A power dispatch control method and system for charging stations

Abstract

This invention belongs to the field of charging station management technology, and relates to a power dispatching and control method and system for charging stations. The method includes: generating the estimated parking time for each vehicle in the charging station based on real-time monitoring of the status of each charging space; comparing the estimated parking time with a preset threshold to classify vehicles as short-term or long-term parking vehicles; generating a differentiated power dispatching strategy based on the vehicle classification results; generating specific power control instructions according to the power dispatching strategy, and issuing them to the corresponding charging piles for execution. This invention significantly optimizes the utilization rate of limited resources within the charging station while meeting the charging requirements of various types of vehicles.

Description

Technical Field

[0001] This invention belongs to the field of charging station management technology, and in particular to a power dispatching and control method and system for charging stations. Background Technology

[0002] Electric vehicle charging stations generally employ power dispatch and control systems to address the conflict between fixed power distribution capacity and fluctuating charging demand. Traditional dispatch methods are mostly based on the current battery charge (SOC) of electric vehicles, the urgency of charging, or the target charge set by the user, aiming to achieve the real-time optimal allocation of power resources.

[0003] However, these methods only focus on the vehicle's battery status, neglecting a crucial but underutilized dimension for scheduling timing: the estimated time a vehicle will remain at the depot. Because the parking duration is unpredictable, existing scheduling systems are prone to short-sighted decisions. For example, they might allocate insufficient power to a vehicle about to leave, preventing it from charging on time; conversely, they might allocate expensive peak-hour power too early to a vehicle that will be parked for an extended period, increasing operating costs and crowding out power resources for other vehicles urgently needing charging. This scheduling strategy fails to deeply integrate with the time dimension, resulting in suboptimal utilization of both power and parking resources at the depot.

[0004] While some studies have attempted to predict user behavior using complex algorithms, the data sources they rely on (historical user data, traffic big data) are costly to acquire and may even involve user privacy, making them difficult to directly apply to real-time power dispatching in actual charging stations. Therefore, this paper proposes a power dispatching and control method and system for charging stations to address the aforementioned problems. Summary of the Invention

[0005] The purpose of this invention is to overcome the shortcomings of the prior art and provide a power dispatch control method and system for charging stations.

[0006] The technical problem solved by this invention is achieved through the following technical solution:

[0007] A power dispatching and control method for charging stations, characterized by the following steps:

[0008] Based on real-time monitoring of the status of each charging space in the station, the estimated parking time for each vehicle in each charging space is generated.

[0009] The estimated parking time is compared with a preset threshold to classify vehicles as either short-term or long-term parked vehicles.

[0010] Based on the vehicle classification results, a differentiated power scheduling strategy is generated. The power scheduling strategy includes prioritizing the allocation of high power for fast charging of short-term parked vehicles and planning charging time windows for long-term parked vehicles, so that they can be charged during the off-peak hours of electricity load or electricity price.

[0011] Specific power control commands are generated based on the power scheduling strategy and sent to the corresponding charging piles for execution.

[0012] As a further aspect of the present invention, the step of generating the estimated parking time for each vehicle in a charging space based on real-time monitoring of the status of each charging space within the station specifically includes:

[0013] The system continuously collects and generates triaxial magnetometer signals corresponding to the vehicle status by deploying geomagnetic sensors in the charging parking spaces.

[0014] Within the first time window after the vehicle stops, the triaxial magnetometer signal is analyzed and an initial behavior pattern feature vector including time interval, number of signal disturbances and steady-state coefficient is extracted. The time interval is the time from when the vehicle stops until the owner opens the door and gets out of the vehicle. The number of signal disturbances is the signal fluctuation caused by the opening and closing of the door. The steady-state coefficient characterizes the signal stability after the person moves away from the vehicle.

[0015] The initial behavior pattern feature vector is input into a pre-trained duration prediction model to obtain the vehicle's estimated parking time.

[0016] The estimated parking time is associated with and bound to the corresponding vehicle, charging space, and charging pile information.

[0017] As a further aspect of the present invention, the step of continuously acquiring and generating triaxial magnetometer signals corresponding to the vehicle state specifically includes:

[0018] The system detects signal interference events by checking whether the triaxial magnetometer signals of adjacent charging spaces exhibit synchronous correlation fluctuations.

[0019] When a signal interference event occurs, the source parking space and the affected parking space are identified based on the amplitude and timing characteristics of the associated triaxial magnetometer signal changes.

[0020] Based on the status determination result of the parking space of the interference source, the triaxial magnetometer signal of the interference parking space is reconstructed through preset logic rules;

[0021] Output the triaxial magnetometer signal after state determination and reconstruction processing.

[0022] As a further aspect of the present invention, the step of analyzing the triaxial magnetometer signal and extracting the initial behavior pattern feature vector, including time interval, number of signal disturbances, and steady-state coefficient, specifically includes:

[0023] When the amplitude change rate of the triaxial magnetometer signal is continuously lower than the first steady-state threshold for more than a set time, it is determined that the vehicle has come to a stop and this time is recorded as the reference time.

