A new energy charging pile power distribution control method based on an intelligent algorithm

Abstract

The application discloses a new energy charging pile power distribution control method based on an intelligent algorithm, relates to the technical field of automobile charging, and comprises the following steps: applying a multi-frequency sinusoidal disturbance current to a battery of an electric vehicle to obtain a voltage response signal; performing Fourier transform decoupling on the voltage response signal, and marking a risk identification code according to the phase shift angle size in the battery electrochemical impedance parameter; extracting an energy integral value from a high-frequency sound wave signal; fusing the risk identification code and the energy integral value to determine the battery risk level of the electric vehicle battery; performing layered power control on the battery risk level and a preset risk power adjustment table to output a charging pile power parameter; and adjusting new energy power according to the charging pile power parameter by using a charging pile power regulator to verify the new energy charging pile power distribution effect. The application realizes multi-modal comprehensive evaluation of charging risks and accurate dynamic regulation and control of new energy charging pile power distribution through fusion determination of the risk identification code and the energy integral value.

Description

Technical Field

[0001] This invention relates to the field of vehicle charging technology, and in particular to a power distribution control method for new energy charging piles based on intelligent algorithms. Background Technology

[0002] In recent years, significant progress has been made in power allocation control technology for new energy charging piles, particularly in grid collaborative optimization and multimodal sensing fusion. Traditional power allocation methods often rely on matrix topologies or fixed power module combinations, driving dynamic power adjustments through preset load thresholds and historical charging data. Flexible scheduling strategies further introduce shared power pools and AI algorithms to achieve dynamic resource allocation. In the field of battery state monitoring, electrochemical impedance spectroscopy (EIS) analyzes the dynamic characteristics of the battery interface through multi-frequency perturbation signals, while ultrasonic sensor arrays reflect changes in internal mechanical stress through acoustic energy integration, constructing battery health assessment systems from electrochemical and acoustic dimensions, respectively. Current research trends focus on multi-source data fusion modeling, aiming to overcome the limitations of single-feature representation and provide more accurate risk assessment basis for power allocation.

[0003] However, existing technologies have shortcomings in multimodal feature coupling analysis and dynamic risk-power mapping mechanisms. Traditional power allocation strategies are mostly based on single-dimensional indicators and fail to effectively integrate the joint characteristics of electrochemical impedance phase shift angle and acoustic energy integral. This results in limited dimensions for battery risk assessment, making it difficult to capture latent failure modes under complex operating conditions. Existing risk-power mapping models mostly use static thresholds or empirical formulas, failing to establish a nonlinear relationship between risk level and power adjustment, and lacking support for dynamic weight allocation and cluster analysis of multi-source data. Consequently, the adaptability of power allocation strategies to the real-time state of the battery is insufficient. Summary of the Invention

[0004] In view of the aforementioned existing problems, the present invention is proposed.

[0005] Therefore, this invention provides a power allocation control method for new energy charging piles based on intelligent algorithms to address the shortcomings in multimodal feature coupling analysis and dynamic risk-power mapping mechanisms.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution:

[0007] In a first aspect, the present invention provides a power allocation control method for new energy charging piles based on intelligent algorithms, comprising: applying a multi-frequency sinusoidal disturbance current to an electric vehicle battery to obtain a voltage response signal; performing Fourier transform decoupling on the voltage response signal, outputting the voltage frequency component and converting it into battery electrochemical impedance parameters, and marking a risk identification code according to the phase shift angle in the battery electrochemical impedance parameters; acquiring a high-frequency sound wave signal through an ultrasonic sensor array and extracting an energy integral value from the high-frequency sound wave signal; fusing the risk identification code and the energy integral value to determine and output the battery risk level of the electric vehicle battery; performing hierarchical power control based on the battery risk level and a preset risk power adjustment table, and outputting charging pile power parameters; and adjusting the new energy power according to the charging pile power parameters, while simultaneously monitoring the voltage change rate and temperature rise curve of the electric vehicle battery to verify the power allocation effect of the new energy charging pile.

[0008] As a preferred embodiment of the intelligent algorithm-based power distribution control method for new energy charging piles described in this invention, the specific steps of applying a multi-frequency sinusoidal disturbance current to the electric vehicle battery and obtaining the voltage response signal are as follows:

[0009] The multi-frequency sinusoidal disturbance current is input into the power amplifier for amplitude adjustment, and the output is the adjusted disturbance current;

[0010] The disturbance current is injected into the electric vehicle battery to monitor the dynamic response of the internal voltage of the electric vehicle battery in real time and obtain the voltage waveform.

