the remaining completion time of the current charging task based on its corresponding rated charging power level and the pulse signal of the accumulated charging amount of the vehicle currently being charged.

[0020] By mapping and associating the expected arrival time with the predicted release time, a pair of predicted time indicators reflecting the intersection of the vehicle movement process and the charging physical process is generated.

[0021] A further aspect of the present invention includes a timing constraint filtering module, which performs the following operations:

[0022] Based on the predicted release time in the predicted time index pair, and combined with the preset traffic fluctuation compensation coefficient, the start and end values ​​of the exclusive service time window for each candidate charging pile are calculated.

[0023] The estimated arrival time is quantitatively compared with the exclusive available service time window to determine whether the estimated arrival time falls within the corresponding exclusive available service time window.

[0024] The vehicle and charging pile combinations that fall within the time window are retained and marked as target pairs with temporal feasibility, forming a target pair set.

[0025] A further aspect of this invention includes a global scheduling optimization module, which performs the following operations:

[0026] Traverse the target pairing set and extract the predicted waiting time for each target pairing. The predicted waiting time is obtained by taking the absolute value of the difference between the predicted release time and the expected arrival time. The absolute value is used to comprehensively characterize the vehicle queuing time and the charging pile idle time.

[0027] Using the total predicted waiting time of vehicles waiting to be charged as the global optimization objective, a combinatorial optimization algorithm is used to perform global matching and optimization on the target pairing set to determine the target charging pile code corresponding to each vehicle waiting to be charged.

[0028] The target charging pile code, estimated arrival time, and predicted release time are encapsulated to generate a clearly targeted global optimization scheduling instruction.

[0029] A further aspect of the present invention includes a resource locking control module, used to perform the following operations:

[0030] The target charging pile code in the global optimization scheduling instruction is sent to the corresponding target charging pile controller via a wireless communication link.

[0031] After receiving the instruction, the target charging pile controller sets the status position to the pre-occupancy flag in the local control register and simultaneously shuts down the real-time status feedback channel of the charging pile to the external public query interface.

[0032] During the period of the pre-occupancy flag, all charging request attempts except for the identity identifier of the vehicle to be charged are blocked, thus completing the logical locking of the physical charging resources.

[0033] A further aspect of the present invention includes a closed-loop execution monitoring module, used to perform the following operations:

[0034] A scheduling confirmation request containing the location of the target charging station and the predicted waiting time is pushed to the user terminal of the vehicle waiting to be charged.

[0035] After obtaining the user's confirmation signal, the location coordinates of the target charging pile are hard-bound to the visual navigation system of the vehicle to be charged, generating a closed-loop directional navigation path;

[0036] The system continuously monitors the real-time trajectory of the vehicle waiting to be charged on the closed-loop directional navigation path. If the path deviation exceeds a preset threshold, a cancellation signal is sent to the target charging pile to release the pre-occupancy flag.

[0037] A further aspect of the present invention involves converting a Global Positioning System (GPS) coordinate sequence into a vehicle displacement vector, comprising the following steps:

[0038] Extract the coordinate points corresponding to two adjacent timestamps from the GPS coordinate sequence;

[0039] After projecting from spherical coordinates to local planar coordinates, the vector difference between the two points is calculated to generate the vehicle displacement vector that characterizes the instantaneous motion state of the vehicle.

[0040] A further aspect of the present invention involves calculating the exclusivity service window cutoff value for candidate charging piles, including the following steps:

[0041] Calculate the time difference between the vehicle's estimated arrival time and the current system time as the estimated travel time;

[0042] The exclusive service window cutoff value is obtained by multiplying the predicted release time by the expected travel time by the preset traffic fluctuation compensation coefficient.

[0043] A further aspect of the present invention, upon detecting a path deviation exceeding a preset threshold, includes the following steps:

[0044] Extract the real-time trajectory location coordinates of the vehicle to be charged, construct a straight line equation based on the coordinates of the two endpoints of the nearest path segment on the closed-loop directional navigation path, and calculate the geometric vertical distance from the location coordinates to the straight line as the path deviation magnitude.

[0045] The path deviation is compared with a preset threshold set according to the road type. If the path deviation exceeds the preset threshold for a certain period of time, it is determined that the path deviation exceeds the preset threshold.

[0046] In summary, the present invention has the following beneficial technical effects:

[0047] 1. This invention utilizes multi-source real-time data to establish a model of estimated arrival time and predicted release time, expanding the reference basis for scheduling decisions from the current state of a single resource to future spatiotemporal connection points. This mechanism allows the system to identify and predict charging resources that are occupied but about to be released, providing vehicles with a time lead time, helping to reduce blind waiting after vehicle arrival, and improving the utilization efficiency of charging resources within continuous time periods.

[0048] 2. Based on timing logic verification, this invention performs global matching optimization with the goal of minimizing the total predicted waiting time of the entire system. Compared to local decisions made independently by individual users, this centralized assignment mode can comprehensively balance the needs of multiple parties and resource distribution, helping to alleviate the queuing pressure caused by concurrent requests at some charging stations, dynamically adjusting the resource occupancy ratio between different geographical nodes, thereby playing a role in peak shaving and valley filling and reducing the overall average waiting time at the system level.

[0049] 3. This invention employs a pre-reservation signal to logically lock physical resources, supplemented by real-time monitoring of directional navigation trajectories. This closed-loop control logic reserves a defined service window for vehicles executing scheduling commands, reducing the uncertainty of resources being unexpectedly occupied during transit. Simultaneously, an automatic release mechanism triggered by path deviation ensures that locked resources are promptly returned to the public resource pool when the user abandons execution or an abnormal deviation occurs, mitigating the risk of resource idleness due to invalid reservations. Attached Figure Description

[0050] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below. The drawings are used to provide a further understanding of the present invention.