[0024] Based on the reference time, the time of the first significant signal disturbance caused by the car owner opening the door and getting out of the car is identified and recorded as the first time, and the time interval is the difference between the first time and the reference time.

[0025] Based on the first moment, a second time window is preset, and the number of signal disturbances caused by human activities is counted within the second time window.

[0026] The steady-state coefficient is obtained by synchronously analyzing the proportion of events in the second time window where the amplitude of the triaxial magnetometer signal is lower than the second steady-state threshold.

[0027] The initial behavior pattern feature vector is obtained by integrating the time interval, the number of signal disturbances, and the steady-state coefficient.

[0028] As a further aspect of the present invention, the step of comparing the expected parking time with a preset threshold specifically includes:

[0029] Obtain the current total charging load power and the maximum power distribution capacity of the power station, and calculate the real-time load factor;

[0030] Obtain the number of available charging spaces and the total number of charging spaces in the station, and calculate the parking space turnover urgency index;

[0031] Based on the real-time load rate and parking space turnover urgency index, the classification threshold for the current time is dynamically calculated using a preset mapping function.

[0032] The estimated parking time of each vehicle is compared with the dynamically generated classification threshold to classify the vehicles.

[0033] As a further aspect of the present invention, when a power grid peak shaving command is received and there are vehicles temporarily parked within the power station, the step of generating a differentiated power scheduling strategy based on the vehicle classification results specifically includes:

[0034] Obtain the maximum total power of the power station corresponding to the power grid peak shaving command, as well as the infrastructure load of the power station and the minimum guaranteed total power allocated to all long-term parked vehicles, and calculate the guaranteed power budget corresponding to short-term parked vehicles.

[0035] For each vehicle that is parked for a short period of time, a charging emergency index is calculated based on its current battery status, the target charging capacity set by the user, and the duration of connection.

[0036] Based on the charging emergency index, short-term parked vehicles are sorted to obtain an emergency dispatch queue and charging power is allocated.

[0037] The power scheduling strategy is obtained by integrating the allocation results, the minimum guaranteed total power of vehicles parked for a long time, and the infrastructure load.

[0038] Another object of the present invention is to provide a power dispatching and control system for charging stations, the system comprising:

[0039] The status monitoring module is used to generate the estimated parking time for each vehicle in each charging space based on real-time monitoring of the status of each charging space in the station.

[0040] The vehicle classification module is used to compare the estimated parking time with a preset threshold to classify vehicles as either short-term or long-term parking vehicles.

[0041] The strategy generation module is used to generate differentiated power scheduling strategies based on vehicle classification results. The power scheduling strategies include prioritizing the allocation of high power for fast charging of short-term parked vehicles and planning charging time windows for long-term parked vehicles, and enabling them to charge during periods of low electricity load or low electricity price.

[0042] The instruction execution module is used to generate specific power control instructions according to the power scheduling strategy and send them to the corresponding charging piles for execution.

[0043] As a further aspect of the present invention, the status monitoring module includes:

[0044] The signal acquisition unit is used to continuously acquire and generate triaxial magnetometer signals corresponding to the vehicle status through geomagnetic sensors deployed in the charging parking spaces.

[0045] The feature extraction unit is used to analyze the triaxial magnetometer signal and extract the initial behavior pattern feature vector, including time interval, number of signal disturbances and steady-state coefficient, within the first time window after the vehicle stops. The time interval is the time from when the vehicle stops until the owner opens the door and gets out of the vehicle. The number of signal disturbances is the signal fluctuation caused by the opening and closing of the door. The steady-state coefficient characterizes the signal stability after the person moves away from the vehicle.

[0046] The duration prediction unit is used to input the initial behavior pattern feature vector into the pre-trained duration prediction model to obtain the estimated parking time of the vehicle.

[0047] The information binding unit is used to associate and bind the estimated parking time with the corresponding vehicle, charging space and charging pile information.

[0048] As a further aspect of the present invention, the vehicle classification module includes:

[0049] The load calculation unit is used to obtain the current total charging load power and the maximum power distribution capacity of the station, and to calculate the real-time load factor.

[0050] The urgency assessment unit is used to obtain the number of vacant charging spaces and the total number of charging spaces in the station, and to calculate the parking space turnover urgency index.

[0051] The threshold calculation unit is used to dynamically calculate the classification threshold for the current time based on the real-time load rate and parking space turnover urgency index using a preset mapping function.

[0052] The dynamic classification unit is used to compare the estimated parking time of each vehicle with the dynamically generated classification threshold to classify the vehicles.

[0053] As a further aspect of the present invention, the strategy generation module includes:

[0054] The budget calculation unit is used to obtain the upper limit total power of the power station corresponding to the power grid peak shaving command, as well as the infrastructure load of the power station and the minimum guaranteed total power allocated to all long-term parked vehicles, and calculate the guaranteed power budget corresponding to short-term parked vehicles.

