Data processing method for parallel charging of charging piles

Abstract

This invention discloses a data processing method for concurrent charging of charging piles, relating to the field of concurrent charging technology. The method includes the following steps: based on power request behavior characteristics, performing feature fusion operations to generate a coordination adaptation factor reflecting the inherent consistency of power request changes, used to characterize the endogenous coordination attributes of each power request when a vehicle alternately issues multiple power requests in concurrent charging. Based on the correspondence between the coordination adaptation factor and a preset confidence interval, the coordination attributes of each power request are determined, and a corresponding coordination confidence level label is generated to distinguish the coordination attribute categories of power requests in concurrent charging. This invention solves the problem of inaccurate identification of alternating power request behavior within a vehicle during concurrent charging, achieving precise determination and control of power request coordination attributes, and improving the stability of concurrent charging scheduling and resource allocation efficiency.

Description

Technical Field

[0001] This invention relates to the field of parallel charging technology, and more specifically to a data processing method for parallel charging of charging piles. Background Technology

[0002] Data processing for parallel charging refers to the system's comprehensive control over the entire process of charging interface functions, including power distribution, status monitoring, charging scheduling, communication coordination, and billing management, when multiple electric vehicles simultaneously connect to the same charging pile or the same group of charging piles for parallel charging. This is achieved through specific data acquisition, processing, and control mechanisms. Existing parallel charging data processing technologies typically utilize a control system embedded in the charging pile's main control module, along with smart meters, power control modules, and communication modules, to monitor the charging status of each connected vehicle in real time. This includes, but is not limited to, vehicle access identification, required charging power assessment, current power supply capacity analysis, dynamic power allocation strategy execution, charging priority adjustment, abnormal power outage handling, and two-way communication with the upper-level monitoring platform. The entire data processing flow generally includes the following key stages: First, the data acquisition stage, which uses sensors and communication modules to acquire information such as current, voltage, charging status, and user ID in real time; second, the data analysis and decision-making stage, where the system formulates concurrent charging strategies based on factors such as vehicle charging demand, grid load, and electricity pricing policies, such as using methods like rotating charging, load balancing, and dynamic time-sharing; third, the control execution stage, which adjusts the output power and on / off status of each charging port according to the analysis results; and finally, the data feedback and recording stage, which synchronously uploads charging data, billing information, and charging logs to the backend management system or cloud platform for user query, operational analysis, and fault diagnosis. Existing technologies generally adopt centralized or distributed architectures, combined with communication standards such as OCPP (Open Charging Protocol), to ensure system stability and data consistency in concurrent multi-vehicle charging scenarios.

[0003] The existing technology has the following shortcomings:

[0004] During the data processing of parallel charging at charging stations, certain vehicle models, due to the special design of their onboard energy management strategies, will alternately send two or more power request values ​​with slight differences during the initial charging phase to achieve internal dynamic power balance control. When such vehicles are simultaneously charging with other vehicles, their power request behavior is processed concurrently with the requests of other vehicles in the system, forming a complex request sequence interaction, which is easily misidentified as a power conflict by the data processing system. Because existing data processing technologies for parallel charging at charging stations are usually based on an immediate response strategy for single power requests, they lack the ability to identify continuous power request sequence change patterns. Therefore, they cannot determine whether each power request belongs to a coordinateable internal power adjustment strategy based on the power request behavior characteristics of vehicles alternately issuing multiple power requests during parallel charging. This lack of identification capability will lead to misjudgments in the system's power scheduling decisions, incorrectly identifying coordinateable internal request behavior as a power conflict, thereby frequently triggering power reallocation operations or interrupting the scheduling process, causing instability in the parallel charging process, affecting the rational allocation efficiency of power resources, ultimately reducing overall charging performance and harming the user experience.

[0005] The information disclosed in the background section is only intended to enhance the understanding of the background of this disclosure, and therefore may include information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0006] The purpose of this invention is to provide a data processing method for charging piles and charging simultaneously, so as to solve the problems in the background art mentioned above.

[0007] To achieve the above objectives, the present invention provides the following technical solution: a data processing method for concurrent charging of charging piles, specifically including the following steps:

[0008] S1. Continuously collect power request data of vehicles in parallel charging state at charging piles, and construct a power request alternation detection sequence according to the time sequence. By analyzing the back-and-forth switching relationship of power requests in the detection sequence, determine whether the vehicle alternately issues multiple power requests in parallel charging state at charging piles.

[0009] S2. In the case of multiple power requests being issued alternately, an evolutionary analysis is performed on the alternating detection sequence of power requests to extract the alternating amplitude features, change rhythm features, and directional consistency features that reflect the changing pattern of power requests, forming power request behavior features used to characterize the changing behavior of power requests.

[0010] S3. Based on the power request behavior characteristics, perform feature fusion operation to generate a coordination adaptation factor that reflects the inherent consistency of power request changes. This factor is used to characterize the endogenous coordination attributes of each power request when the vehicle alternately issues multiple power requests in the state of charging piles in parallel charging.

[0011] S4. Based on the correspondence between the coordination adaptation factor and the preset confidence interval, determine the coordination attribute of each power request and generate the corresponding coordination confidence level mark to distinguish the coordination attribute category of the power request in the charging pile parallel charging state.

[0012] S5. Based on the coordination trust level label and the power resource occupancy status of the charging pile in the parallel charging state, dynamically adjust the response order and response rhythm of power requests so that power requests participate in power allocation according to the coordination attribute category during the parallel charging process.

[0013] Preferably, S1 is as follows:

[0014] The system continuously collects power request data issued by vehicles in parallel charging state during the charging initialization phase, records each power request data in the actual order of its generation, and forms a power request data stream arranged in chronological order.

[0015] Based on the power request data stream, adjacent power request data are compared one by one to identify the direction and magnitude of change in the numerical value of the power request data. Request nodes with numerical rotation characteristics are extracted from the power request data stream, and a power request alternation detection sequence is constructed in chronological order.

[0016] The alternating power request detection sequence is traversed and analyzed. When the change amplitude between two or more power request data in the detection sequence is lower than the preset difference threshold and the change direction is periodically alternating, it is determined that the vehicle is alternately issuing multiple power requests in the charging pile's parallel charging state.

[0017] Preferably, S2 is as follows:

[0018] When it is confirmed that the vehicle is issuing multiple power requests alternately, the alternating power request detection sequence is continuously scanned, the difference between adjacent power request data is extracted, and the mean and standard deviation of the difference within a preset fixed time window are calculated to form the alternation amplitude feature.

[0019] Based on the acquisition time points of adjacent power requests in the alternating power request detection sequence, the average interval period of the alternation direction of power requests is calculated, and the number of occurrences of the alternation direction within the same period is counted to construct the change rhythm characteristics.

[0020] The alternation amplitude feature and the change rhythm feature are jointly segmented and processed. The consistency of the alternation direction of power request is statistically analyzed to generate the direction consistency feature. The three features are then combined into a power request behavior feature to characterize the power request change behavior.

[0021] Preferably, S3 specifically includes the following steps:

[0022] S301. Normalize the alternating amplitude features, change rhythm features, and direction consistency features, convert them into vector components under a unified scale, and construct a power request behavior feature vector for feature fusion operation.

[0023] S302. Using the power request behavior feature vector as input, each feature component is fused according to the weighted calculation rules to extract stability index and correlation index, which are used to generate a coordination adaptation factor that reflects the inherent consistency of power request changes.