[0011] The voltage waveform is sampled synchronously in time to generate segmented voltage data;

[0012] The segmented voltage data is interpolated to generate a voltage response signal.

[0013] As a preferred embodiment of the intelligent algorithm-based power distribution control method for new energy charging piles described in this invention, the specific steps of performing Fourier transform decoupling on the voltage response signal, outputting the voltage frequency component, and converting it into battery electrochemical impedance parameters are as follows:

[0014] The voltage response signal is divided into multiple overlapping time windows, and segmented signals are output.

[0015] Perform a Fast Fourier Transform on the segmented signal to output component amplitude and component phase information;

[0016] The component amplitude is compared with the disturbance current amplitude in the adjustment disturbance current at each frequency point to generate the impedance amplitude;

[0017] The phase information of the component is compared with the phase information of the disturbance current in the regulating disturbance current to generate a phase shift angle;

[0018] The impedance magnitude and phase shift angle are combined to output the battery electrochemical impedance parameters.

[0019] As a preferred embodiment of the intelligent algorithm-based power allocation control method for new energy charging piles described in this invention, the specific steps of marking risk identification codes according to the phase shift angle in the battery electrochemical impedance parameters are as follows:

[0020] The battery electrochemical impedance parameters are classified according to the phase shift angle, and the risk classification parameters are output.

[0021] Based on the risk classification parameters, the corresponding risk identifier is matched to generate a risk identifier code.

[0022] As a preferred embodiment of the intelligent algorithm-based power distribution control method for new energy charging piles described in this invention, the specific steps of acquiring high-frequency sound wave signals through an ultrasonic sensor array and extracting energy integral values ​​from the high-frequency sound wave signals are as follows:

[0023] High-frequency acoustic signals from new energy charging piles and electric vehicle batteries are acquired using an ultrasonic sensor array.

[0024] The high-frequency acoustic signal is smoothed and weighted point-by-point using the Hamming window function to generate a weighted coefficient sequence.

[0025] The weighted coefficient sequence is combined point by point with the amplitude of the high-frequency acoustic signal to output a weighted high-frequency acoustic signal.

[0026] The weighted high-frequency acoustic signal is subjected to amplitude squaring, and the results of amplitude squaring are integrated to form an instantaneous energy sequence;

[0027] Accumulate the instantaneous energy sequence over a fixed time window and obtain the energy integral value for the corresponding fixed time window.

[0028] As a preferred embodiment of the intelligent algorithm-based power allocation control method for new energy charging piles described in this invention, the specific steps for fusing the risk identification code and energy integral value to determine and output the battery risk level of the electric vehicle battery are as follows:

[0029] The risk identification codes are numerically mapped to generate an electrochemical risk score.

[0030] The electrochemical risk score and energy integral value are normalized respectively, and the standard risk score and standard energy integral are output.

[0031] Based on the statistical results of historical failures, dynamic weights are assigned to the standard risk score and standard energy integral, and then weighted and fused to generate a comprehensive risk index.

[0032] The K-means clustering algorithm is used to group the comprehensive risk index and assign corresponding level labels to the groups to output the battery risk level.

[0033] As a preferred embodiment of the intelligent algorithm-based power allocation control method for new energy charging piles described in this invention, the specific steps of performing hierarchical power control based on battery risk levels and a preset risk power adjustment table to output charging pile power parameters are as follows:

[0034] Extract correlation data between battery risk and actual power adjustment from the historical operation logs of new energy charging piles;

[0035] Based on the power adjustment magnitude in the associated data, a risk-power mapping is performed on the temperature and charging efficiency changes of the electric vehicle battery to generate a preset risk power adjustment table.

[0036] According to the battery risk level, the power adjustment value of each new energy charging pile is obtained by searching in the preset risk power adjustment table in layers.

[0037] The various power adjustment values ​​are integrated to output the power parameters of the charging pile.

[0038] As a preferred embodiment of the intelligent algorithm-based power allocation control method for new energy charging piles described in this invention, the charging pile power regulator adjusts the new energy power according to the charging pile power parameters, while simultaneously monitoring the voltage change rate and temperature rise curve of the electric vehicle battery to verify the power allocation effect of the new energy charging pile. The specific steps are as follows:

[0039] Obtain the real-time charging power of the electric vehicle from the charging pile interface, and adjust the real-time power to new energy power according to the charging pile power parameters;

[0040] The voltage value of the vehicle after the new energy power is adjusted is obtained in real time by a voltage sensor, and differential calculation is performed to generate the voltage change rate.