[0055] The emergency quantification unit is used to calculate the charging emergency index for each short-term parked vehicle based on its current battery status, the user-set target charging capacity, and the duration of connection.

[0056] The sorting and allocation unit is used to sort the short-term parked vehicles according to the charging emergency index, obtain an emergency dispatch queue, and allocate charging power.

[0057] The strategy integration unit is used to integrate the allocation results, the minimum guaranteed total power of long-term parked vehicles, and the infrastructure load to obtain the power scheduling strategy.

[0058] The advantages and beneficial effects of this invention are as follows:

[0059] This invention deeply integrates the time dimension into the power dispatching decision-making process by monitoring the status of charging spaces in real time and generating corresponding estimated vehicle parking times. This effectively solves the short-sightedness of traditional methods that ignore the duration of vehicle stays at the depot. Based on the comparison between the estimated parking time and a preset threshold, vehicles can be accurately classified into short-term and long-term parking types, and differentiated power dispatching strategies are implemented accordingly. For short-term parked vehicles, high-power charging is prioritized for fast charging, ensuring that they can complete recharging and leave the depot within the planned time, thereby significantly improving the turnover efficiency of charging spaces and user satisfaction. For long-term parked vehicles, their charging time windows are intelligently planned, and their main charging process is scheduled during off-peak hours of electricity load or price. This effectively reduces the overall electricity cost of the depot and alleviates the power supply pressure on the grid during peak hours, achieving a synergistic effect of peak shaving and valley filling.

[0060] In summary, this invention significantly optimizes the utilization rate of limited resources within the charging station while meeting the charging requirements of various vehicles. Attached Figure Description

[0061] Figure 1 This is a flowchart of a power dispatch control method for charging stations;

[0062] Figure 2 A flowchart for generating the estimated parking time in a power dispatch control method for charging stations;

[0063] Figure 3 This is a flowchart illustrating the generation of triaxial magnetometer signals in a power dispatch control method for charging stations.

[0064] Figure 4 This is a flowchart illustrating the analysis of triaxial magnetometer signals in a power dispatch control method for charging stations.

[0065] Figure 5 This is a flowchart illustrating the comparison between the expected parking time and a preset threshold in a power dispatch control method for charging stations.

[0066] Figure 6 A flowchart illustrating the generation of differentiated power dispatch strategies under special circumstances in a power dispatch control method for charging stations;

[0067] Figure 7 This is a schematic diagram of a power dispatch and control system for charging stations.

[0068] Figure 8 This is a schematic diagram of the structure of a status monitoring module in a power dispatch control system for charging stations;

[0069] Figure 9This is a schematic diagram of the vehicle classification module in a power dispatch control system for charging stations.

[0070] Figure 10 This is a schematic diagram of the strategy generation module in a power dispatch control system for charging stations. Detailed Implementation

[0071] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0072] The specific implementation of the present invention will be described in detail below with reference to specific embodiments.

[0073] like Figure 1 As shown in the figure, this embodiment of the invention provides a power dispatching and control method for charging stations, the innovation of which is that the method includes the following steps:

[0074] S100 generates the estimated parking time for each vehicle in each charging space based on real-time monitoring of the status of each charging space in the station.

[0075] S200 compares the estimated parking time with a preset threshold to classify vehicles as either short-term or long-term parked vehicles.

[0076] S300, generates a differentiated power scheduling strategy based on the vehicle classification results. The power scheduling strategy includes prioritizing the allocation of high power for fast charging of the short-term parked vehicles and planning the charging time window for the long-term parked vehicles, and enabling them to charge during the off-peak hours of electricity load or electricity price.

[0077] S400 generates specific power control commands based on the power scheduling strategy and sends them to the corresponding charging piles for execution.

[0078] It should be noted that the step of generating specific power control commands based on the power scheduling strategy and sending them to the corresponding charging piles for execution is the final execution stage of the entire scheduling and control logic. It transforms the optimized strategy into actual operation commands that the charging piles can recognize. Based on the generated power scheduling strategy, a specific target output power value and charging start / stop time command are calculated for each controlled charging pile. These commands are encapsulated into standard data frames conforming to the charging pile control protocol through the site's internal communication network (such as Ethernet or a wireless communication module). Subsequently, based on the unique identifier of the charging pile, the corresponding power control command is sent point-to-point to the target charging pile. After receiving the command, the charging pile's internal controller parses and executes the command, precisely adjusting the operating parameters of its power electronic modules (such as AC / DC or DC / DC converters), thereby achieving precise control of the output power and timely start / stop of the charging process.