[0024] S303. The coordination adaptation factor is associated and matched with the power request behavior feature vector to identify the inherent pattern in the power request behavior, and the coordination adaptation factor is used to characterize the endogenous coordination attribute of each power request when the vehicle alternately issues multiple power requests in the state of charging piles.

[0025] Preferably, S302 is as follows:

[0026] Using the power request behavior feature vector as input, the alternating amplitude feature, the change rhythm feature and the direction consistency feature are respectively processed by interval discretization to obtain the value distribution sequence of each feature component in multiple time segments, which is used to characterize the degree of concentration of the feature component changes.

[0027] Based on the value distribution sequence, the discrete distribution parameter and time fluctuation parameter are calculated for each feature component. According to the pre-set weighting calculation rules, the discrete distribution parameter and time fluctuation parameter are mapped to the corresponding weight coefficients, and a weighted fusion operation is performed to generate a stability index.

[0028] Based on the generation of stability indices, a pairwise analysis is performed on the changing trends of each feature component in the power request behavior feature vector to calculate the degree of trend consistency, generate correlation indices, and combine the stability indices with the correlation indices to form a coordination adaptation factor, which is used to characterize the inherent consistency of power request changes.

[0029] Preferably, S303 is as follows:

[0030] The range of values ​​for the coordination adaptation factor is divided into multiple fixed-level segments. Each feature dimension in the power request behavior feature vector is segmented over time. The average value and fluctuation value of the feature values ​​in each segment are calculated, and this set of data is used as the feature expression vector.

[0031] The numerical distance between each feature expression vector and the center value of each level segment is calculated to determine the distance relationship between it and each level segment. The level segment number that meets the minimum distance requirement is extracted as the candidate level index, and the corresponding coordination matching label is constructed.

[0032] A label fitting operation is performed on the behavioral feature trajectory of each power request node. Power request nodes with fitting errors less than a preset error limit are selected. Based on label consistency, the corresponding coordination adaptation factor is assigned to the node. The coordination adaptation factor is used to characterize the endogenous coordination attribute of each power request when the vehicle alternately issues multiple power requests in the state of parallel charging at the charging pile.

[0033] Preferably, S4 is as follows:

[0034] Based on the coordination adaptation factor data of multiple historical power requests, multiple preset confidence intervals are established, and a unique coordination attribute category number is assigned to each preset confidence interval to construct the correspondence between coordination adaptation factors and confidence intervals.

[0035] For each power request in the parallel charging state of the charging pile, the corresponding coordination adaptation factor value is extracted, and the value is compared with multiple preset confidence intervals. Based on the preset confidence interval range in which it falls, the coordination attribute judgment is performed to determine the coordination attribute category to which the power request belongs.

[0036] Based on the coordination attribute category to which the power request belongs, a coordination trust level tag matching the power request is generated as a unique code to identify the coordination attribute category of the power request, thus completing the differentiation of the coordination attribute category of the power request in the charging pile's parallel charging state.

[0037] Preferably, S5 is as follows:

[0038] The system collects real-time power resource occupancy data of charging piles in parallel charging mode, divides the current total power resources into multiple coordination attribute categories for allocation channels, and establishes a candidate response sequence based on the coordination trust level label corresponding to each power request.

[0039] The response order of the candidate response sequences is adjusted according to the coordination trust level label. The matching response rhythm control parameters are calculated by combining the allocated channel capacity and power resource occupancy ratio corresponding to each coordination attribute category, and a control configuration containing order and rhythm is generated.

[0040] Based on the dynamic control of the regulation configuration, the response order and response rhythm of each power request during the parallel charging process are controlled, so that power requests with different coordination trust level labels participate in power resource allocation according to the coordination attribute category.

[0041] The technical effects and advantages provided by the present invention in the above technical solution are as follows:

[0042] 1. This invention achieves accurate identification and classification of multiple alternating power request behaviors issued by vehicles in parallel charging scenarios by constructing a feature extraction and fusion mechanism based on power request behavior sequences. By modeling the power request data in three dimensions—alternating amplitude, rhythm of change, and directional consistency—and further generating a coordination adaptation factor reflecting the inherent consistency of power request changes, the system can determine from a behavioral perspective whether a power request belongs to a coordinateable behavior caused by the vehicle's internal control strategy, thereby effectively avoiding misidentifying normal power adjustment behavior as a power conflict. Compared to the immediate response strategy based on a single request in existing technologies, this solution introduces a continuous, multi-dimensional behavior analysis method, significantly improving the accuracy and context awareness of the data processing system in complex parallel charging scenarios.

[0043] 2. This invention constructs a coordination trust level labeling system to dynamically associate power requests of different coordination attribute categories with the current power resource occupancy of charging piles. Power resource allocation channels are divided according to coordination attribute categories, and the response order and rhythm are dynamically adjusted based on the allocation channel capacity and power request level priority. This differentiated scheduling mechanism not only improves the efficiency of power resource allocation but also enhances the system's adaptability to various vehicle interaction behaviors, reduces unnecessary power reallocation or scheduling interruptions, and improves the stability and consistency of the overall parallel charging process. It has good engineering practicality and user experience optimization value. Attached Figure Description

[0044] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this invention. For those skilled in the art, other drawings can be obtained based on these drawings.

[0045] Figure 1 This is a flowchart illustrating the data processing method for parallel charging of the charging pile of the present invention. Detailed Implementation

[0046] Exemplary embodiments will now be described more fully with reference to the accompanying drawings. However, these exemplary embodiments can be implemented in many forms and should not be construed as limited to the examples set forth herein; rather, they are provided so that the description of this disclosure will be more complete and fully convey the concept of the exemplary embodiments to those skilled in the art.

[0047] This invention provides, for example Figure 1 The data processing method for parallel charging of the charging piles shown includes the following steps:

[0048] S1. Continuously collect power request data of vehicles in parallel charging state at charging piles, and construct a power request alternation detection sequence according to the time sequence. By analyzing the back-and-forth switching relationship of power requests in the detection sequence, determine whether the vehicle alternately issues multiple power requests in parallel charging state at charging piles.

[0049] In this embodiment, S1 specifically refers to:

[0050] The system continuously collects power request data issued by vehicles in parallel charging state during the charging initialization phase, records each power request data in the actual order of its generation, and forms a power request data stream arranged in chronological order.

[0051] To accurately identify the power request behavior of vehicles charging in parallel at charging stations, it is first necessary to continuously monitor all power request data issued by each vehicle during the charging initialization phase. This data can be obtained by reading the power request field in the real-time communication messages between the vehicle and the charging station. This communication process is typically based on protocols such as ISO 15118 or DIN 70121. After the vehicle initiates the charging handshake, it periodically reports the desired charging power. By adding a timestamp recording mechanism to the communication parsing process, it can be ensured that each power request data corresponds to its issuance time. Arranging these power request data in chronological order can construct a power request data stream reflecting the power change trajectory. Taking a certain vehicle as an example, it may send data such as "42kW, 43kW, 42kW, 43kW, 42kW" in sequence during the initialization phase. If the time series is ignored, this repetitive behavior cannot be identified. Therefore, recording the power request data in the actual order of generation and organizing it into a time series is a necessary prerequisite for subsequent determination of whether the vehicle is alternately issuing multiple power request behaviors.