[0041] The surface temperature of the electric vehicle battery is collected in real time by a temperature sensor and plotted as a temperature rise curve over time.

[0042] Synchronize and align the voltage change rate with the temperature rise curve to generate temperature-voltage data pairs;

[0043] By comparing the power parameters and temperature and voltage data of charging piles on the same time axis, the dynamic relationship between power adjustment and battery state changes can be determined, and the power distribution effect of new energy charging piles can be verified.

[0044] In a second aspect, the present invention provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the computer program, when executed by the processor, implements any step of the power distribution control method for new energy charging piles based on intelligent algorithms as described in the first aspect of the present invention.

[0045] Thirdly, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, it implements any step of the power allocation control method for new energy charging piles based on intelligent algorithms as described in the first aspect of the present invention.

[0046] The beneficial effects of this invention are as follows: By fusing risk identification codes and energy integral values, a multimodal comprehensive assessment of charging risks and precise dynamic control of power allocation for new energy charging piles are achieved. High-frequency acoustic signals are acquired using an ultrasonic sensor array, from which energy integral values ​​are extracted. These values ​​are then combined with weighted processing and energy accumulation quantification of physical dimension signals, and normalized and dynamically weighted with risk identification codes to output a unified battery risk level. This achieves complementary enhancement of electrochemical and acoustic data, reduces single-factor misjudgments, and forms a comprehensive risk diagnosis basis. The verification mechanism monitors the voltage change rate and aligns the power adjustment effect with the temperature rise curve, optimizing the power strategy in a closed loop to adapt to actual charging conditions, ultimately ensuring the safe and reliable operation of electric vehicle batteries under dynamic conditions. Attached Figure Description

[0047] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0048] Figure 1 This is a flowchart of a power allocation control method for new energy charging piles based on intelligent algorithms.

[0049] Figure 2 A flowchart for voltage response signal processing;

[0050] Figure 3 This is a flowchart for high-frequency acoustic signal processing and fusion determination.

[0051] Figure 4 This is a flowchart for hierarchical power control and verification. Detailed Implementation

[0052] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0053] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0054] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.

[0055] Reference Figures 1-4 This is one embodiment of the present invention, which provides a power allocation control method for new energy charging piles based on intelligent algorithms, including the following steps:

[0056] S1. Apply a multi-frequency sinusoidal perturbation current to the electric vehicle battery and obtain the voltage response signal.

[0057] The multi-frequency sinusoidal disturbance current is input into the power amplifier for amplitude adjustment, and the output is the adjusted disturbance current;

[0058] Specifically, the multi-frequency sinusoidal perturbation current is generated by a signal generator, containing multiple sinusoidal wave components of different frequencies, covering the low-frequency to mid-frequency range, to excite the electrochemical response inside the battery. The multi-frequency sinusoidal perturbation current is first input to a power amplifier, which adjusts the amplitude of the multi-frequency sinusoidal perturbation current. The power amplifier adjusts the gain of each frequency component of the sinusoidal wave independently through an internal control circuit to ensure that the amplitude of the output signal is stable and distortion-free. During the adjustment process, the power amplifier monitors the waveform quality of the multi-frequency sinusoidal perturbation current in real time, eliminates noise interference, and ensures the purity of the output signal. After the adjustment is completed, the power amplifier outputs an adjusted perturbation current, which retains the frequency characteristics of the multi-frequency sinusoidal perturbation current.

[0059] The disturbance current is injected into the electric vehicle battery to monitor the dynamic response of the internal voltage of the electric vehicle battery in real time and obtain the voltage waveform.

[0060] After the regulating disturbance current is output from the power amplifier, it is injected into the electric vehicle battery through the current output interface of the new energy charging pile. The injection process is completed through a dedicated current transmission circuit to ensure that the regulating disturbance current is not attenuated or distorted during transmission. After the regulating disturbance current enters the electric vehicle battery, it triggers a dynamic response of the electrochemical substances inside the battery, causing fluctuations in the battery terminal voltage. In real time, a high-precision voltage sensor is connected to the positive and negative terminals of the electric vehicle battery to continuously collect data on changes in the battery terminal voltage. The voltage sensor operates at a high sampling rate to capture the dynamic changes in battery voltage under the action of the regulating disturbance current, forming continuous voltage time series data. During the monitoring process, the collected voltage time series data is filtered to remove environmental noise and external interference, and the output voltage waveform reflects the true response of the electric vehicle battery to the regulating disturbance current, providing reliable basic data for subsequent signal processing.