[0079] In this embodiment of the invention, the present invention deeply integrates the time dimension into the power dispatch decision-making process by real-time monitoring of the charging space status and generating corresponding estimated vehicle parking time. This effectively solves the problem of short-sighted scheduling caused by traditional methods neglecting the duration of vehicle stay at the depot. Based on the comparison between the estimated parking time and a preset threshold, vehicles can be accurately classified into short-term and long-term parking types, and differentiated power dispatch strategies can be implemented accordingly. For short-term parking vehicles, high power is prioritized for fast charging to ensure that they can complete recharging and leave the depot within the planned time, thereby significantly improving the turnover efficiency of charging spaces and user satisfaction. For long-term parking vehicles, their charging time windows are intelligently planned, and their main charging process is scheduled to be carried out during off-peak hours of electricity load or price. This effectively reduces the overall electricity cost of the depot and alleviates the power supply pressure on the grid during peak hours, achieving a synergistic effect of peak shaving and valley filling.

[0080] like Figure 2 As shown, in a preferred embodiment of the present invention, the step of generating the estimated parking time for each vehicle in a charging space based on real-time monitoring of the status of each charging space within the charging station specifically includes:

[0081] S101 continuously collects and generates triaxial magnetometer signals corresponding to the vehicle status through geomagnetic sensors deployed in charging parking spaces.

[0082] S102, within the first time window after the vehicle stops, analyze the triaxial magnetometer signal and extract the initial behavior pattern feature vector including time interval, number of signal disturbances and steady-state coefficient. The time interval is the time from when the vehicle stops until the owner opens the door and gets out of the vehicle. The number of signal disturbances is the signal fluctuation caused by the opening and closing of the door. The steady-state coefficient characterizes the signal stability after the person moves away from the vehicle.

[0083] S103, input the initial behavior pattern feature vector into the pre-trained duration prediction model to obtain the estimated parking time of the vehicle;

[0084] S104, associate and bind the estimated parking time with the corresponding vehicle, charging space and charging pile information.

[0085] In this embodiment of the invention, a triaxial magnetometer sensor, pre-deployed in each charging parking space, continuously collects changes in the spatial magnetic field caused by the vehicle's entry, parking, and departure, and outputs a triaxial magnetometer signal containing dynamic data of the X, Y, and Z axial components. When the signal strength stabilizes, indicating a vehicle has come to a complete stop, a pre-set time window for analysis (e.g., the first 3 minutes after stopping) is immediately initiated. High-precision time-domain analysis is performed on the triaxial magnetometer signal within this window. Within this window, the system accurately extracts three key behavioral features to form an initial behavioral pattern feature vector, including the time interval Δt, the number of signal disturbances N, and the steady-state coefficient R. Subsequently, the system inputs this feature vector V = [Δt, N, R] into a machine learning model (such as a gradient boosting decision tree regression model) pre-trained with a large amount of sample data (covering various vehicle models and different user behavior patterns). This model learns and maps the complex nonlinear relationship between the behavioral features and the final actual parking time, outputting a quantified estimated parking time. Finally, this prediction result is uniquely associated with the charging space number corresponding to the sensor that generated the signal and the ID of the charging pile connected to that space, thereby providing accurate, space-level data support for subsequent vehicle classification and power scheduling.

[0086] like Figure 3 As shown, in a preferred embodiment of the present invention, the step of continuously acquiring and generating triaxial magnetometer signals corresponding to the vehicle state specifically includes:

[0087] S1011, detect signal interference events based on whether the triaxial magnetometer signals of adjacent charging spaces show synchronous correlation fluctuations;

[0088] S1012, When a signal interference event occurs, the source parking space and the affected parking space are identified based on the amplitude and timing characteristics of the associated triaxial magnetometer signal changes.

[0089] S1013, Based on the state determination result of the interference source parking space, the triaxial magnetometer signal of the interference parking space is reconstructed through preset logic rules;

[0090] S1014 outputs the triaxial magnetometer signal after state determination and reconstruction processing.

[0091] This invention is specifically designed to address the problem of cross-interference between signals from adjacent parking spaces. First, signal interference event detection is performed: the system periodically compares the triaxial magnetometer signals of adjacent parking spaces (e.g., spaces numbered A01, A02, and A03). When a drastic change is detected in the signal of one sensor (e.g., parking space A01), and simultaneously, the signals of one or more adjacent sensors (e.g., parking space A02) also exhibit highly correlated fluctuations exceeding the normal noise range, a suspected signal interference event is identified. Then, the combined vector amplitude of these correlated signals is compared. ( The order of their changes, and the sequence of time, will determine the signal amplitude at its maximum (i.e., The parking space corresponding to the sensor that first changes (A01) is identified as the source of interference, while other parking spaces exhibiting similar fluctuations (A02) are marked as affected parking spaces. Then, the system enters the interference signal logic reconstruction stage. For the marked affected parking spaces, their distorted original signals are no longer used directly. Instead, based on the final state determination of the source parking space, preset logic rules are applied to reconstruct the signal. For example, if the source parking space (A01) is determined to change from "idle" to "occupied," the monitoring signal of the affected parking space (A02) will be logically reconstructed into a stable signal representing the "idle" state, and vice versa. The final output is a calibrated signal, a set of triaxial magnetometer signals that includes the real signal from the source parking space and the reconstructed signal from the affected parking spaces, excluding cross-interference. This provides a reliable and accurate data foundation for subsequent behavioral feature extraction and parking duration prediction.