[0052] Concurrent charging refers to the operational state where a single charging station simultaneously provides charging services to multiple electric vehicles. In this state, a single device needs to collaboratively process multiple power requests. The charging initialization phase refers to the process from when a vehicle connects to the charging gun until the system confirms that charging can begin. It includes steps such as authentication, parameter negotiation, and power request, and is a crucial stage for the initial expression of the vehicle's power intentions. Power request data refers to the numerical information sent by the vehicle to the charging station to express its desired electrical output power, usually in kilowatts, and is dynamic and real-time. The power request data stream is an ordered data set formed by combining continuously collected power request data in chronological order. It accurately reflects the changing trend of the vehicle's power demand within a specific period and is the core foundation for identifying alternating characteristics. By constructing this data stream, not only can numerical fluctuations be observed, but their rhythm and patterns can also be analyzed, providing initial evidence for subsequent alternating behavior identification.

[0053] Based on the power request data stream, adjacent power request data are compared one by one to identify the direction and magnitude of change in the numerical value of the power request data. Request nodes with numerical rotation characteristics are extracted from the power request data stream, and a power request alternation detection sequence is constructed in chronological order.

[0054] Before constructing the alternating power request probe sequence, it is necessary to compare adjacent power request data in the power request data stream one by one. The comparison method is as follows: for power request data arranged chronologically, compare each pair sequentially, recording the direction (increasing or decreasing) and magnitude (absolute difference) of the numerical change of the subsequent item compared to the previous item. For example, if a vehicle sends five consecutive power request data during the initialization phase: 41.8, 42.1, 41.7, 42.2, and 41.6, the direction of change between adjacent data is increasing, decreasing, increasing, decreasing, with magnitudes of 0.3, 0.4, 0.5, and 0.6 respectively. When the direction of change alternates continuously (i.e., increasing then decreasing, then increasing again, then decreasing again), and the magnitudes fluctuate within a small range, these nodes can be identified as exhibiting numerical reversal characteristics. After extracting these request data nodes that meet the criteria, the alternating power request probe sequence can be organized chronologically. The purpose of constructing this sequence is to extract datasets with stable alternating behavior from the entire data stream, providing accurate input for subsequent behavioral feature modeling.

[0055] One-by-one comparison refers to using a paired comparison method to analyze the differences between each data point and its predecessor in a continuously arranged power request data set, thereby identifying the trend of change. The direction of change in the numerical value of the power request data is used to characterize whether the power request is increasing or decreasing, and is a prerequisite feature for alternating behavior; while the magnitude of change reflects the degree of fluctuation between adjacent data points, and is the core parameter for judging the stability of change. Request nodes with numerical rotation characteristics are those power request data points that repeatedly alternate in the direction of change and maintain a continuous small fluctuation range in the magnitude of change. The power request alternation detection sequence is an ordered set composed of these data points with numerical rotation characteristics. It can extract the part with regular change characteristics from the full power request data, and is the key foundation for subsequent behavior recognition, feature extraction, and coordination judgment.

[0056] The alternating power request detection sequence is traversed and analyzed. When the change amplitude between two or more power request data in the detection sequence is lower than the preset difference threshold and the change direction is periodically alternating, it is determined that the vehicle is alternately issuing multiple power requests in the charging pile's parallel charging state.

[0057] The specific implementation of the traversal analysis of the alternating power request detection sequence is as follows: each power request data is read sequentially according to time order, and the numerical change between the current data and the previous data is calculated. The magnitude of the change can be obtained by calculating the absolute value of the difference between the two, while the direction of change is determined by the magnitude relationship between the previous data and the current data. If multiple power request data show a change magnitude lower than a preset difference threshold in the entire detection sequence, and the change direction presents an alternating pattern of "rise-fall-rise-fall" or "fall-rise-fall-rise", and this pattern lasts for at least three sets, then the detection sequence can be considered to have a periodic alternating relationship. For example, if the data in the detection sequence are 42.0, 42.3, 41.9, 42.2, and 41.8, and the change magnitude of adjacent data is all less than the set 0.6kW threshold, and the change direction is rise, fall, rise, fall, then it can be preliminarily determined that the vehicle is alternately issuing multiple power requests in the current parallel charging state. This behavior is most likely a dynamic power adjustment performed by the vehicle to adapt to the current load or battery state, rather than a real power conflict request. The purpose of this judgment is to distinguish between coordination and resource competition behaviors, thus providing a basis for subsequent regulatory strategies.

[0058] Erratic analysis refers to the process of scanning and comparing all data points in the alternating power request detection sequence item by item. By continuously calculating the amplitude and direction of change for each pair of data, the inherent change pattern of the sequence is identified. The preset difference threshold is a numerical limit used to determine whether the fluctuation of power request data is within an acceptable range. It is usually set based on historical data statistical analysis or charging control requirements and has a fixed numerical boundary, such as 0.5kW or 0.6kW. Periodic alternation relationship refers to the stable up-and-down switching pattern of power request data in the direction of change, and this pattern repeats in a certain number of data points, exhibiting a clear rhythm. This relationship is an important basis for determining whether a vehicle has inherent dynamic power adjustment behavior and is also the dividing line between coordinated and conflicting request patterns. By identifying periodic alternation relationships, incorrect power redistribution judgments can be avoided for such requests, thereby improving the stability and accuracy of the scheduling system in parallel charging scenarios.

[0059] S2. In the case of multiple power requests being issued alternately, an evolutionary analysis is performed on the alternating detection sequence of power requests to extract the alternating amplitude features, change rhythm features, and directional consistency features that reflect the changing pattern of power requests, forming power request behavior features used to characterize the changing behavior of power requests.

[0060] In this embodiment, S2 specifically refers to:

[0061] When it is confirmed that the vehicle is issuing multiple power requests alternately, the alternating power request detection sequence is continuously scanned, the difference between adjacent power request data is extracted, and the mean and standard deviation of the difference within a preset fixed time window are calculated to form the alternation amplitude feature.

[0062] Assuming a vehicle is issuing multiple power requests alternately, the process of continuously scanning the alternating power request detection sequence can be accomplished by iterating through the power request data arranged in chronological order. Specifically, during the scanning process, the current power request data is sequentially selected and compared with the power request data corresponding to its previous moment, and the numerical difference between the two is calculated to reflect the variation amplitude between adjacent power requests. Subsequently, multiple differences obtained over a continuous period are aggregated into the same time interval, and the differences within this time interval are statistically analyzed to calculate their average variation level and dispersion. For example, when a vehicle issues power requests of 42.1, 42.4, 42.0, 42.3, and 41.9 consecutively during the initialization phase, the differences between adjacent data show small fluctuations. By statistically analyzing these differences within a fixed time window, it can be found that the overall variation amplitude is concentrated and stable. The purpose of this processing method is to transform the instantaneous fluctuations of a single power request into quantifiable and comparable overall variation characteristics, thereby providing a reliable basis for distinguishing between random fluctuations and regular alternating behavior.

[0063] The difference between adjacent power request data refers to the numerical difference between two consecutive power request values ​​in time. This difference directly reflects the degree of change in power requests at adjacent moments and is a fundamental parameter for characterizing the strength of power request fluctuations. A preset fixed time window refers to a continuously defined time range on the time axis, used for unified analysis of multiple differences generated within a certain time period. The purpose of this time window is to avoid excessive influence of a single abnormal change on the overall judgment, making the analysis results more stable and representative. Alternating amplitude characteristics are a comprehensive description formed by statistically processing multiple adjacent differences within the time window. They characterize the concentration and dispersion of the overall fluctuation amplitude of power requests during alternating changes. This feature reflects whether power requests consistently change within a small range, which is an important basis for judging whether power request changes have internal regulatory attributes and a key prerequisite for subsequently extracting change rhythm characteristics and directional consistency characteristics.