[0061] The voltage waveform is sampled synchronously in time to generate segmented voltage data;

[0062] Specifically, voltage waveforms, as continuous voltage time series data collected by real-time monitoring equipment, need to be time-synchronized sampled to ensure data consistency and accuracy. Time-synchronization sampling is controlled by a high-precision clock signal, and the sampling frequency is set according to the highest frequency component of the disturbance current to meet the requirements of the Nyquist sampling theorem (e.g., the sampling frequency must be greater than twice the highest frequency component in the sampled signal in order to completely preserve the spectral characteristics of the voltage waveform and achieve distortion-free signal reconstruction). The sampling process discretizes the voltage waveform into digital signals at fixed time intervals to form discrete voltage data points. The discrete voltage data points are divided according to fixed time windows to generate multiple segmented voltage data. Each segmented voltage data corresponds to a local time period of the voltage waveform, preserving the dynamic characteristics of the voltage waveform. During the segmentation process, adjacent time windows are kept overlapping to avoid the loss of signal characteristics at the segmentation boundary.

[0063] The segmented voltage data is interpolated to generate a voltage response signal;

[0064] Segmented voltage data consists of discrete data sequences sampled in time. Interpolation is required to generate a continuous voltage response signal, improving signal smoothness and analytical accuracy. The interpolation process employs a spline function interpolation algorithm, fitting discrete voltage data points within each segment to generate a smooth, continuous curve. During interpolation, the spline function identifies transition points within each time window based on the local characteristics of the segmented voltage data, ensuring the interpolation result matches the dynamic trend of the original voltage waveform. The interpolation algorithm considers the connection between adjacent segments, eliminating abrupt changes or discontinuities at segment boundaries to form a smooth overall voltage curve. After interpolation, the segmented voltage data forms a complete voltage response signal, retaining all frequency components and dynamic characteristics of the voltage waveform, providing high-quality input data for subsequent Fourier transform decoupling.

[0065] S2. Decouple the voltage response signal by performing a Fourier transform, output the voltage frequency component and convert it into battery electrochemical impedance parameters, and mark the risk identification code according to the phase shift angle in the battery electrochemical impedance parameters.

[0066] The voltage response signal is divided into multiple overlapping time windows, and segmented signals are output.

[0067] Specifically, the segmentation process first determines the length of the time window, which is set according to the frequency of the adjustment disturbance current in the voltage response signal. It is usually a multiple of the lowest frequency period to cover the complete period of all frequency components. Each time window maintains a fixed overlap with the previous time window, such as an overlap ratio of 50% to 75%, to ensure signal continuity and avoid information loss at the boundaries. During the segmentation process, the voltage response signal is divided into multiple consecutive time segments, each time segment corresponding to a segment signal. Each segment signal contains all the data points of the voltage response signal within the corresponding time window, preserving the dynamic characteristics of the voltage response signal. After the segmentation is completed, the segment signals are stored in an ordered sequence to ensure that the start and end times of each segment signal are clearly recorded.

[0068] Perform a Fast Fourier Transform on the segmented signal to output component amplitude and component phase information;

[0069] The Fast Fourier Transform (FFT) is performed separately for each segmented signal, converting the time-domain segmented signal into frequency-domain spectral data. The transformation involves decoupling the segmented signal into several single-frequency sinusoidal wave components and linearly superimposing them to obtain the composite frequency components of the voltage waveform in the segmented signal. The real and imaginary parts of the spectral data are used to characterize the intensity and phase shift of the corresponding frequency points, respectively. The component amplitude is obtained from the ratio of the real and imaginary parts, and the component phase information is obtained by obtaining the ratio of the real and imaginary parts through the arctangent function. After the transformation, component amplitude and component phase information containing multiple frequency points are generated. The component amplitude represents the voltage signal intensity corresponding to each frequency point, and the component phase information reflects the phase shift of the voltage signal relative to the time origin.

[0070] The component amplitude is compared with the disturbance current amplitude in the adjustment disturbance current at each frequency point to generate the impedance amplitude;

[0071] Specifically, the amplitude of the disturbance current is recorded when the power amplifier outputs, including the current signal strength at each frequency point, which corresponds one-to-one with the frequency points of the segmented signal. During the comparison process, for each frequency point, the ratio of the component amplitude to the corresponding disturbance current amplitude is statistically calculated to generate the impedance amplitude at that frequency point. During the comparison, it is ensured that the frequency points of the component amplitude and the disturbance current amplitude are strictly aligned to avoid comparison errors caused by frequency deviation. The impedance amplitude reflects the electrochemical response characteristics of the electric vehicle battery at a specific frequency. The larger the value, the higher the impedance of the battery at different frequencies. After the comparison is completed, the impedance amplitudes of all frequency points are integrated into an impedance amplitude sequence and arranged in frequency order.