[0051] Figure 1 This is a schematic diagram of the framework in the embodiments of this application.

[0052] Figure 2 This is a flowchart illustrating an embodiment of this application. Detailed Implementation

[0053] The following is in conjunction with the appendix Figures 1-2 A preferred description of the present invention is provided below.

[0054] See attached document Figures 1-2 This invention proposes a smart scheduling and queuing optimization system for charging piles based on big data analysis, which includes the following modules:

[0055] The state parameter aggregation module is used to analyze vehicle-mounted sensor signals and pile-end electrical signals to establish a dynamic state parameter set for the vehicle to be charged and the candidate charging piles.

[0056] The predictive index calculation module performs time-series extrapolation calculations based on a dynamic state parameter set to generate a predictive time index pair for each vehicle and charging pile pair. The predictive time index pair includes the expected arrival time and the predicted release time.

[0057] The timing constraint filtering module is used to establish an exclusive serviceable time window, perform timing logic verification based on the predicted time indicators, and filter the target pair set that meets the timing constraints.

[0058] The global scheduling optimization module takes the total predicted waiting time of vehicles waiting to be charged as the global optimization target. It uses a combinatorial optimization algorithm to perform global matching and optimization in the target pairing set, determines the target charging pile code corresponding to each vehicle waiting to be charged, and generates a global optimization scheduling instruction.

[0059] The resource locking control module sends a pre-occupancy signal to the target charging pile according to the global optimization scheduling instruction. It sets the status to the pre-occupancy flag in the local control register of the target charging pile controller and blocks all charging request attempts except for the identity of the vehicle to be charged during the duration of the pre-occupancy flag, thus completing the logical locking of the physical charging resources.

[0060] The closed-loop execution monitoring module is used to convert global optimization scheduling instructions into closed-loop directional navigation paths and monitor the execution status in real time. If the deviation of the vehicle to be charged on the closed-loop directional navigation path exceeds a preset threshold, a cancellation signal is sent to the target charging pile to release the pre-occupancy flag, thereby realizing the dynamic closed-loop adjustment of the scheduling scheme.

[0061] In one embodiment of the present invention, the state parameter aggregation module is configured to perform the following steps:

[0062] The system acquires the GPS coordinate sequence, battery state of charge percentage, and battery rated capacity of the vehicle to be charged, and converts the GPS coordinate sequence into a vehicle displacement vector. Simultaneously, it acquires the real-time operating status indicator, rated charging power level, and accumulated charging amount pulse signal of each candidate charging pile. The system aggregates the vehicle displacement vector, battery state of charge percentage, real-time operating status indicator, and accumulated charging amount pulse signal to generate a dynamic state parameter set containing spatiotemporal and energy state characteristics.

[0063] Specifically, a central dispatch system deployed on a cloud server acquires multi-source heterogeneous real-time data streams and establishes a dynamic state parameter set for the vehicle to be charged and the candidate charging pile by analyzing vehicle-mounted sensor signals and charging pile electrical signals. The central dispatch system first establishes a long connection with the communication terminal of the vehicle to be charged registered in the system via the MQTT protocol (Message Queuing Telemetry Transmission Protocol), receiving data frames reported in real-time by the vehicle-mounted data unit built into the vehicle in a subscription manner. The vehicle-mounted data unit collects and encapsulates the GPS coordinate sequence, battery state of charge percentage, and battery rated capacity at a preset sampling frequency, such as once per second. It should be noted that the GPS coordinate sequence is a series of latitude and longitude coordinates with timestamps output by the vehicle-mounted GPS module. In one embodiment of the invention, to eliminate positioning errors caused by GPS signal drift, this embodiment uses a Kalman filter algorithm to smooth the original coordinates, with a position weight coefficient set to 0.7-0.85 to ensure the stability of displacement calculation. The battery state-of-charge percentage is a floating-point number ranging from 0.0 to 100.0, representing the ratio of the battery's current remaining charge to its total capacity.

[0064] After receiving the coordinate sequence, the central dispatch system extracts the coordinate points corresponding to the two most recent timestamps. Through a projection transformation from spherical coordinates to local planar coordinates, a Gauss-Kruger projection transformation is used to convert the latitude and longitude spherical coordinates to local planar Cartesian coordinates. The vector difference between the two points is calculated to generate the vehicle displacement vector, and the magnitude of this displacement vector is divided by the corresponding timestamp difference. This allows us to obtain the instantaneous speed, which characterizes the vehicle's instantaneous motion state. In this embodiment, the vehicle displacement vector is a two-dimensional vector, where its magnitude represents the displacement and its direction represents the direction of travel. The velocity dimension extracted from the time difference will be directly used for subsequent calculations of the estimated arrival time. Its transient displacement calculation can be expressed by the following formula. Assuming the vehicle is in a Cartesian coordinate system at time... The position coordinates are At any moment The position coordinates are Then the vehicle displacement vector The calculation formula is:

[0065]

[0066] Meanwhile, a data acquisition thread in the central dispatch system initiates data requests via HTTPS (Hypertext Transfer Security Protocol) to the data acquisition modules at the charging piles of a group of candidate charging piles geographically adjacent to the vehicle to be charged. The charging pile data acquisition modules respond to the requests, returning a data packet containing their real-time operating status indicator, rated charging power level, and a pulse signal indicating the accumulated charge amount for the current vehicle if it is charging. The real-time operating status indicator is an integer value; for example, 0 represents idle, 1 represents charging, 2 represents a fault, and 3 represents a reserved charging point. The rated charging power level is a parameter specifying the maximum output power of the charging pile, for example, a value in kW, such as 7, 50, or 150. The accumulated charge amount pulse signal is a count value accumulated since the start of the current charging task; the physical amount of electricity represented by it needs to be multiplied by a preset pulse equivalent coefficient, for example, 0.01 kWh / pulse.