[0064] Based on the acquisition time points of adjacent power requests in the alternating power request detection sequence, the average interval period of the alternation direction of power requests is calculated, and the number of occurrences of the alternation direction within the same period is counted to construct the change rhythm characteristics.

[0065] After constructing the alternating power request detection sequence, continuous time interval data can be obtained by extracting the acquisition time point of each power request data point in the sequence and comparing the time difference between two adjacent changes in the alternating direction of power requests. For example, when the power request direction changes from rising to falling or from falling to rising, the time of this change is recorded, and the difference is calculated with the time of the previous change to obtain the interval period of this alternation. By statistically processing several time intervals of alternating direction changes and calculating the average of these time intervals, the average interval period of alternating direction changes can be obtained. Subsequently, using this average interval period as the benchmark unit, multiple equal-length period intervals are divided within the time range of the entire alternating detection sequence, and the number of alternating direction changes occurring in each period interval is counted. If the frequency of alternating direction changes is relatively regular in each period, it indicates that the power request sequence has strong rhythmic consistency. In this way, the rhythmic characteristics of the power request change process can be extracted, which can then be used to determine whether the power request is due to the vehicle's intentional control logic rather than abnormal disturbances.

[0066] The average interval period of alternating power request directions is obtained by measuring the time difference between power request changes in adjacent directions and averaging the values. This characteristic reflects the periodic regularity of power request throughout the entire change process and is a core indicator for assessing rhythm stability. The frequency of alternating directions occurring within equal periods refers to the frequency of power request direction switching in each time period divided by the average interval period. This frequency is used to measure whether there are regular fluctuations in power request within each period. The change rhythm characteristic is a dual description of the time interval distribution characteristics of alternating directions and the frequency of change within the period, possessing dual dimensions of time and direction. It is a key dimension for determining whether power request behavior originates from the vehicle's intrinsic rhythm regulation mechanism. If a vehicle's power request exhibits regular alternating behavior across multiple periods, it indicates that its power request changes are not occasional disturbances but the result of stable rhythm control. This rhythmicity is an important reference basis for subsequent coordination judgments.

[0067] The alternation amplitude feature and the change rhythm feature are jointly segmented and processed. The consistency of the alternation direction of power request is statistically analyzed to generate the direction consistency feature. The three features are then combined into a power request behavior feature to characterize the power request change behavior.

[0068] After obtaining the alternation amplitude and rhythm characteristics, a synchronous comparison of the two over time is required. The two types of features are then jointly segmented according to the same time window. Each segment corresponds to a time period, and the average alternation amplitude and rhythm frequency values ​​within that time period are labeled. Subsequently, within each time period, the change path of the alternation direction of power requests is further extracted, and the behavior of whether the direction switching exhibits periodic rotation without deviating from the initial direction is statistically analyzed and quantified as a direction consistency index. For example, if the power request direction cycles between rising and falling within a certain time period, and the change pattern remains basically consistent, it can be considered as having high direction consistency. The direction consistency indices from different time periods are collected and integrated to form direction consistency features, which are finally combined with the alternation amplitude and rhythm characteristics to construct a complete power request behavior feature. This process achieves a multi-dimensional dynamic characterization of the power request change pattern, accurately reflecting whether the vehicle's power request possesses the inherent attributes of coordinated and regular control.

[0069] Joint segmented processing is a method that aligns different types of features along the time dimension to form a unified analysis unit. Its purpose is to ensure that data from different dimensions participate in the judgment collaboratively within the same time frame, which is beneficial for capturing potential correlations in changes. The consistency of alternating power request directions refers to whether the direction of power request changes in different time segments exhibits regular repetition, maintains a specific change pattern, or continuously fluctuates around a central value. This characteristic is used to identify whether the power request's adjustment behavior is predictable. Directional consistency features are the overall representation of the consistency results of directional changes extracted from multiple time windows, used to measure the overall stability of the power request at the directional level. Power request behavior features, used to characterize the changing behavior of power requests, are a composite set of behavioral features formed by integrating alternating amplitude features, change rhythm features, and directional consistency features. This set has the ability to comprehensively evaluate power request changes from three levels: amplitude, rhythm, and direction, and is a key criterion for subsequently determining whether the power request can be coordinated.

[0070] S3. Based on the power request behavior characteristics, perform feature fusion operation to generate a coordination adaptation factor that reflects the inherent consistency of power request changes. This factor is used to characterize the endogenous coordination attributes of each power request when the vehicle alternately issues multiple power requests in the state of charging piles in parallel charging.

[0071] In this embodiment, S3 specifically includes the following steps:

[0072] S301. Normalize the alternating amplitude features, change rhythm features, and direction consistency features, convert them into vector components under a unified scale, and construct a power request behavior feature vector for feature fusion operation.

[0073] To construct a power request behavior feature vector for subsequent weighted calculations, the alternating amplitude feature, the rhythm of change feature, and the directional consistency feature need to be normalized to ensure comparability and calculability at the same scale. Specifically, linear normalization or Z-score normalization can be used to map the values ​​in different feature dimensions to a specified scale range, for example, mapping the values ​​to between 0 and 1. In actual processing, the mean of alternating amplitude, the periodic frequency of rhythm of change, and the ratio of directional consistency are first collected over a fixed time period. Then, their maximum and minimum values ​​are calculated as the upper and lower limits of the normalization interval, and the original feature values ​​are linearly compressed. After normalization, the three types of feature values ​​are arranged sequentially to form a three-dimensional vector, serving as the basic expression of power request behavior. This is done to avoid computational bias caused by different feature units and numerical distribution ranges, ensuring that the weighted results in subsequent fusion processing have a unified expression space.

[0074] Normalization refers to the mathematical transformation of numerical data to a unified numerical scale. Its purpose is to eliminate the dimensional differences between features caused by varying measurement standards, ensuring that each feature has equal weight in computation. Vector components under a unified scale mean that, after normalization, each power request behavior feature is represented as vector elements with a consistent numerical range, facilitating vector-level algebraic processing. The power request behavior feature vector is an ordered vector composed of multiple normalized values ​​reflecting the dynamic changes in power requests. It characterizes the behavior of vehicles alternately issuing multiple power requests while charging at charging stations. This vector not only carries information about the correlation between the amplitude, rhythm, and direction of power changes but also provides a unified computational structure for subsequent feature fusion operations. By constructing this feature vector, the inherent coordination of power requests can be evaluated and modeled at the overall behavioral pattern level without relying on single data features.

[0075] S302. Using the power request behavior feature vector as input, each feature component is fused according to the weighted calculation rules to extract stability index and correlation index, which are used to generate a coordination adaptation factor that reflects the inherent consistency of power request changes.

[0076] S303. The coordination adaptation factor is associated and matched with the power request behavior feature vector to identify the inherent pattern in the power request behavior, and the coordination adaptation factor is used to characterize the endogenous coordination attribute of each power request when the vehicle alternately issues multiple power requests in the state of charging piles.