[0092] like Figure 4 As shown, in a preferred embodiment of the present invention, the step of analyzing the triaxial magnetometer signal and extracting the initial behavior pattern feature vector including time interval, number of signal perturbations, and steady-state coefficient specifically includes:

[0093] S1021, when the amplitude change rate of the triaxial magnetometer signal is continuously lower than the first steady-state threshold for more than a set time, it is determined that the vehicle has reached a stationary state and this time is recorded as the reference time.

[0094] S1022, Based on the reference time, identify the time of the first significant signal disturbance caused by the car owner opening the door and getting out of the car and record it as the first time, wherein the time interval is the difference between the first time and the reference time;

[0095] S1023, based on the first moment, a second time window is preset, and the number of signal disturbances caused by human activities is counted within the second time window;

[0096] S1024, synchronously analyze the proportion of events in the second time window where the amplitude of the triaxial magnetometer signal is lower than the second steady-state threshold, and obtain the steady-state coefficient;

[0097] S1025, the time interval, the number of signal disturbances, and the steady-state coefficient are integrated to obtain the feature vector of the initial behavior pattern.

[0098] In this embodiment of the invention, the amplitude of the composite vector of the triaxial magnetometer signal is continuously monitored. When its rate of change remains below a first steady-state threshold for a set period of time, the vehicle is determined to have entered a stable parking state, and this moment is marked as the reference time for subsequent analysis. Next, event identification is performed: with Starting from the first moment, scan and identify the first significant signal disturbance caused by the owner opening the car door for the first time, and record its starting point as the first moment. And calculate the first key feature time interval. A pre-defined second time window is then initiated. Within this window, density analysis is performed to count the number of signal disturbances exceeding the event determination threshold caused by human activities (such as opening and closing car doors, retrieving and placing items), resulting in the second key feature: the number of signal disturbances, N. Simultaneously, steady-state analysis is performed in parallel within the same time window. Using a more sensitive second steady-state threshold as a benchmark, the total time the signal amplitude is below this threshold is accumulated, and the proportion of this total time in the second time window is calculated to obtain the third key feature: the steady-state coefficient R. This coefficient quantifies the stability of the magnetic field environment after the person leaves the vehicle. Finally, these three feature parameters (time interval ΔT, number of disturbances N, and steady-state coefficient R) characterizing the user's initial departure behavior from different dimensions are combined to construct a comprehensive initial behavior pattern feature vector V = [ΔT, N, R]. This feature vector effectively captures the essential differences in user behavior. For example, a typical "stop and go" behavior pattern is often characterized by a shorter ΔT (indicating the user is eager to leave), fewer N (indicating concise actions), and a higher R (indicating the user quickly moves away); conversely, a "long-term parking" behavior pattern may exhibit a moderate ΔT, more N (indicating actions such as tidying up items), and a relatively low R. This feature extraction method provides highly discriminative and interpretable input features for subsequent duration prediction models.

[0099] like Figure 5 As shown in the preferred embodiment of the present invention, the step of comparing the expected parking time with a preset threshold specifically includes:

[0100] S201, obtain the current total charging load power and the maximum power distribution capacity of the station, and calculate the real-time load factor;

[0101] S202, obtain the number of available charging spaces and the total number of charging spaces in the station, and calculate the parking space turnover urgency index;

[0102] S203, Based on the real-time load rate and parking space turnover urgency index, dynamically calculate the classification threshold for the current time using a preset mapping function;

[0103] S204, compare the estimated parking time of each vehicle with the dynamically generated classification threshold to classify the vehicles.

[0104] In this embodiment of the invention, the current total charging load power of the power station is first obtained in real time through load calculation. With maximum power distribution capacity Calculate the real-time load factor This accurately reflects the current power supply pressure level of the depot. Simultaneously, it allows for the urgent assessment of the number of available parking spaces within the depot. Total number of parking spaces Calculate the parking space turnover urgency index This is used to quantify the scarcity of parking space resources. Subsequently, a threshold calculation is performed, inputting the real-time load rate η and the parking space turnover urgency index θ into a preset mapping function F(η,θ) to dynamically calculate the most suitable classification threshold for the current moment. For example, when the station becomes busier (i.e., the values ​​of η and θ increase). The threshold is lowered accordingly, thus classifying more vehicles as "short-term parking vehicles" to accelerate resource turnover; conversely, the threshold is raised if the parking time is longer. During the dynamic classification phase, the system compares the estimated parking time of each vehicle with this dynamically generated classification threshold. The comparison is performed to complete the final vehicle classification. For example, when η > 0.85 and θ > 0.7, The timeframe may automatically adjust from the usual 2 hours to 1 hour, ensuring that vehicles expected to be parked for 70 minutes are correctly classified as short-term parking vehicles requiring priority service. This dynamic threshold mechanism ensures that vehicle classification always maintains an optimal match with the actual operational status of the depot.