[0067] Finally, the data aggregation module of the central dispatch system correlates and matches the vehicle displacement vector and battery state of charge percentage from the same vehicle to be charged with the real-time operating status indicator and accumulated charge pulse signal from each candidate charging station, generating one or more data structures stored in key-value pairs. This data structure is a dynamic state parameter set containing spatiotemporal and energy state characteristics, and is stored in the real-time database for subsequent steps. It should be understood that the dynamic state parameter set is a structured data object containing at least the following fields: vehicle ID, charging station ID, timestamp, vehicle displacement vector, battery state of charge percentage, real-time operating status indicator, rated charging power level, and accumulated charge pulse signal.

[0068] For example, suppose at time t=10:00:01, the central dispatch system receives data reported by vehicle V001, whose GPS coordinates are (116.4074°E, 39.9042°N), battery state of charge percentage is 25.5%, and battery rated capacity is 80 kWh. The next second, at t=10:00:02, it receives its new GPS coordinates (116.4078°E, 39.9045°N). The system first converts these two sets of latitude and longitude coordinates into local plane coordinates, obtaining... and Assuming the conversion The unit is meters. The unit is meters (m). Next, the vehicle displacement vector is calculated as ((407.8-407.4), (904.5-904.2)), which is (0.4, 0.3) m. Simultaneously, data requests are sent to nearby candidate charging stations C005 and C006. C005 returns a real-time operating status indicator of 1, with a rated charging power level of 50 kW and 1500 accumulated charging pulse signals. C006 returns a real-time operating status indicator of 0, with a rated charging power level of 150 kW and 0 accumulated charging pulse signals. Finally, two sets of dynamic state parameters are generated. The first parameter set associates V001 and C005, containing the data {vehicle_id:"V001", pile_id:"C005", timestamp:10:00:02, displacement_vector:(0.4, 0.3), soc:25.5, pile_status:1, pile_power:50, pulse_count:1500}. The second parameter set associates V001 and C006, containing the data {vehicle_id:"V001", pile_id:"C006", timestamp:10:00:02, displacement_vector:(0.4, 0.3), soc:25.5, pile_status:0, pile_power:150, pulse_count:0}.

[0069] In one embodiment of the present invention, the predictive index calculation module is configured to perform the following steps:

[0070] Based on the vehicle displacement vector in the dynamic state parameter set and combined with the real-time road network congestion coefficient, the estimated arrival time of the vehicle to be charged to each candidate charging pile is calculated. For candidate charging piles that are occupied, the remaining completion time of the current charging task is extrapolated based on the corresponding rated charging power level and the pulse signal of the accumulated charging amount of the currently charging vehicle to determine the predicted release time of the candidate charging pile. The estimated arrival time and the predicted release time are mapped and correlated to generate a predicted time index pair that reflects the intersection of the vehicle movement process and the charging physical process.

[0071] Specifically, the physical process trend prediction module within the central dispatch system receives and processes the generated dynamic state parameter set to perform time-series extrapolation calculations and generate predicted time index pairs for each vehicle-charging pile pair. First, for each dynamic state parameter set, the module extracts the current geographical coordinates of the vehicle to be charged and the geographical coordinates of the candidate charging piles. The current coordinates of the vehicle to be charged are directly obtained from the dynamic state parameter set, while the coordinates of the candidate charging piles are retrieved from the system's pre-set charging station facility database. The module obtains the navigation path planning distance between the two by calling an external map service application interface, inputting the start and end coordinates, and simultaneously obtaining the real-time road network congestion coefficient along the path. Combining the current driving speed calculated from the vehicle displacement vector in the dynamic state parameter set, the module calculates the estimated arrival time of the vehicle to be charged to each candidate charging pile. In this embodiment, the estimated arrival time... The calculation formula is as follows:

[0072]

[0073] In the formula, This is the current system timestamp. This refers to the navigation distance from the vehicle to the charging station, obtained through the map service interface. The average driving speed is determined based on a combination of road speed limits and vehicle displacement vectors, for example, set to 40 km / h. The real-time road network congestion coefficient is a floating-point number between 0.1 and 1.0. A smaller value indicates more severe congestion. This coefficient is provided by a third-party traffic information platform, and is set in stages based on the ratio of the average vehicle speed to the design speed limit: when the speed ratio is ≥0.8, A value of 0.9-1.0 is acceptable, representing smooth traffic flow; when the speed ratio is between 0.4-0.8, A value of 0.5-0.8 can be used to represent slow movement; when the speed ratio is <0.4, A value of 0.2-0.4 represents congestion.

[0074] Secondly, for candidate charging piles whose real-time working status indicator is "occupied" in the dynamic status parameter set, the module performs extrapolation calculations of the charging process based on its corresponding rated charging power level and the accumulated charging amount pulse signal. The module queries the vehicle battery characteristic database for the total battery capacity and typical charging curve of the vehicle currently being charged, and sets a target state of charge, such as 80%, to determine the total target charging amount. Using the difference between the accumulated charging amount and the total target charging amount, and the rated charging power level, the module calculates the remaining completion time of the current charging task and adds it to the current system time to obtain the predicted release time of the candidate charging pile. When a charging pile is occupied, the calculation formula is:

[0075]

[0076] In the formula, The target charging energy for vehicles currently occupying the charging station is set based on the historical average value of charging the vehicle model to 80% capacity, for example, 50 kWh. This refers to the accumulated charge pulse signal extracted from the dynamic state parameter set. This is the pulse equivalent coefficient, whose value has been defined in the state parameter aggregation module. This refers to the rated charging power level extracted from the dynamic state parameter set.