[0077] When charging at multiple charging stations simultaneously, some vehicles, due to their unique internal energy management strategies, may alternately issue multiple power requests with similar but slightly different values. This behavior can easily be misjudged as resource conflicts during power scheduling. To avoid misjudgment, it is necessary to accurately identify whether these alternating power requests possess inherent coordination, thereby determining whether they constitute acceptable internal adjustment behavior. By using the power request behavior feature vector as input and fusing each feature component according to weighted calculation rules, stability and correlation indicators can be effectively extracted, comprehensively reflecting the consistency and regularity of power requests in both time and amplitude dimensions. This consistency is quantified through a coordination adaptation factor, which is then further matched and analyzed with the original feature vector to identify whether stable, repetitive behavioral patterns exist. This operation not only constructs a quantifiable judgment mechanism for complex power request behavior but also provides a reliable basis for subsequent power allocation response strategies, making scheduling more flexible and precise, minimizing erroneous intervention in normal dynamic adjustment behavior, and thus improving the stability and charging efficiency of the parallel charging process.

[0078] In this embodiment, S302 specifically refers to:

[0079] Using the power request behavior feature vector as input, the alternating amplitude feature, the change rhythm feature and the direction consistency feature are respectively processed by interval discretization to obtain the value distribution sequence of each feature component in multiple time segments, which is used to characterize the degree of concentration of the feature component changes.

[0080] To characterize the concentration of power request behavior features over time, it is necessary to perform interval discretization on the alternating amplitude feature, the rhythm of change feature, and the directional consistency feature, using the power request behavior feature vector as input. Interval discretization involves dividing the continuous numerical feature into several discrete segments of fixed width and counting the frequency of feature values ​​falling into each segment within each time segment, forming a value distribution sequence reflecting the distribution trend of that feature component. Specifically, the time series of the feature vector is first divided into segments using a sliding time window of fixed length, for example, every 5 seconds. The alternating amplitude feature values ​​in each segment are then divided and counted according to the interval boundaries, recording their distribution density in different intervals. The rhythm of change and directional consistency features are similarly segmented and divided into intervals to obtain their interval distribution within each time segment. This method effectively captures the concentrated areas and dispersion of each feature component, laying the foundation for subsequent calculations of distribution dispersion parameters and time fluctuation parameters. Interval discretization helps remove interference from occasional fluctuations, strengthens the expressive power of regular behavior patterns in the feature sequence, and thus improves the accuracy of recognizing the inherent consistency of power request behavior. The value distribution sequence is a numerical sequence formed on the basis of discrete statistical results, used to express the fluctuation structure and central tendency of a certain feature in the process of time evolution.

[0081] Based on the value distribution sequence, the discrete distribution parameter and time fluctuation parameter are calculated for each feature component. According to the pre-set weighting calculation rules, the discrete distribution parameter and time fluctuation parameter are mapped to the corresponding weight coefficients, and a weighted fusion operation is performed to generate a stability index.

[0082] To measure the intensity of fluctuations and the degree of concentration in the distribution of power request behavior characteristics over time, two key parameters need to be calculated for each feature component based on the value distribution sequence: the distribution dispersion parameter and the temporal fluctuation parameter. The distribution dispersion parameter measures the concentration of feature values ​​in the interval dispersion results, typically achieved by calculating the interval variance or the proportion of the maximum frequency band within each time segment. The temporal fluctuation parameter reflects the magnitude of change of the feature between adjacent time segments, for example, by calculating the trend of the difference in mean and standard deviation between adjacent time segments. After these two parameters are calculated, they are mapped to a unified metric, and the weight of each parameter in the fusion operation is determined according to a pre-defined weighted calculation rule. The weight coefficients are set based on empirical rules or model optimization results; for example, if the distribution dispersion is more critical in judging stability, it can be given a higher weight. Subsequently, a weighted fusion operation is used to synthesize these two parameters into a unified stability index, used to characterize whether the numerical distribution of the feature component is stable and whether the changes are controllable throughout the entire time period. The stability index is an important basis for judging whether power request behavior has coordination potential; the closer the value is to the stable range, the more predictable and manageable the behavior pattern is. Such fusion methods help quantify the regularity in characteristic changes and improve the ability to intelligently judge the coordination of power requests.

[0083] Based on the generation of stability indices, a pairwise analysis is performed on the changing trends of each feature component in the power request behavior feature vector to calculate the degree of trend consistency, generate correlation indices, and combine the stability indices with the correlation indices to form a coordination adaptation factor, which is used to characterize the inherent consistency of power request changes.

[0084] To further identify the inherent consistency of power request behavior, it is necessary to perform paired analysis on each feature component in the power request behavior feature vector based on the generated stability index. This analysis identifies the changing trends of these components in the time series and calculates the degree of consistency among these trends. Specifically, firstly, time-varying curves are established for alternating amplitude features, changing rhythm features, and directional consistency features. The trend direction of each feature within different time windows is extracted, such as rising, falling, or remaining stable. Then, these trend sequences are combined pairwise to calculate the consistency rate of their changing directions, the degree of overlap of time nodes, and the degree of synchronization of trend changes, thereby quantifying the degree of trend consistency. Based on the results of these paired analyses, correlation indicators reflecting the mutual synergy between features are extracted to determine whether each feature exhibits a shared driving or jointly regulated behavioral logic. Finally, the previously generated stability index and the currently obtained correlation index are combined and processed to generate a coordination adaptation factor through linear weighting or nonlinear aggregation. This factor characterizes the consistency of power request changes in the time dimension and behavioral patterns. The more concentrated the coordination adaptation factor values ​​are in the high-consistency range, the greater the potential for coordinated processing of the power request behavior, which helps to achieve more reasonable power resource matching in concurrent charging scheduling.

[0085] In this embodiment, S303 specifically refers to:

[0086] The range of values ​​for the coordination adaptation factor is divided into multiple fixed-level segments. Each feature dimension in the power request behavior feature vector is segmented over time. The average value and fluctuation value of the feature values ​​in each segment are calculated, and this set of data is used as the feature expression vector.

[0087] To accurately determine power request behavior, the range of values ​​for the coordination adaptation factor needs to be divided into multiple fixed-level segments, and a matching feature representation system needs to be constructed. Specifically, each feature dimension in the power request behavior feature vector is first divided into several equal-length time segments according to the time series, thus completing time-series segmentation, where each segment represents the behavior state within a certain time period. Then, the feature values ​​within each segment are statistically analyzed, and the arithmetic mean and standard deviation of all feature values ​​within that segment are calculated. The former characterizes the overall performance level of the feature within that segment, while the latter reflects the degree of fluctuation of the feature within that segment. Through these calculations, an overview of the performance of each segment on each feature dimension is obtained. Finally, the average and fluctuation values ​​of multiple feature dimensions over the same time period are combined into a feature representation vector. This vector can be used to subsequently match the coordination adaptation factor for different level segments, identifying the coordination potential of vehicle power request behavior. In this way, behavioral analysis can be refined along the time dimension, forming a behavioral profile that can be quantitatively compared, improving the accuracy of identifying the coordination attributes of power request behavior.

[0088] The numerical distance between each feature expression vector and the center value of each level segment is calculated to determine the distance relationship between it and each level segment. The level segment number that meets the minimum distance requirement is extracted as the candidate level index, and the corresponding coordination matching label is constructed.