[0072] The phase information of the component is compared with the phase information of the disturbance current in the regulating disturbance current to generate a phase shift angle;

[0073] The disturbance current phase information is recorded when the power amplifier outputs and adjusts the disturbance current. It includes the phase shift of the current signal at each frequency point, which corresponds one-to-one with the frequency points of the segmented signal. The phase comparison process identifies the difference between the component phase information and the disturbance current phase information at each frequency point and generates the phase shift angle of the frequency point. The phase shift angle represents the phase delay of the voltage response signal relative to the adjusted disturbance current, reflecting the electrochemical dynamic characteristics of the electric vehicle battery. When comparing the component phase information and the disturbance current phase information, the frequency points are aligned and the phase difference is normalized to eliminate possible periodic deviations. After the phase shift angle is generated, it is stored as a phase shift angle sequence indexed by the frequency point, maintaining the correspondence with the impedance amplitude sequence.

[0074] The impedance amplitude and phase shift angle at each frequency point are combined to output the battery electrochemical impedance parameters.

[0075] Specifically, the combination process pairs the impedance amplitude and phase shift angle at each frequency point to form complete impedance data points. Each impedance data point contains two parameters: the impedance amplitude and the phase shift angle at the frequency point. During combination, it is ensured that the frequency points of the impedance amplitude and the phase shift angle are strictly aligned to avoid data misalignment. The impedance data points of all frequency points are integrated to form ordered battery electrochemical impedance parameters. The battery electrochemical impedance parameters reflect the electrochemical behavior of electric vehicle batteries at different frequencies. The impedance amplitude represents the battery's resistance characteristics, and the phase shift angle reflects the battery's dynamic response characteristics.

[0076] The battery electrochemical impedance parameters are classified according to the phase shift angle, and the risk classification parameters are output.

[0077] The risk classification process first extracts the phase shift angle sequence from the battery's electrochemical impedance parameters, performs numerical analysis on the phase shift angle at each frequency point, and divides the phase shift angle into multiple intervals based on the range of the maximum and minimum values ​​of the phase shift angle. Each interval corresponds to a different risk level, which is usually determined based on the battery's electrochemical characteristics. For example, a larger phase shift angle may indicate a sluggish electrochemical reaction inside the battery, which is a higher risk. The classification process judges the phase shift angle at each frequency point one by one, determines the risk interval to which it belongs, and assigns a corresponding risk level label. The risk level labels of all frequency points are then integrated into risk classification parameters.

[0078] Based on the risk classification parameters, match the corresponding risk identifier to generate a risk identifier code;

[0079] The matching process first defines a risk identifier set, which contains multiple predefined identifier symbols. Each symbol corresponds to a specific risk level, such as low risk, medium risk, and high risk. During matching, each risk level label in the risk classification parameters is traversed, and the corresponding risk identifier is assigned according to the correspondence between the risk level label and the risk identifier set. The matching process ensures that the risk level label of each frequency point is mapped to a unique and consistent risk identifier to avoid ambiguity. The risk identifiers of all frequency points are integrated and summarized into a risk identifier code.

[0080] S3. Acquire high-frequency sound wave signals through an ultrasonic sensor array, extract energy integral values ​​from the high-frequency sound wave signals, fuse the risk identification code with the energy integral values ​​to determine the battery risk level of the electric vehicle battery.

[0081] High-frequency acoustic signals from new energy charging piles and electric vehicle batteries are acquired using an ultrasonic sensor array.

[0082] Specifically, the ultrasonic sensor array consists of multiple high-sensitivity ultrasonic sensors, which are arranged around the new energy charging piles and electric vehicle batteries to capture high-frequency sound wave signals generated during operation. The ultrasonic sensor array operates in a high-frequency sampling mode. Through precise positioning and orientation configuration, the ultrasonic sensor array collects high-frequency sound wave signals generated by electrochemical reactions, current flow or mechanical vibration during the charging process of new energy charging piles and electric vehicle batteries in real time. During the acquisition process, the ultrasonic sensor array performs preliminary filtering of environmental noise. After the acquisition is completed, the high-frequency sound wave signals are stored in the form of a continuous time series.