[0077] For candidate charging stations that are in an idle state, their predicted release time is the current system time. Finally, the module combines the estimated arrival time and predicted release time calculated for the same pair of vehicles and charging stations, encapsulating them into a data tuple. This tuple reflects the intersection of the vehicle's movement and the physical charging process, and is then output to the next processing stage. It should be noted that the predicted time tuple is a tuple containing two timestamp elements. , representing the expected arrival time and the predicted release time, respectively.

[0078] For example, continuing from the previous embodiment, the central dispatch system in At 10:00:02, the two received dynamic state parameter sets are processed. First, the pairing of V001 and C005 is processed, and the navigation distance between the current position of V001 and C005 is retrieved from the database. The distance is 5 km. The traffic data interface is called to obtain the real-time road network congestion coefficient for this route. It is 0.6. Assuming average driving speed... If the speed is 40 km / h, the travel time is This is approximately 12 minutes and 30 seconds. Therefore, the estimated arrival time is... The time is 10:00:02 + 12 minutes 30 seconds = 10:12:32. Since C005 is in an occupied state, the system query reveals the target charging energy for the currently charging vehicle. The value is 50 kWh, and its accumulated charge pulse signal is known. The pulse equivalent coefficient is 1500. 0.01 kWh / pulse, rated charging power The power is 50 kW. Therefore, the remaining completion time is calculated as follows: Therefore, the predicted release time is 10:00:02 + 42 minutes = 10:42:02. The predicted time index pair for V001-C005 is generated as (10:12:32, 10:42:02). Next, the pairing of V001 and C006 is processed, and the navigation distance is found to be 6 km, with a real-time road network congestion coefficient of 0.9. The travel time is calculated as follows: The estimated arrival time is 10:00:02 + 10 minutes = 10:10:02. Since C006 is in an idle state, its predicted release time is the current time 10:00:02. The predicted time index pair for V001-C006 is generated as (10:10:02, 10:00:02). Finally, these two predicted time index pairs are output for subsequent timing logic verification.

[0079] In one embodiment of the present invention, the time-series constraint filtering module is used to perform the following steps:

[0080] Based on the predicted release time in the predicted time index pair, and combined with the preset traffic fluctuation compensation coefficient, the start and end values ​​of the exclusive service time window for each candidate charging pile are calculated. The expected arrival time is quantitatively compared with the exclusive service time window to determine whether the expected arrival time falls within the corresponding exclusive service time window. The vehicle and charging pile combinations that fall within the time window are retained and marked as target pairs with temporal feasibility, forming a target pair set.

[0081] Specifically, the timing logic verification module within the central dispatch system is responsible for receiving predicted time indicator pairs and establishing exclusive serviceable time windows based on these indicator pairs. It then uses timing logic verification to filter out a set of target pairs that meet the timing constraints. For each received predicted time indicator pair, the module first calculates the cutoff value of the corresponding candidate charging pile's exclusive serviceable time window based on the predicted release time and a preset traffic fluctuation compensation coefficient. In this embodiment, the cutoff value of the exclusive serviceable time window... The calculation formula is as follows:

[0082]

[0083] in, This is the traffic fluctuation compensation coefficient. In this embodiment, It is not a fixed value, but rather dynamically adjusted based on the route environment: if the route includes highway sections, Set to 0.15 to reflect the stability of travel time; if the route is entirely urban center roads, The value will be increased to 0.35 to compensate for the uncertainty of traffic lights and random congestion.

[0084] The starting value of this time window is set to the predicted release time itself. In other words, the exclusive serviceability time window is a time interval. In physical terms, it refers to a valid service time reserved for a designated vehicle. The starting value of the exclusive service time window is the predicted release time. Exclusivity service availability window cutoff value This refers to the latest time after a charging station is released, at which point the charging station can be reserved for a designated vehicle. The calculation logic for this cutoff value is to add a buffer time determined by the traffic fluctuation compensation coefficient and the estimated travel time to the predicted release time. The preset traffic fluctuation compensation coefficient... It is a dimensionless floating-point number, and its value is usually between 0.1 and 0.3. This coefficient is used to quantify the uncertainty of travel time introduced by dynamic changes in traffic conditions. Its specific value is set based on the statistical analysis results of historical traffic data. For example, it is set to 0.1 in the scenario of smooth traffic on highways, and 0.3 in the urban center area with complex road conditions.

[0085] After calculating the time window, the module quantitatively compares the estimated arrival time in the predicted time index with the calculated exclusive serviceable time window. The core logic of this comparison is to verify whether the estimated arrival time is earlier than or equal to the cutoff value of the exclusive serviceable time window. Specifically, the judgment condition for the timing logic verification is as follows:

[0086]

[0087] This step aims to filter out infeasible pairings where the vehicle cannot arrive before the charging station is occupied by another random user due to long distances or severe traffic congestion. All vehicle-charging station combinations that pass this timing constraint verification are marked as time-feasible target pairings and added to a set data structure, ultimately forming a target pairing set. This set will serve as input for the next step of global optimization scheduling. It should be noted that the target pairing set is a list of several target pairings that have passed the timing logic verification. Each target pairing contains at least a vehicle ID, a charging station ID, and a corresponding predicted time indicator pair.