[0089] To determine the level of power request behavior features, a matching relationship can be established by calculating the numerical distance between each feature expression vector and the center values ​​of multiple coordination and adaptation factor level segments. First, a center value needs to be preset for each level segment, obtained statistically from the typical power request behavior feature vectors corresponding to that level. Then, the target feature expression vector is compared with each center value using Euclidean distance or Manhattan distance methods to obtain a list of distances for all segments. Based on this, the level segment number corresponding to the center value with the shortest distance to the feature expression vector is selected as a candidate level index. It is then determined whether this minimum distance is less than a set threshold. If the condition is met, the level segment number is assigned as the matching result of the current feature expression vector, and a unique coordination matching label is generated to identify the level category. This label will serve as the core basis for subsequent behavior pattern recognition and coordination determination. This process enables the accurate mapping of complex behavioral features to different coordination level intervals, helping to improve the classification clarity and processing accuracy of subsequent power request identification.

[0090] A label fitting operation is performed on the behavioral feature trajectory of each power request node. Power request nodes with fitting errors less than a preset error limit are selected. Based on label consistency, the corresponding coordination adaptation factor is assigned to the node. The coordination adaptation factor is used to characterize the endogenous coordination attribute of each power request when the vehicle alternately issues multiple power requests in the state of parallel charging at the charging pile.

[0091] To achieve accurate identification of the coordination of power request nodes, a label fitting operation can be performed on the behavioral feature trajectory of each power request node. The label fitting operation involves comparing the feature representation vector of each node with previously constructed coordination matching labels one by one, and using error analysis to determine the degree of fit between the node and each level of labels. In the implementation process, firstly, based on the changes in the node's power request behavior over time, a feature trajectory vector corresponding to that node is constructed. Then, the error value between the vector and the features represented by different coordination matching labels is calculated using numerical differences, and then compared with a preset error limit. When the fitting error corresponding to a label is less than the preset error limit, the label is considered to accurately characterize the node's coordination behavior pattern. Furthermore, by statistically analyzing the label matching results of the node over multiple time periods, its label consistency level is evaluated. If the label remains stable, a matching coordination adaptation factor is assigned to the node, and this factor is used as the core parameter characterizing its endogenous coordination attributes. This process can significantly improve the analytical accuracy of complex power request behaviors, avoid misjudgments due to short-term disturbances, and ensure the identification of the actual power control strategy characteristics adopted by the vehicle in parallel charging mode.

[0092] S4. Based on the correspondence between the coordination adaptation factor and the preset confidence interval, determine the coordination attribute of each power request and generate the corresponding coordination confidence level mark to distinguish the coordination attribute category of the power request in the charging pile parallel charging state.

[0093] In this embodiment, S4 specifically refers to:

[0094] Based on the coordination adaptation factor data of multiple historical power requests, multiple preset confidence intervals are established, and a unique coordination attribute category number is assigned to each preset confidence interval to construct the correspondence between coordination adaptation factors and confidence intervals.

[0095] To enhance the intelligent recognition capability of power request behavior, a large amount of vehicle power request data collected from historical periods under concurrent charging conditions can be used to extract the coordination adaptability factor value corresponding to each power request, and statistical analysis can be performed on all adaptability factor data. Through cluster analysis, density partitioning, or equal-interval segmentation, the overall distribution of coordination adaptability factor values ​​is divided into multiple continuous, non-overlapping confidence intervals. Each confidence interval represents a stable coordination level, which can represent different intrinsic coordination intensities of vehicle power request behavior. To ensure consistency in subsequent judgments, each confidence interval needs to be assigned a fixed coordination attribute category number; for example, number 1 represents a high-consistency request, number 2 represents a medium-consistency request, and number 3 represents a request with no obvious coordination attribute. Finally, the numerical ranges of coordination adaptability factors are mapped one-to-one with these numbers to construct a complete correspondence between coordination adaptability factors and confidence intervals. The purpose of this setup is to ensure that each subsequent power request, when entering the judgment logic, can be automatically mapped to a predefined coordination attribute category through its adaptability factor value, improving the judgment efficiency and consistency of data processing.

[0096] The coordination adaptation factor data for multiple historical power requests refers to a dataset extracted from a large number of past concurrent charging instances, used to characterize the inherent coordination attributes of each power request. This dataset is representative and statistically grounded. Preset confidence intervals are continuous numerical ranges, either manually set or automatically generated by algorithms, based on a thorough analysis of the distribution characteristics of the adaptation factor data. Each interval reflects a specific range of coordination degree. A unique coordination attribute category number is a manually defined, non-repeating numerical label for each confidence interval, used to indicate a specific coordination attribute category in subsequent classification and identification. The correspondence between coordination adaptation factors and confidence intervals refers to mapping each specific adaptation factor value to its corresponding confidence interval through a judgment of its endpoint, and further linking it to its associated attribute number, forming a complete three-element binding structure of value-interval-category, used to support the classification and identification of coordination attributes of power requests. This structure ensures the logical continuity and scalability of the judgment process and is the core foundation of the coordination attribute judgment mechanism.

[0097] For each power request in the parallel charging state of the charging pile, the corresponding coordination adaptation factor value is extracted, and the value is compared with multiple preset confidence intervals. Based on the preset confidence interval range in which it falls, the coordination attribute judgment is performed to determine the coordination attribute category to which the power request belongs.

[0098] In parallel charging scenarios, to accurately classify each power request, it's necessary to extract the corresponding coordination adaptation factor value for each power request node. This factor, derived from prior feature fusion and association matching, comprehensively reflects the inherent consistency of power request changes. After extraction, the coordination adaptation factor value is compared one by one with multiple pre-established confidence intervals. By determining the position of the value on the number axis, its confidence interval range is determined. For example, when the coordination adaptation factor value of a power request is in the middle interval, it can be directly mapped to the corresponding medium coordination attribute category. This interval comparison method transforms continuously changing numerical factors into discrete and stable category determination results, avoiding unstable judgments based directly on instantaneous numerical fluctuations. This processing logic facilitates rapid, unified, and reproducible classification of power requests in parallel charging scenarios, providing clear attribute basis for subsequent scheduling and management.

[0099] Coordination attribute determination refers to the process of classifying and confirming the inherent behavioral attributes of power requests based on the correspondence between coordination fitness factors and confidence intervals. This process does not rely on a single instantaneous power value, but rather on a stable judgment result formed by comprehensive behavioral characteristics. Coordination attribute categories are abstract classifications of the behavioral characteristics exhibited by power requests in a concurrent charging state. Each category corresponds to a specific coordination level, used to distinguish whether a power request possesses the potential conditions for coordinated processing. By mapping coordination fitness factors to confidence intervals, complex behavioral consistency information can be transformed into clear category identifiers, enabling different power requests to have distinguishable and comparable attribute labels in a concurrent charging environment. This classification method provides a structured foundation for concurrent charging data processing, making it possible to subsequently adopt differentiated processing for different coordination attribute categories, while ensuring the consistency and stability of the determination process.

[0100] Based on the coordination attribute category to which the power request belongs, a coordination trust level tag matching the power request is generated as a unique code to identify the coordination attribute category of the power request, thus completing the differentiation of the coordination attribute category of the power request in the charging pile's parallel charging state.

[0101] After determining the coordination attribute category of a power request, to achieve standardized data management and subsequent control logic, each power request needs to be assigned a unique identifier corresponding to its coordination attribute category. This can be achieved by establishing a one-to-one mapping relationship between coordination attribute categories and coding rules, assigning a unique coded value to each coordination attribute category, and generating a coordination trust level label for identification. For example, when a power request is classified into the third coordination attribute category, it can be assigned the label "C3" according to the mapping rules, indicating that the request has a medium level of endogenous coordination potential under the current behavioral characteristics. This level label not only possesses structured attributes but can also be used in multiple processing stages such as data filtering and power scheduling weight control, ensuring that subsequent processing operations for requests can respond differently based on their behavioral stability. This coding method has high scalability and system compatibility in multi-vehicle parallel charging environments, helping charging platforms build a unified request labeling system.