[0083] The high-frequency acoustic signal is smoothed and weighted point-by-point using the Hamming window function to generate a weighted coefficient sequence.

[0084] The Hamming window function dynamically generates a weighted coefficient sequence that perfectly matches the total number of sampling points of a high-frequency acoustic signal by acquiring the total number of sampling points in real time. The specific generation process is as follows: based on the total number of sampling points of the high-frequency acoustic signal, the weight value corresponding to each sampling point position is automatically assigned. The weight of the points in the weighted coefficient sequence is the highest (e.g., close to 1.0), and the weight value is gradually decreased along the direction of the beginning and end of the signal, with the weight of the two ends dropping to the lowest value (e.g., 0.08). During the generation process, the weight value of each sampling point is precisely bound to the position of the sampling point of the high-frequency acoustic signal, forming a continuous weighted distribution with center enhancement and boundary attenuation, i.e., the weighted coefficient sequence. Each coefficient of the weighted coefficient sequence corresponds to a sampling point of the high-frequency acoustic signal. The coefficient value is determined by the mathematical properties of the Hamming window function. The center point has a high weight, and the weight gradually decreases at both ends to achieve a smooth transition. The Hamming window function is applied point by point to each sampling point of the high-frequency acoustic signal, combining the amplitude of the high-frequency acoustic signal with the corresponding weighted coefficient to generate a smoothed weighted coefficient sequence. After the weighted coefficient sequence is generated, it is stored in the form of an ordered sequence.

[0085] The weighted coefficient sequence is combined point by point with the amplitude of the high-frequency acoustic signal to output a weighted high-frequency acoustic signal.

[0086] For each sampling point of the high-frequency acoustic signal, the amplitude of the sampling point is weighted and calculated with the corresponding weighting coefficient in the weighting coefficient sequence to generate a weighted amplitude. This point-by-point combination ensures that each sampling point of the high-frequency acoustic signal has undergone weighting processing, preserving the dynamic characteristics of the signal. At the same time, the smoothing effect of the weighting coefficient sequence reduces the impact of noise and abrupt changes. During the combination process, the time series structure of the high-frequency acoustic signal remains unchanged, and only the amplitude is adjusted to form a smooth weighted high-frequency acoustic signal. After the combination is completed, the weighted high-frequency acoustic signal is stored in time series form, containing the same number of sampling points as the high-frequency acoustic signal. The amplitude reflects the weighted acoustic intensity, providing high-quality input data for subsequent energy calculations.

[0087] The formula for weighted magnitude is:

[0088] ;

[0089] in, An identifier representing each sampling point in a high-frequency acoustic signal. Indicates the first The weighted amplitude of each sampling point Indicates the first The amplitude of each sampling point The Hamming window function is represented at the th... Weighting coefficients for each sampling point;

[0090] The weighted high-frequency acoustic signal is subjected to amplitude squaring, and the results of amplitude squaring are integrated to form an instantaneous energy sequence;

[0091] The weighted high-frequency acoustic signal, as a smoothed signal after point-by-point combination, undergoes weighted amplitude squared processing to obtain the energy characteristics of the signal. For each sampling point of the weighted high-frequency acoustic signal, the weighted amplitude squared processing statistically calculates the square of the amplitude at the sampling point to generate the instantaneous energy value of the corresponding sampling point. During the processing, the time series structure of the weighted high-frequency acoustic signal remains unchanged, generating an instantaneous energy value sequence with the same length as the weighted high-frequency acoustic signal. The instantaneous energy value sequence integrates the amplitude squared results of all sampling points to form an ordered instantaneous energy sequence, reflecting the energy distribution of the weighted high-frequency acoustic signal at each time point.

[0092] Accumulate the instantaneous energy sequence over a fixed time window and obtain the energy integral value for the corresponding fixed time window;

[0093] The accumulation process involves statistically summarizing all instantaneous energy values ​​within each fixed time window of the instantaneous energy sequence to generate the energy integral value for the corresponding time window. During accumulation, it is ensured that the boundaries of the time window are aligned with the sampling points of the instantaneous energy sequence to avoid data loss or duplication. The energy integral value of each fixed time window reflects the cumulative energy intensity of the weighted high-frequency acoustic signal within a fixed time period, embodying the acoustic characteristic intensity of new energy charging piles and electric vehicle batteries.

[0094] The risk identification codes are numerically mapped to generate an electrochemical risk score.