[0088] For example, continuing from the previous embodiment, the timing logic verification module processes two pairs of predicted time indicators. The preset traffic fluctuation compensation coefficient... The value is 0.2. First, the pairing of V001-C005 is processed, and its prediction time index pair is... Current time The timeframe is 10:00:02. The module calculates the vehicle's estimated travel time as 10:12:32 minus 10:00:02, resulting in 750 seconds. Next, the buffer time for the time window is calculated as 0.2 multiplied by 750 seconds, yielding 150 seconds. Therefore, the exclusive service availability time window cutoff value is... Add 150 seconds to 10:42:02, resulting in 10:44:32. Then, a check is performed to determine if the estimated arrival time 10:12:32 is less than or equal to the cutoff value 10:44:32. The result is yes, therefore the V001-C005 combination is considered a target pair. Next, the pairing of V001-C006 is processed, with its predicted time index pair being... The module calculates the estimated travel time as 10:10:02 minus 10:00:02, resulting in 600 seconds. The buffer time is calculated as 0.2 multiplied by 600 seconds, yielding 120 seconds. Exclusivity-available service window cutoff value. Add 120 seconds to 10:00:02, resulting in 10:02:02. A check is performed to determine if the estimated arrival time 10:10:02 is less than or equal to the cutoff time 10:02:02. The result is no. Therefore, the V001-C006 combination is filtered out because it does not meet the timing constraints. Ultimately, the target pairing set generated by the module contains only one record: {Vehicle ID: V001, Charging Pile ID: C005}, and this record is sent to the global optimization scheduling module.

[0089] In one embodiment of the present invention, the global scheduling optimization module is configured to perform the following steps:

[0090] The target pairing set is traversed, and the predicted waiting time for each target pairing is extracted. The predicted waiting time is obtained by taking the absolute value of the difference between the predicted release time and the expected arrival time. The absolute value is used to comprehensively represent the vehicle queuing time and the charging pile idle time. Taking the total predicted waiting time of the vehicles waiting to be charged as the global optimization objective, the target pairing set is globally matched and optimized using a combinatorial optimization algorithm to determine the target charging pile code corresponding to each vehicle waiting to be charged. The target charging pile code, the expected arrival time, and the predicted release time are encapsulated to generate a clearly targeted global optimization scheduling instruction.

[0091] Specifically, the global optimization scheduling module within the central scheduling system receives the target pairing set output by the timing logic verification module and executes global optimization scheduling with the objective function of maximizing system throughput to determine the optimal scheduling instruction. This module first iterates through the target pairing set, extracting the predicted arrival time and predicted release time from the corresponding predicted time index pair for each target pair. Then, the module calculates the difference between the predicted release time and the predicted arrival time, and obtains the predicted waiting time for that target pair by taking the absolute value. This optimized waiting time index not only quantifies the queuing time a vehicle needs to wait for the charging station to release after arrival, but also penalizes idle time caused by premature idling of charging stations. In this embodiment, the predicted waiting time... It is a non-negative time quantity, and its calculation formula is:

[0092]

[0093] Next, the module constructs a combinatorial optimization problem with the global optimization objective of minimizing the total predicted waiting time of all vehicles waiting to be charged within the current scheduling cycle. This problem is typically modeled as a bipartite graph matching or assignment problem. Specifically, the objective function for global optimization is... That is, minimizing the total prediction waiting time, its mathematical expression is:

[0094]

[0095] The optimization problem needs to satisfy the following constraints: For each vehicle i to be charged, there are For each candidate charging station ,have ;and This is a binary variable, taking the value 0 or 1. (Variable) Represents a single entity in the set of vehicles to be charged, variable This represents an entity in the set of candidate charging stations P involved in the target pairing set. This is a decision variable; when its value is 1, it indicates that vehicle i will be assigned to a charging station. When the value is 0, it means that this allocation will not be performed.

[0096] The module utilizes its built-in integer linear programming solver to solve the problem. Through iterative calculations, it finds the optimal solution among all possible vehicle-charging-pile matching combinations that minimizes the total predicted waiting time. It should be understood that the combinatorial optimization algorithm in this scheme specifically refers to the integer linear programming algorithm used to solve the assignment problem. This optimal solution clearly defines the target charging pile code that should be assigned to each vehicle waiting to be charged. The target charging pile code is a unique device identifier for the charging pile. Finally, based on the solution results, the module encapsulates the corresponding target charging pile code, estimated arrival time, and predicted release time for each vehicle waiting to be charged, generating one or more structured, clearly targeted global optimization scheduling instructions, and transmitting these instructions to the downstream resource logic locking module. It should be noted that the global optimization scheduling instruction is a data message whose data structure contains key scheduling information such as vehicle ID, assigned target charging pile code, estimated arrival time, predicted release time, and predicted waiting time.

[0097] For example, to more fully illustrate the optimization process, assume that in addition to V001, another vehicle V002 also initiated a charging request, and after processing by the timing constraint filtering module, the resulting target pairing set contains two feasible target pairs: {Vehicle ID: V001, Charging Pile ID: C005} and {Vehicle ID: V002, Charging Pile ID: C005}. The predicted time index pair for V001 remains... The prediction time index pair for V002 is... First, the modules calculate their respective predicted waiting times. For the V001-C005 pairing, the predicted waiting time is... Seconds. For V002-C005 pairing, the predicted waiting time is... Seconds. Next, the optimization solver makes a decision. Since only one charging pile C005 is available for allocation, the system needs to choose one of the two vehicles. With the objective of minimizing the total waiting time, the total waiting time for allocating C005 to V001 is 1770 seconds, and the total waiting time for allocating it to V002 is 1322 seconds. To minimize the total system waiting time, the solver determines that the optimal scheduling is to allocate charging pile C005 to vehicle V002. Finally, a global optimized scheduling instruction is generated, with the content {Vehicle ID: V002, Target charging pile code: "C005", Estimated arrival time: "10:20:00", Predicted release time: "10:42:02"}, and this instruction is sent to the execution module. For vehicle V001, no scheduling instruction is generated for the time being, and it is placed in the next round of scheduling pool.