[0105] like Figure 6 As shown in the preferred embodiment of the present invention, when a power grid peak shaving command is received and there are vehicles temporarily parked within the power station, the step of generating a differentiated power scheduling strategy based on the vehicle classification results specifically includes:

[0106] S301, obtain the upper limit total power of the substation corresponding to the power grid peak shaving command, as well as the infrastructure load of the substation and the minimum guaranteed total power allocated to all long-term parked vehicles, and calculate the guaranteed power budget corresponding to short-term parked vehicles.

[0107] S302 calculates the charging emergency index for each short-term parked vehicle based on its current battery status, the user-set target charging capacity, and the duration of connection.

[0108] S303, sort the short-term parked vehicles according to the charging emergency index to obtain an emergency dispatch queue and allocate charging power;

[0109] S304, the allocation results, the minimum guaranteed total power of long-term parked vehicles, and the infrastructure load are integrated to obtain the power scheduling strategy.

[0110] In this embodiment of the invention, when the system receives a power grid peak shaving command and there are vehicles classified as short-term parked within the power station, a dedicated conflict resolution mechanism will be activated. This mechanism first obtains the maximum total power limit of the power station as specified in the power grid command. And deduct the basic load of the station infrastructure from it. The sum of the minimum guaranteed power of all vehicles parked for extended periods This allows for the acquisition of a dedicated power budget for vehicles parked for short periods. Its calculation formula is Then, emergency quantification is performed. For each short-term parked vehicle, a charging emergency index is calculated using a pre-set algorithm based on its current battery state of charge, the user-set target charging capacity, and the duration of connection. This index is directly proportional to the remaining demand for electricity and inversely proportional to the duration of connection; that is, vehicles with more urgent needs and shorter waiting times receive a higher index. Then, arbitration allocation is performed. All short-term parked vehicles are arranged in descending order of their charging emergency index to form an emergency dispatch queue. Charging power is then allocated sequentially within this queue until the guaranteed power budget is reached. All allocations are complete. Finally, through strategy integration, the above allocation results, the guaranteed power scheme for long-term parked vehicles, and the infrastructure load are unified to generate a final power scheduling strategy that satisfies both grid peak shaving requirements and maximizes the guarantee of emergency charging needs. For example, in a certain peak shaving command, the calculated... With a power rating of 50kW, the maximum acceptable power will be allocated to vehicles with the highest emergency index (such as ride-hailing vehicles with a SOC of only 15% that have just joined the network), while vehicles with a lower emergency index (such as private cars with a SOC of 50% that have been connected for 30 minutes) will be allocated less power or not at all, thereby achieving the optimal allocation of limited resources under extreme power constraints.

[0111] like Figure 7 As shown in the figure, this embodiment of the invention also provides a power dispatching and control system for charging stations, the system comprising:

[0112] The status monitoring module 100 is used to generate the estimated parking time for each vehicle in each charging space based on real-time monitoring of the status of each charging space in the station.

[0113] The vehicle classification module 200 is used to compare the estimated parking time with a preset threshold to classify vehicles as short-term parking vehicles or long-term parking vehicles.

[0114] The strategy generation module 300 is used to generate differentiated power scheduling strategies based on vehicle classification results. The power scheduling strategies include prioritizing the allocation of high power for fast charging of short-term parked vehicles and planning charging time windows for long-term parked vehicles, and enabling them to charge during periods of low electricity load or low electricity price.

[0115] The instruction execution module 400 is used to generate specific power control instructions according to the power scheduling strategy and send them to the corresponding charging piles for execution.

[0116] In this embodiment of the invention, intelligent power dispatching of charging stations is achieved through the coordinated operation of four core modules. The status monitoring module 100 collects parking space status data in real time using a geomagnetic sensor and generates the estimated parking duration for vehicles; the vehicle classification module 200 receives the estimated parking duration data and dynamically adjusts the classification threshold based on the station's real-time operating parameters to complete the intelligent classification of vehicle types; the strategy generation module 300 generates a differentiated power dispatching scheme that balances efficiency and cost based on the classification results, taking into account grid commands, user needs, and economic factors; and the command execution module 400 converts the optimized strategy into executable control commands to drive the charging piles to complete power allocation.

[0117] like Figure 8 As shown, in a preferred embodiment of the present invention, the status monitoring module 100 includes:

[0118] The signal acquisition unit 101 is used to continuously acquire and generate triaxial magnetometer signals corresponding to the vehicle status through a geomagnetic sensor deployed in the charging parking space.

[0119] Feature extraction unit 102 is used to analyze the triaxial magnetometer signal and extract the initial behavior pattern feature vector, including time interval, number of signal disturbances and steady-state coefficient, within the first time window after the vehicle stops. The time interval is the time from when the vehicle stops until the owner opens the door and gets out of the vehicle. The number of signal disturbances is the signal fluctuation caused by the opening and closing of the door. The steady-state coefficient characterizes the signal stability after the person moves away from the vehicle.

[0120] The duration prediction unit 103 is used to input the initial behavior pattern feature vector into the pre-trained duration prediction model to obtain the estimated parking time of the vehicle.