[0095] Specifically, the risk identification code, as an identification sequence generated based on the phase shift angle in the battery's electrochemical impedance parameters, is converted into an electrochemical risk score through numerical mapping to quantify the degree of risk. The numerical mapping process first defines mapping rules, which associate each risk identifier in the risk identification code with a specific numerical value. The mapping rules are determined according to the definition of low risk, medium risk, and high risk levels in the risk identifier set. The higher the risk identifier level, the larger the numerical value assigned to the risk identifier. For example, low risk is mapped to a smaller numerical value, and high risk is mapped to a larger numerical value. During mapping, each risk identifier in the risk identification code is traversed, and the corresponding numerical value is assigned according to the mapping rules to generate a numerical sequence with the same length as the risk identification code. The numerical value in the numerical sequence is the electrochemical risk score.

[0096] The electrochemical risk score and energy integral value are normalized respectively, and the standard risk score and standard energy integral are output.

[0097] Electrochemical risk scores and energy integral values, as quantitative data reflecting electrochemical risk and acoustic energy respectively, need to be normalized to unify their dimensions for easy subsequent fusion. Normalization is performed separately on electrochemical risk scores and energy integral values. The minimum-maximum normalization method is used to obtain the minimum and maximum values ​​of all electrochemical risk scores for the electrochemical risk scores. Each electrochemical risk score is linearly mapped to the range of 0 to 1 to generate a standard risk score.

[0098] For the energy integral value, each energy integral value is linearly mapped to the range of 0 to 1 to generate a standard energy integral. During the normalization process, the original sequence length and time of the electrochemical risk score and energy integral value remain unchanged, and only the numerical range is adjusted.

[0099] Based on the statistical results of historical failures, dynamic weights are assigned to the standard risk score and standard energy integral, and then weighted and fused to generate a comprehensive risk index.

[0100] The statistical results of historical faults are based on the operation records of new energy charging piles and electric vehicle batteries. The correlation between electrochemical risk and acoustic energy and the occurrence of faults is statistically analyzed. The weight allocation process is based on the numerical distribution of standard risk scores and standard energy integrals to obtain weight values. For example, when the number of statistical correlations between electrochemical risk and fault occurrences increases, the weight value of the standard risk score also increases dynamically. Similarly, when the number of statistical correlations between acoustic energy and fault occurrences increases, the weight value of the standard energy integral increases dynamically. The comprehensive risk index is generated by weighted fusion based on the standard risk score, standard energy integral, standard risk score weight value, and standard energy integral weight value.

[0101] The formula for the comprehensive risk index is as follows:

[0102] ;

[0103] in, This represents the comprehensive risk index generated by the weighted fusion of the standard risk score and the standard energy integral. This represents the standard risk score weight value. This represents the normalized standard risk score. This represents the normalized standard energy integral weight value. Represents the standard energy integral;

[0104] The K-means clustering algorithm is used to group the comprehensive risk index, and corresponding level labels are assigned to each group to output the battery risk level.

[0105] The K-means clustering algorithm iteratively optimizes the comprehensive risk index into risk level groups. The specific process revolves around the distribution characteristics of data points and the dynamic adjustment of cluster centers. First, the K-means algorithm sets the number of clusters based on the definition of risk levels. For example, it divides the comprehensive risk index into three groups: low risk, medium risk, and high risk. Then, it initializes three cluster centers using each value of the comprehensive risk index as an independent data point. During iteration, the K-means algorithm calculates the Euclidean distance between each data point and the current cluster center, assigns the data point to the group of the nearest cluster center, and updates the cluster center's position based on the mean of the data points within the group. This process is repeated until the cluster centers no longer change significantly. Finally, each cluster group is assigned a corresponding risk level label (such as low risk, medium risk, or high risk) based on the range of comprehensive risk index values ​​it contains. All data points form an ordered sequence of level labels according to their assigned cluster groups.

[0106] S4. Perform hierarchical power control based on the battery risk level and the preset risk power adjustment table, and output the charging pile power parameters.

[0107] Extract correlation data between battery risk and actual power adjustment from the historical operation logs of new energy charging piles;

[0108] Specifically, the historical operation logs of new energy charging piles contain detailed data on battery status, power adjustment parameters, and operation results recorded during the charging process. The correlation data between battery risk and actual power adjustment is extracted from the historical operation logs. The extraction process first filters the historical operation logs to identify records related to electric vehicle battery risk, including battery risk level, charging power value, and corresponding operating conditions. For each charging event, actual power adjustment data corresponding to the battery risk level is collected. The actual power adjustment data includes the amount of power change applied by the charging pile under a specific risk level. The extraction process ensures the integrity of the data, covering records from multiple charging cycles and different battery states, forming correlation data that includes battery risk level and actual power adjustment data.