[0098] In one embodiment of the present invention, the resource locking control module is configured to perform the following steps:

[0099] The target charging pile code in the global optimization scheduling instruction is sent to the corresponding target charging pile controller via a wireless communication link. After receiving the instruction, the target charging pile controller sets the status position to the pre-occupancy flag in the local control register and simultaneously shuts down the real-time status feedback channel of the charging pile to the external public query interface. During the period of the pre-occupancy flag, all charging request attempts except for the identity identifier of the vehicle to be charged are blocked, thus completing the logical locking of the physical charging resources.

[0100] Specifically, the resource logic locking module of the central dispatch system sends a pre-occupancy signal to the target charging pile and rewrites its access control logic based on the generated global optimized scheduling instructions. First, this module extracts the target charging pile code and the associated vehicle identification identifier from the global optimized scheduling instructions. The vehicle identification identifier can be the vehicle's VIN, the unique identifier bound to the charging card, or the user ID assigned by the vehicle-to-everything (V2X) platform; this identifier is automatically reported by the vehicle during the charging handshake protocol phase. Next, the module sends a pre-occupancy instruction data packet to the network address corresponding to the target charging pile code via a secure wireless communication link.

[0101] In this embodiment, the wireless communication link is a narrowband IoT or LTE Cat-1 cellular network dedicated to industrial-grade IoT, ensuring low latency and high reliability of command issuance. After receiving the data packet and verifying its validity, the embedded controller of the target charging pile immediately rewrites its local control logic. The target charging pile controller is an embedded main control board integrating a communication module, a billing unit, and high-voltage relay control logic. Specifically, in the local control register of the target charging pile controller, the status bit storing the current operating state is rewritten from its current value to a preset pre-emption flag value. It should be noted that the local control register is a volatile or non-volatile storage area within the microcontroller unit on the main control board, used to store the charging pile's operating status word. The pre-emption flag is an independent status code distinct from idle, charging, or fault states. For example, if 0x00 represents idle and 0x01 represents charging, then 0x02 can be defined as the pre-emption flag.

[0102] Simultaneously, the controller stores the vehicle's identity identifier contained in the instruction into a temporary authentication whitelist and sets a pre-occupancy validity period, which can be determined based on the predicted release time and the exclusivity service window deadline. After completing the status setting, the controller will synchronously adjust the behavior of its external communication protocol stack so that all status query requests to external public query interfaces return a pre-occupancy flag, thereby temporarily masking its actual physical status. Furthermore, the external public query interface typically refers to a full-duplex communication WebSocket or HTTP service port conforming to open charging protocols such as the Open Charging Protocol (OCPP), allowing third-party platforms to query the charging station status. During the duration of this pre-occupancy flag, when any vehicle attempts to start charging, the controller will force the vehicle to authenticate and compare the obtained vehicle identity identifier with the previously stored authentication whitelist. If the comparison does not match, the charging request is rejected. This logically locks the physical charging resource, ensuring that it is used only by the designated vehicle.

[0103] For example, following the output of the previous embodiment, the resource logic locking module receives a globally optimized scheduling instruction generated for vehicle V002, which includes the target charging pile code C005. First, the module extracts the target charging pile code "C005" and the vehicle ID "V002" from the instruction. Then, based on the exclusive serviceable time window deadline of 10:46:01 corresponding to V002 calculated in the timing constraint filtering module, a pre-occupancy instruction in JSON format is encapsulated, with the content being {"command":"LOCK", "target_vehicle_id":"V002", "expire_at":"10:46:01"}. This instruction is sent to the controller of the target charging pile C005 via the LTE network. After receiving and verifying the instruction at 10:00:10, the controller of C005 immediately updates the value of its status register at address 0x1F04 in memory from 0x01 (representing "charging") to 0x02 (representing "reservation flag") after the current charging task ends, and writes the vehicle ID "V002" into the authentication whitelist. At this time, if an app from a charging aggregation platform initiates a status query to the public query interface of C005, C005 will no longer return "charging" or "idle," but will uniformly return "reserved." When the time reaches 10:35:00, an unscheduled vehicle V004 connects to C005 and attempts to charge, and its onboard controller reports the identity identifier "V004." The controller of C005 compares "V004" with "V002" in the whitelist. If they do not match, the charging program will not be started, and "This charging station has been reserved" will be displayed on the charging station screen. This achieves the logical locking of the charging station C005 and ensures that it is reserved for vehicle V002.

[0104] In one embodiment of the present invention, the closed-loop execution monitoring module is used to perform the following steps:

[0105] A scheduling confirmation request containing the location of the target charging station and the predicted waiting time is pushed to the user terminal of the vehicle to be charged. After obtaining the user confirmation signal, the location coordinates of the target charging station are hard-bound to the visual navigation system of the vehicle to be charged to generate a closed-loop directional navigation path. The real-time trajectory of the vehicle to be charged on the closed-loop directional navigation path is continuously monitored. If the path deviation exceeds a preset threshold, a cancellation signal is sent to the target charging station to release the pre-occupancy flag.

[0106] Specifically, the execution closed-loop and path binding module of the central dispatch system receives the global optimization scheduling instructions and transforms them into closed-loop directional navigation paths, while simultaneously monitoring the execution status in real time to form closed-loop control. This module first parses the global optimization scheduling instructions, extracting the geographical coordinates of the target charging station, the predicted waiting time, and the user terminal identifier of the vehicle to be charged. The user terminal is a smartphone or tablet with the accompanying charging service software installed.