[0102] The coordination trust level label is a unique encoded result used to identify the coordination attribute category of a power request. It is typically expressed using a combination of letters and numbers, ensuring that different categories of requests have clearly identifiable labels at the data level. This label is generated based on the judgment result of the coordination attribute category, and its generation logic depends on the trust interval mapped by the coordination adaptability factor, combined with specific level classification rules. Each level label represents a coordination attribute level segment, reflecting the strength of the coordination characteristics exhibited by the power request in continuous behavior. By assigning coordination trust level labels to power requests, differentiated management of the attributes of different power requests can be achieved in the context of concurrent charging at charging stations. This provides the system with a quantifiable, traceable, and dynamically processable request hierarchy foundation, thereby supporting the orderly execution of subsequent power response strategies and resource allocation mechanisms.

[0103] S5. Based on the coordination trust level label and the power resource occupancy status of the charging pile in the parallel charging state, dynamically adjust the response order and response rhythm of power requests so that power requests participate in power allocation according to the coordination attribute category during the parallel charging process.

[0104] In this embodiment, S5 specifically refers to:

[0105] The system collects real-time power resource occupancy data of charging piles in parallel charging mode, divides the current total power resources into multiple coordination attribute categories for allocation channels, and establishes a candidate response sequence based on the coordination trust level label corresponding to each power request.

[0106] Real-time power resource occupancy during concurrent charging can be obtained by collecting instantaneous output power data from each concurrent charging gun, the currently adjustable redundant power capacity, and the load status of the main power supply circuit. To accurately allocate resources for power requests of different coordination attribute categories, the current total power resources can be divided into multiple physical or logical channels proportionally or according to rules, with each channel corresponding to a coordination attribute category. For example, if the total power is 300 kW, three channels can be allocated according to the system strategy, with 150 kW, 100 kW, and 50 kW allocated respectively to respond to power requests with high, medium, and low coordination trust labels. Then, all currently pending power requests are classified according to their coordination trust label, and a candidate response sequence is constructed based on time order, request timestamp, or power demand value, thus laying the data foundation for dynamic adjustment of the subsequent response order and rhythm. This approach can prioritize requests with high stability and strong inherent coordination under resource constraints, thereby improving concurrent charging scheduling efficiency.

[0107] Real-time power resource occupancy in concurrent charging refers to the ratio of the total allocatable power to the current power usage of each connection channel at any given moment when multiple vehicles are simultaneously connected to a charging station. Coordination attribute category allocation channels are power allocation channels divided according to different coordination attribute categories. Each channel represents a specific coordination trust level marker and the schedulable resource range of the request, with clear capacity limits and scheduling priorities. The candidate response sequence is a sequence of responses arranged according to a certain sorting rule, constructed based on all pending power requests and considering their coordination trust level markers. This sequence is used to subsequently determine the scheduling order and control strategy. By constructing such a structured response queue, orderly allocation of power resources can be achieved in a dynamic environment.

[0108] The response order of the candidate response sequences is adjusted according to the coordination trust level label. The matching response rhythm control parameters are calculated by combining the allocated channel capacity and power resource occupancy ratio corresponding to each coordination attribute category, and a control configuration containing order and rhythm is generated.

[0109] To rationally allocate power resources during parallel charging, the candidate response sequences must first be sorted according to the coordination trust level tag carried by each power request. The priority of the sorting can be arranged from high to low according to a predefined trust level; for example, level 1 is superior to level 2, and level 2 is superior to level 3. After sorting, the scheduling frequency for each category needs to be restricted or relaxed based on the allocation channel capacity and current power resource occupancy ratio corresponding to each coordination attribute category. For example, if the allocation channel capacity for category 1 is 150 kW and the current occupancy rate is 80%, its response pace should be appropriately slowed down; conversely, if the channel capacity for category 3 is 50 kW and the current occupancy rate is only 30%, its response pace can be appropriately increased. To this end, the response cycle, delay start time, and interval between adjacent requests for each category of power request can be calculated to form the corresponding response pace control parameters. Finally, the response order of power requests and the response pace parameters for each category are combined to form a control configuration that includes request sorting and time control elements, used for subsequent execution of scheduling instructions. This approach not only reflects priority based on resource allocation but also dynamically adapts to real-time load changes, thereby achieving a balance between resource utilization and scheduling fairness.

[0110] Channel capacity allocation refers to the upper limit of power resource allocation defined for different coordination attribute categories, ensuring that different levels of power requests can obtain corresponding power channel support during parallel charging. Power resource occupancy ratio is the ratio between the currently used power of a channel and its total capacity, reflecting the current resource scarcity. Matching response rhythm control parameters are a set of parameters used to control the scheduling frequency, typically including response period, initial response delay, and minimum interval, used to dynamically adjust the speed and frequency of power request responses. Sequence and rhythm-based control configuration refers to the scheduling execution plan formed by integrating the scheduling order of sorted requests with its response rhythm control parameters, used to guide the charging pile control system to respond to different power requests sequentially according to preset rules, achieving a precise and dynamic power allocation strategy.

[0111] Based on the dynamic control of the regulation configuration, the response order and response rhythm of each power request during the parallel charging process are controlled, so that power requests with different coordination trust level labels participate in power resource allocation according to the coordination attribute category.

[0112] To ensure effective allocation of power resources in parallel charging mode, dynamic control of the response order and rhythm of each power request is required based on the control configuration. This control process first categorizes each power request into its corresponding coordination attribute category channel based on its coordination trust level label, and then arranges the requests according to the response order defined in the control configuration. For example, level 1 requests are responded to first, followed by level 2 and level 3 requests in sequence. Based on this, the system adjusts the scheduling sequence according to the response rhythm control parameters allocated to each channel, such as the response period and minimum time interval. Specifically, the system can set the response period for level 1 requests to 60 milliseconds, level 2 to 120 milliseconds, and level 3 to 200 milliseconds, thus varying the response density of different request categories. In this way, even under resource constraints, high-level requests can be prioritized for resource acquisition, while low-level requests are gradually inserted after resources are released, thereby implementing a strategy of responding according to coordination attribute categories, improving overall scheduling efficiency and power allocation fairness.

[0113] The control configuration refers to the scheduling execution scheme generated by combining sorting and time parameters. It includes sequential configurations to distinguish response priorities and rhythm parameters to adjust response frequency. Dynamic control refers to the system reading the control configuration in real time and making rapid scheduling adjustments based on the current power request status and resource allocation, allowing the response time and order of requests to be updated at any time. Response order indicates the order in which power requests enter the execution queue according to their priority, while response rhythm indicates the time frequency and interval control strategy for each request during the actual response process. The coordination trust level label is a level code carried by the power request behavior evaluation result, used to identify the coordination level of the request's behavior. The coordination attribute category is a classification unit based on the level label, used to determine which power allocation channel each request should enter for scheduling. Through the organic combination of these technical elements, hierarchical response and dynamic control of power requests during the parallel charging process are achieved.

[0114] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, as a computer program product. A computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, all or part of the processes or functions according to the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. Computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired or wireless means (e.g., infrared, wireless, microwave, etc.). A computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more sets of available media. Available media can be magnetic media (e.g., floppy disks, hard disks, magnetic tapes), optical media (e.g., DVDs), or semiconductor media. Semiconductor media can be solid-state drives.