[0121] The information binding unit 104 is used to associate and bind the estimated parking time with the corresponding vehicle, charging space and charging pile information.

[0122] like Figure 9 As shown, in a preferred embodiment of the present invention, the vehicle classification module 200 includes:

[0123] The load calculation unit 201 is used to obtain the current total charging load power and the maximum power distribution capacity of the station, and to calculate the real-time load factor.

[0124] Urgency assessment unit 202 is used to obtain the number of vacant charging spaces and the total number of charging spaces in the station, and to calculate the parking space turnover urgency index.

[0125] The threshold calculation unit 203 is used to dynamically calculate the classification threshold for the current time based on the real-time load rate and parking space turnover urgency index using a preset mapping function.

[0126] The dynamic classification unit 204 is used to compare the estimated parking time of each vehicle with the dynamically generated classification threshold to classify the vehicles.

[0127] like Figure 10 As shown, in a preferred embodiment of the present invention, the strategy generation module 300 includes:

[0128] Budget calculation unit 301 is used to obtain the upper limit total power of the station corresponding to the power grid peak shaving command, as well as the infrastructure load of the station and the minimum guaranteed total power allocated to all long-term parked vehicles, and calculate the guaranteed power budget corresponding to short-term parked vehicles.

[0129] Emergency quantification unit 302 is used to calculate the charging emergency index for each short-term parked vehicle based on its current battery status, the target charging capacity set by the user, and the duration of connection.

[0130] The sorting and allocation unit 303 is used to sort the short-term parked vehicles according to the charging emergency index, obtain an emergency dispatch queue, and allocate charging power.

[0131] The strategy integration unit 304 is used to integrate the allocation results, the minimum guaranteed total power of long-term parked vehicles, and the infrastructure load to obtain the power scheduling strategy.

[0132] The above description only details the preferred embodiments of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

[0133] It should be understood that although the steps in the flowcharts of the various embodiments of the present invention are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the various embodiments may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least a portion of the sub-steps or stages of other steps.

[0134] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0135] Other embodiments of this disclosure will readily occur to those skilled in the art upon consideration of the disclosure in the specification and embodiments. This application is intended to cover any variations, uses, or adaptations of this disclosure that follow the general principles of this disclosure and include common knowledge or customary techniques in the art not disclosed herein. The specification and embodiments are to be considered exemplary only, and the true scope and spirit of this disclosure are indicated by the claims.

Claims

1. A power dispatch control method for a charging station, characterized by, The method includes the following steps: Based on real-time monitoring of the status of each charging space in the station, the estimated parking time for each vehicle in each charging space is generated. The estimated parking time is compared with a preset threshold to classify vehicles as either short-term or long-term parked vehicles. Based on the vehicle classification results, a differentiated power scheduling strategy is generated. The power scheduling strategy includes prioritizing the allocation of high power for fast charging of short-term parked vehicles and planning charging time windows for long-term parked vehicles, so that they can be charged during the off-peak hours of electricity load or electricity price. Specific power control commands are generated according to the power scheduling strategy and sent to the corresponding charging piles for execution. The step of generating the estimated parking time for each vehicle in a charging space based on real-time monitoring of the status of each charging space within the station specifically includes: The system continuously collects and generates triaxial magnetometer signals corresponding to the vehicle status by deploying geomagnetic sensors in the charging parking spaces. Within the first time window after the vehicle stops, the triaxial magnetometer signal is analyzed and an initial behavior pattern feature vector including time interval, number of signal disturbances and steady-state coefficient is extracted. The time interval is the time from when the vehicle stops until the owner opens the door and gets out of the vehicle. The number of signal disturbances is the signal fluctuation caused by the opening and closing of the door. The steady-state coefficient characterizes the signal stability after the person moves away from the vehicle. The initial behavior pattern feature vector is input into a pre-trained duration prediction model to obtain the vehicle's estimated parking time. The estimated parking time is associated with and bound to the corresponding vehicle, charging space, and charging pile information; The step of comparing the expected parking time with a preset threshold specifically includes: Obtain the current total charging load power and the maximum power distribution capacity of the power station, and calculate the real-time load factor; Obtain the number of available charging spaces and the total number of charging spaces in the station, and calculate the parking space turnover urgency index; Based on the real-time load rate and parking space turnover urgency index, the classification threshold for the current time is dynamically calculated using a preset mapping function. The estimated parking time of each vehicle is compared with the dynamically generated classification threshold to classify the vehicles.

2. The power dispatch control method for a charging station of claim 1, wherein, The step of continuously acquiring and generating triaxial magnetometer signals corresponding to the vehicle state specifically includes: The system detects signal interference events by checking whether the triaxial magnetometer signals of adjacent charging spaces exhibit synchronous correlation fluctuations. When a signal interference event occurs, the source parking space and the affected parking space are identified based on the amplitude and timing characteristics of the associated triaxial magnetometer signal changes. Based on the status determination result of the parking space of the interference source, the triaxial magnetometer signal of the interference parking space is reconstructed through preset logic rules; Output the triaxial magnetometer signal after state determination and reconstruction processing.