[0107] Subsequently, the module sends a dispatch confirmation request to the user terminal of the vehicle to be charged via a mobile application push service. This request displays the target charging station's location and waiting time information in the form of an interactive card, and provides confirmation or cancellation options. Upon receiving the user confirmation signal from the user terminal, the module immediately sends a navigation setting command to the remote information control unit of the vehicle to be charged via the vehicle communication network interface. This command includes the target charging station's location coordinates and a forced binding flag. The user confirmation signal is a data packet containing a confirmation flag sent from the user terminal to the central dispatch system's backend server via an HTTP POST request.

[0108] Upon receiving the instruction, the visual navigation system of the vehicle awaiting charging sets the location coordinates of the target charging station as the navigation destination with exclusive priority and plans a closed-loop directional navigation path. It should be understood that the closed-loop directional navigation path is a route locked within the navigation system, consisting of a series of geographic coordinate points; the user cannot modify the destination or waypoints during the journey. Subsequently, the central dispatch system continuously receives real-time trajectory data reported by the vehicle at high frequency and calculates the deviation between the vehicle's current position and the closed-loop directional navigation path.

[0109] In this embodiment, the path deviation magnitude The calculation can be simplified to calculating the vehicle's current point. The perpendicular distance to the nearest path segment L on the closed-loop directional navigation path. Assume line segment L consists of two points... and By definition, distance The calculation formula is:

[0110]

[0111] Among them, the path deviation magnitude is a physical quantity used to quantify whether a vehicle follows a specified route.

[0112] If the path deviation exceeds a preset threshold for a certain period of time, the vehicle is determined to have abandoned the dispatch command. In this embodiment, a certain period of time refers to at least three consecutive instances of abnormal path deviation collected, or a duration of 10 to 30 seconds, to rule out misjudgments caused by instantaneous GPS signal drift. At this time, the trigger condition for dynamic closed-loop adjustment is: Among them, the preset threshold This distance value is set based on a combination of road type and navigation accuracy. For example, it is set to 50-100 m on urban roads and 200-300 m on highways, used to distinguish between normal driving fine-tuning and intentional route deviation. A cancellation signal is then automatically generated and sent to the target charging station. The cancellation signal is a command message, functioning the opposite of the pre-occupancy command, used to instruct the charging station controller to release its logical lock. Upon receiving the cancellation signal, the target charging station controller immediately clears its local pre-occupancy flag, thereby releasing the charging resource and returning it to the public scheduling pool, thus achieving dynamic closed-loop adjustment of the scheduling scheme.

[0113] For example, following the previous embodiment, the closed-loop and path binding module processes the global optimization scheduling command for vehicle V002. First, at 10:00:12, a message is pushed to the user terminal of V002, such as: "Scheduling successful. Charging station C005 has been locked for you. The estimated waiting time is 22 minutes and 2 seconds. Please confirm whether to proceed." The user clicks "confirm" at 10:00:20. Upon receiving the confirmation signal, the module immediately sends a command to the navigation system of vehicle V002, hard-binding the target charging station C005 as the destination and generating a unique closed-loop directional navigation path. Vehicle V002 begins to travel along this path. At 10:08:30, the module receives the real-time GPS coordinates of V002 and calculates its path deviation from the closed-loop directional navigation path, d = 150 m. A preset threshold for urban roads is set. The distance was 80 m. Since 150 m exceeded 80 m, the system determined that V002 had deviated from the planned route. Therefore, it immediately generated a cancellation signal with the content {"command":"UNLOCK", "reason":"DEVIATION"} and sent it to the controller of the target charging pile C005. After receiving the signal, C005 cleared the status bit of its local control register from the pre-emptive flag state of 0x02, restoring it to the scheduleable state, thus completing a dynamic closed-loop adjustment caused by the execution deviation, and removed V002 from the current scheduling task, notifying it to re-initiate the request.

[0114] Each of the modules can be implemented in whole or in part through software, hardware, or a combination thereof. It supports hardware embedded in or independent of the processor in the computer device, and also supports software stored in the memory of the computer device, so that the processor can call and execute the operations corresponding to each of the above modules.

[0115] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included within the protection scope of the present invention.

Claims

1. A charging pile intelligent scheduling and queuing optimization system based on big data analysis, characterized in that, include: The state parameter aggregation module is used to analyze vehicle-mounted sensor signals and pile-end electrical signals to establish a dynamic state parameter set for the vehicle to be charged and the candidate charging piles. The predictive index calculation module performs time-series extrapolation calculations based on a dynamic state parameter set to generate a predictive time index pair for each vehicle and charging pile pair. The predictive time index pair includes the expected arrival time and the predicted release time. The timing constraint filtering module is used to establish an exclusive serviceable time window, perform timing logic verification based on the predicted time indicators, and filter the target pair set that meets the timing constraints. The global scheduling optimization module takes the total predicted waiting time of vehicles waiting to be charged as the global optimization target. It uses a combinatorial optimization algorithm to perform global matching and optimization in the target pairing set, determines the target charging pile code corresponding to each vehicle waiting to be charged, and generates a global optimization scheduling instruction. The resource locking control module sends a pre-occupancy signal to the target charging pile according to the global optimization scheduling instruction. It sets the status to the pre-occupancy flag in the local control register of the target charging pile controller and blocks all charging request attempts except for the identity of the vehicle to be charged during the duration of the pre-occupancy flag, thus completing the logical locking of the physical charging resources. The closed-loop execution monitoring module is used to convert global optimization scheduling instructions into closed-loop directional navigation paths and monitor the execution status in real time. If the deviation of the vehicle to be charged on the closed-loop directional navigation path exceeds a preset threshold, a cancellation signal is sent to the target charging pile to release the pre-occupancy flag, thereby realizing the dynamic closed-loop adjustment of the scheduling scheme.