[0115] It should be understood that in the various embodiments of this application, the order of the above-mentioned processes does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0116] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0117] In the several embodiments provided in this application, it should be understood that the disclosed systems and methods can be implemented in other ways. For example, the embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection through some interfaces, devices, or units, and may be electrical, mechanical, or other forms.

[0118] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0119] In addition, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0120] The above are merely specific embodiments of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A data processing method for concurrent charging of charging piles, characterized in that, Specifically, the following steps are included: S1. Continuously collect power request data of vehicles in parallel charging state at charging piles, and construct a power request alternation detection sequence according to the time sequence. By analyzing the back-and-forth switching relationship of power requests in the detection sequence, determine whether the vehicle alternately issues multiple power requests in parallel charging state at charging piles. S2. In the case of multiple power requests being issued alternately, an evolutionary analysis is performed on the alternating detection sequence of power requests to extract the alternating amplitude features, change rhythm features, and directional consistency features that reflect the changing pattern of power requests, forming power request behavior features used to characterize the changing behavior of power requests. S3. Based on the power request behavior characteristics, perform feature fusion operation to generate a coordination adaptation factor that reflects the inherent consistency of power request changes. This factor is used to characterize the endogenous coordination attributes of each power request when the vehicle alternately issues multiple power requests in the state of charging piles in parallel charging. S3 specifically includes the following steps: S301. Normalize the alternating amplitude features, change rhythm features, and direction consistency features, convert them into vector components under a unified scale, and construct a power request behavior feature vector for feature fusion operation. S302. Using the power request behavior feature vector as input, each feature component is fused according to the weighted calculation rules to extract stability index and correlation index, which are used to generate a coordination adaptation factor that reflects the inherent consistency of power request changes. S302 specifically refers to: Using the power request behavior feature vector as input, the alternating amplitude feature, the change rhythm feature and the direction consistency feature are respectively processed by interval discretization to obtain the value distribution sequence of each feature component in multiple time segments, which is used to characterize the degree of concentration of the feature component changes. Based on the value distribution sequence, the discrete distribution parameter and time fluctuation parameter are calculated for each feature component. According to the pre-set weighting calculation rules, the discrete distribution parameter and time fluctuation parameter are mapped to the corresponding weight coefficients, and a weighted fusion operation is performed to generate a stability index. Based on the generation of stability indices, a pairwise analysis is performed on the changing trends of each feature component in the power request behavior feature vector to calculate the degree of trend consistency, generate correlation indices, and combine the stability indices with the correlation indices to form a coordination adaptation factor, which is used to characterize the inherent consistency of power request changes. S303. The coordination adaptation factor is associated and matched with the power request behavior feature vector to identify the inherent pattern in the power request behavior, and the coordination adaptation factor is used to characterize the endogenous coordination attribute of each power request in the case of the vehicle alternately issuing multiple power requests in the charging pile charging state. S4. Based on the correspondence between the coordination adaptation factor and the preset confidence interval, determine the coordination attribute of each power request and generate the corresponding coordination confidence level mark to distinguish the coordination attribute category of the power request in the charging pile parallel charging state. S5. Based on the coordination trust level label and the power resource occupancy status of the charging pile in the parallel charging state, dynamically adjust the response order and response rhythm of power requests so that power requests participate in power allocation according to the coordination attribute category during the parallel charging process.

2. The data processing method for parallel charging of charging piles according to claim 1, characterized in that, S1 specifically refers to: The system continuously collects power request data issued by vehicles in parallel charging state during the charging initialization phase, records each power request data in the actual order of its generation, and forms a power request data stream arranged in chronological order. Based on the power request data stream, adjacent power request data are compared one by one to identify the direction and magnitude of change in the numerical value of the power request data. Request nodes with numerical rotation characteristics are extracted from the power request data stream, and a power request alternation detection sequence is constructed in chronological order. The alternating power request detection sequence is traversed and analyzed. When the change amplitude between two or more power request data in the detection sequence is lower than the preset difference threshold and the change direction is periodically alternating, it is determined that the vehicle is alternately issuing multiple power requests in the charging pile's parallel charging state.

3. The data processing method for parallel charging of charging piles according to claim 1, characterized in that, S2 specifically refers to: When it is confirmed that the vehicle is issuing multiple power requests alternately, the alternating power request detection sequence is continuously scanned, the difference between adjacent power request data is extracted, and the mean and standard deviation of the difference within a preset fixed time window are calculated to form the alternation amplitude feature. Based on the acquisition time points of adjacent power requests in the alternating power request detection sequence, the average interval period of the alternation direction of power requests is calculated, and the number of occurrences of the alternation direction within the same period is counted to construct the change rhythm characteristics. The alternation amplitude feature and the change rhythm feature are jointly segmented and processed. The consistency of the alternation direction of power request is statistically analyzed to generate the direction consistency feature. The three features are then combined into a power request behavior feature to characterize the power request change behavior.

4. The data processing method for parallel charging of charging piles according to claim 1, characterized in that, S303 specifically refers to: The range of values ​​for the coordination adaptation factor is divided into multiple fixed-level segments. Each feature dimension in the power request behavior feature vector is segmented over time. The average value and fluctuation value of the feature values ​​in each segment are calculated as feature expression vectors. The numerical distance between each feature expression vector and the center value of each level segment is calculated to determine the distance relationship between it and each level segment. The level segment number that meets the minimum distance requirement is extracted as the candidate level index, and the corresponding coordination matching label is constructed. A label fitting operation is performed on the behavioral feature trajectory of each power request node. Power request nodes with fitting errors less than a preset error limit are selected. Based on label consistency, the corresponding coordination adaptation factor is assigned to the node. The coordination adaptation factor is used to characterize the endogenous coordination attribute of each power request when the vehicle alternately issues multiple power requests in the state of parallel charging at the charging pile.

5. The data processing method for parallel charging of charging piles according to claim 1, characterized in that, S4 specifically refers to: Based on the coordination adaptation factor data of multiple historical power requests, multiple preset confidence intervals are established, and a unique coordination attribute category number is assigned to each preset confidence interval to construct the correspondence between coordination adaptation factors and confidence intervals. For each power request in the parallel charging state of the charging pile, the corresponding coordination adaptation factor value is extracted, and the value is compared with multiple preset confidence intervals. Based on the preset confidence interval range in which it falls, the coordination attribute judgment is performed to determine the coordination attribute category to which the power request belongs. Based on the coordination attribute category to which the power request belongs, a coordination trust level tag matching the power request is generated as a unique code to identify the coordination attribute category of the power request, thus completing the differentiation of the coordination attribute category of the power request in the charging pile's parallel charging state.

6. The data processing method for parallel charging of charging piles according to claim 1, characterized in that, S5 specifically refers to: The system collects real-time power resource occupancy data of charging piles in parallel charging mode, divides the current total power resources into multiple coordination attribute categories for allocation channels, and establishes a candidate response sequence based on the coordination trust level label corresponding to each power request. The response order of the candidate response sequences is adjusted according to the coordination trust level label. The matching response rhythm control parameters are calculated by combining the allocated channel capacity and power resource occupancy ratio corresponding to each coordination attribute category, and a control configuration containing order and rhythm is generated. Based on the dynamic control of the regulation configuration, the response order and response rhythm of each power request during the parallel charging process are controlled, so that power requests with different coordination trust level labels participate in power resource allocation according to the coordination attribute category.

Images