3. The power dispatch control method for a charging station of claim 1, wherein, The steps of analyzing the triaxial magnetometer signal and extracting the initial behavior pattern feature vector, including time interval, number of signal disturbances, and steady-state coefficient, specifically include: When the amplitude change rate of the triaxial magnetometer signal is continuously lower than the first steady-state threshold for more than a set time, it is determined that the vehicle has come to a stop and this time is recorded as the reference time. Based on the reference time, the time of the first significant signal disturbance caused by the car owner opening the door and getting out of the car is identified and recorded as the first time, and the time interval is the difference between the first time and the reference time. Based on the first moment, a second time window is preset, and the number of signal disturbances caused by human activities is counted within the second time window. The steady-state coefficient is obtained by synchronously analyzing the proportion of events in the second time window where the amplitude of the triaxial magnetometer signal is lower than the second steady-state threshold. The initial behavior pattern feature vector is obtained by integrating the time interval, the number of signal disturbances, and the steady-state coefficient.

4. The power dispatch control method for a charging station of claim 1, wherein, When a power grid peak shaving command is received and there are vehicles temporarily parked within the power station, the step of generating a differentiated power scheduling strategy based on the vehicle classification results specifically includes: Obtain the maximum total power of the power station corresponding to the power grid peak shaving command, as well as the infrastructure load of the power station and the minimum guaranteed total power allocated to all long-term parked vehicles, and calculate the guaranteed power budget corresponding to short-term parked vehicles. For each vehicle that is parked for a short period of time, a charging emergency index is calculated based on its current battery status, the target charging capacity set by the user, and the duration of connection. Based on the charging emergency index, short-term parked vehicles are sorted to obtain an emergency dispatch queue and charging power is allocated. The power scheduling strategy is obtained by integrating the allocation results, the minimum guaranteed total power of vehicles parked for a long time, and the infrastructure load.

5. A power dispatch control system for a charging station, characterized by, The system includes: The status monitoring module is used to generate the estimated parking time for each vehicle in each charging space based on real-time monitoring of the status of each charging space in the station. The vehicle classification module is used to compare the estimated parking time with a preset threshold to classify vehicles as either short-term or long-term parking vehicles. The strategy generation module is used to generate differentiated power scheduling strategies based on vehicle classification results. The power scheduling strategies include prioritizing the allocation of high power for fast charging of short-term parked vehicles and planning charging time windows for long-term parked vehicles, and enabling them to charge during periods of low electricity load or low electricity price. The instruction execution module is used to generate specific power control instructions according to the power scheduling strategy and send them to the corresponding charging piles for execution. The status monitoring module includes: The signal acquisition unit is used to continuously acquire and generate triaxial magnetometer signals corresponding to the vehicle status through geomagnetic sensors deployed in the charging parking spaces. The feature extraction unit is used to analyze the triaxial magnetometer signal and extract the initial behavior pattern feature vector, including time interval, number of signal disturbances and steady-state coefficient, within the first time window after the vehicle stops. The time interval is the time from when the vehicle stops until the owner opens the door and gets out of the vehicle. The number of signal disturbances is the signal fluctuation caused by the opening and closing of the door. The steady-state coefficient characterizes the signal stability after the person moves away from the vehicle. The duration prediction unit is used to input the initial behavior pattern feature vector into the pre-trained duration prediction model to obtain the estimated parking time of the vehicle. The information binding unit is used to associate and bind the estimated parking time with the corresponding vehicle, charging space and charging pile information; The vehicle classification module includes: The load calculation unit is used to obtain the current total charging load power and the maximum power distribution capacity of the station, and to calculate the real-time load factor. The urgency assessment unit is used to obtain the number of vacant charging spaces and the total number of charging spaces in the station, and to calculate the parking space turnover urgency index. The threshold calculation unit is used to dynamically calculate the classification threshold for the current time based on the real-time load rate and parking space turnover urgency index using a preset mapping function. The dynamic classification unit is used to compare the estimated parking time of each vehicle with the dynamically generated classification threshold to classify the vehicles.

6. The power dispatch control system for a charging station of claim 5, wherein, The strategy generation module includes: The budget calculation unit is used to obtain the upper limit total power of the power station corresponding to the power grid peak shaving command, as well as the infrastructure load of the power station and the minimum guaranteed total power allocated to all long-term parked vehicles, and calculate the guaranteed power budget corresponding to short-term parked vehicles. The emergency quantification unit is used to calculate the charging emergency index for each short-term parked vehicle based on its current battery status, the user-set target charging capacity, and the duration of connection. The sorting and allocation unit is used to sort the short-term parked vehicles according to the charging emergency index, obtain an emergency dispatch queue, and allocate charging power. The strategy integration unit is used to integrate the allocation results, the minimum guaranteed total power of long-term parked vehicles, and the infrastructure load to obtain the power scheduling strategy.

Images