2. The intelligent scheduling and queuing optimization system for charging piles based on big data analysis according to claim 1, characterized in that, The status parameter aggregation module is used to perform the following operations: The system acquires the GPS coordinate sequence, battery state of charge percentage, and battery rated capacity of the vehicle to be charged, and converts the GPS coordinate sequence into a vehicle displacement vector. Simultaneously acquire the real-time working status indicator bit, rated charging power level, and the accumulated charging amount pulse signal of the vehicle currently being charged for each candidate charging pile; By aggregating vehicle displacement vector, battery state of charge percentage, real-time operating status indicator bit, and accumulated charge pulse signal, a dynamic state parameter set containing spatiotemporal characteristics and energy state characteristics is generated.

3. The intelligent scheduling and queuing optimization system for charging piles based on big data analysis according to claim 1, characterized in that, The predictive metric calculation module is used to perform the following steps: Based on the vehicle displacement vector in the dynamic state parameter set and combined with the real-time road network congestion coefficient, the estimated arrival time of the vehicle to be charged to each candidate charging station is calculated. For candidate charging piles that are currently occupied, the predicted release time of the candidate charging pile is determined by extrapolating the remaining completion time of the current charging task based on its corresponding rated charging power level and the pulse signal of the accumulated charging amount of the vehicle currently being charged. By mapping and associating the expected arrival time with the predicted release time, a pair of predicted time indicators reflecting the intersection of the vehicle movement process and the charging physical process is generated.

4. The intelligent scheduling and queuing optimization system for charging piles based on big data analysis according to claim 1, characterized in that, The timing constraint filtering module is used to perform the following steps: Based on the predicted release time in the predicted time index pair, and combined with the preset traffic fluctuation compensation coefficient, the start and end values ​​of the exclusive service time window for each candidate charging pile are calculated. The estimated arrival time is quantitatively compared with the exclusive available service time window to determine whether the estimated arrival time falls within the corresponding exclusive available service time window. The vehicle and charging pile combinations that fall within the time window are retained and marked as target pairs with temporal feasibility, forming a target pair set.

5. The intelligent scheduling and queuing optimization system for charging piles based on big data analysis according to claim 1, characterized in that, The global scheduling optimization module is used to perform the following steps: Traverse the target pairing set and extract the predicted waiting time for each target pairing. The predicted waiting time is obtained by taking the absolute value of the difference between the predicted release time and the expected arrival time. The absolute value is used to comprehensively characterize the vehicle queuing time and the charging pile idle time. Using the total predicted waiting time of vehicles waiting to be charged as the global optimization objective, a combinatorial optimization algorithm is used to perform global matching and optimization on the target pairing set to determine the target charging pile code corresponding to each vehicle waiting to be charged. The target charging pile code, estimated arrival time, and predicted release time are encapsulated to generate a clearly targeted global optimization scheduling instruction.

6. The intelligent scheduling and queuing optimization system for charging piles based on big data analysis according to claim 1, characterized in that, The resource locking control module is used to perform the following operations: The target charging pile code in the global optimization scheduling instruction is sent to the corresponding target charging pile controller via a wireless communication link. After receiving the instruction, the target charging pile controller sets the status position to the pre-occupancy flag in the local control register and simultaneously shuts down the real-time status feedback channel of the charging pile to the external public query interface. During the period of the pre-occupancy flag, all charging request attempts except for the identity identifier of the vehicle to be charged are blocked, thus completing the logical locking of the physical charging resources.

7. The intelligent scheduling and queuing optimization system for charging piles based on big data analysis according to claim 1, characterized in that, The closed-loop execution monitoring module is used to perform the following operations: A scheduling confirmation request containing the location of the target charging station and the predicted waiting time is pushed to the user terminal of the vehicle waiting to be charged. After obtaining the user's confirmation signal, the location coordinates of the target charging pile are hard-bound to the visual navigation system of the vehicle to be charged, generating a closed-loop directional navigation path; The system continuously monitors the real-time trajectory of the vehicle waiting to be charged on the closed-loop directional navigation path. If the path deviation exceeds a preset threshold, a cancellation signal is sent to the target charging pile to release the pre-occupancy flag.

8. The intelligent scheduling and queuing optimization system for charging piles based on big data analysis according to claim 2, characterized in that, Converting a GPS coordinate sequence into a vehicle displacement vector includes the following steps: Extract the coordinate points corresponding to two adjacent timestamps from the GPS coordinate sequence; After projecting from spherical coordinates to local planar coordinates, the vector difference between the two points is calculated to generate the vehicle displacement vector that characterizes the instantaneous motion state of the vehicle.

9. The intelligent scheduling and queuing optimization system for charging piles based on big data analysis according to claim 4, characterized in that, Calculating the exclusivity service window cutoff value for candidate charging stations includes the following steps: Calculate the time difference between the vehicle's estimated arrival time and the current system time as the estimated travel time; The exclusive service window cutoff value is obtained by multiplying the predicted release time by the expected travel time by the preset traffic fluctuation compensation coefficient.

10. The intelligent scheduling and queuing optimization system for charging piles based on big data analysis according to claim 7, characterized in that, If a path deviation exceeding a preset threshold is detected, the following steps are included: Extract the real-time trajectory location coordinates of the vehicle to be charged, construct a straight line equation based on the coordinates of the two endpoints of the nearest path segment on the closed-loop directional navigation path, and calculate the geometric vertical distance from the location coordinates to the straight line as the path deviation magnitude. The path deviation is compared with a preset threshold set according to the road type. If the path deviation exceeds the preset threshold for a certain period of time, it is determined that the path deviation exceeds the preset threshold.

Images