[0109] Based on the power adjustment magnitude in the associated data, a risk-power mapping is performed on the temperature and charging efficiency changes of the electric vehicle battery to generate a preset risk power adjustment table.

[0110] The associated data contains the correspondence between battery risk levels and actual power adjustment values. By analyzing the impact of power adjustment magnitude on changes in electric vehicle battery temperature and charging efficiency, a preset risk power adjustment table is generated. The risk-power mapping process first statistically analyzes the actual power adjustment value corresponding to each battery risk level in the associated data, and then statistically obtains the average impact of the power adjustment value on changes in battery temperature and charging efficiency. The risk-power mapping process divides the battery risk level into multiple intervals, such as low risk, medium risk, and high risk, and assigns a corresponding power adjustment value to each interval. The power adjustment value is determined based on the statistical results of temperature and charging efficiency changes, ensuring that high-risk levels correspond to smaller power adjustments to reduce the risk of battery overheating or efficiency degradation. After the risk-power mapping is completed, a preset risk power adjustment table is generated.

[0111] According to the battery risk level, the power adjustment value of each new energy charging pile is obtained by searching in the preset risk power adjustment table in layers.

[0112] The hierarchical search is based on the battery risk level obtained during the current charging process. It iterates through the preset risk power adjustment table and matches the power adjustment value corresponding to the battery risk level. In the preset risk power adjustment table, each battery risk level is associated with a set of recommended power adjustment values. The power adjustment values ​​are expressed as a percentage or absolute value, reflecting the amount of power increase or decrease that the charging pile should apply. During the search, it is ensured that the battery risk level and the risk level range in the preset risk power adjustment table are accurately matched to avoid matching errors. For the case of multiple new energy charging piles operating at the same time, the battery risk level of each charging pile is searched separately to generate the power adjustment value corresponding to each charging pile.

[0113] The various power adjustment values ​​are integrated to output the power parameters of the charging pile.

[0114] The power adjustment value, as the recommended adjustment amount generated for each new energy charging pile after hierarchical lookup, needs to be integrated to output a unified charging pile power parameter. The integration process first collects the power adjustment values ​​of all new energy charging piles, ensuring that each power adjustment value is associated with the corresponding charging pile identifier to form the charging pile power parameter.

[0115] S5, the charging pile power regulator adjusts the new energy power according to the charging pile power parameters, and at the same time monitors the voltage change rate and temperature rise curve of the electric vehicle battery to verify the power distribution effect of the new energy charging pile.

[0116] Obtain the real-time charging power of the electric vehicle from the charging pile interface, and adjust the real-time power to new energy power according to the charging pile power parameters;

[0117] Specifically, the power monitoring equipment of the charging pile is connected to the charging pile interface to collect the real-time power applied to the electric vehicle battery during the current charging process. The collected real-time power is stored in the form of a time series. The adjustment process adds the power adjustment value to the real-time power according to the power parameters of each charging pile, thereby realizing the adjustment of the new energy power. The adjustment is executed by the charging pile power regulator. The power regulator precisely controls the output current or voltage according to the difference to ensure that the adjusted power is consistent with the power parameters of the charging pile.

[0118] The voltage value of the vehicle after the new energy power is adjusted is obtained in real time by a voltage sensor, and differential calculation is performed to generate the voltage change rate.

[0119] The voltage value of the electric vehicle battery is acquired in real time by a voltage sensor to assess the impact of power adjustment on the battery state. The voltage sensor is connected to the positive and negative terminals of the electric vehicle battery and continuously collects the change data of the battery terminal voltage after the power adjustment of the new energy vehicle at a high sampling rate. The collected voltage values ​​are stored in the form of a time series, recording the battery voltage at each time point to form a continuous voltage data sequence. Differential operation is performed on the voltage data sequence to calculate the difference between the voltage values ​​at adjacent time points and generate the voltage change rate at each time point according to the time interval. The differential operation is completed by processing point by point to ensure that the voltage change rate reflects the dynamic change trend of the battery voltage.

[0120] The formula for the rate of voltage change is:

[0121] ;

[0122] in, Identifiers representing points in time within a time interval. Indicates the first Voltage change rate at each time point Indicates the first The voltage value of the electric vehicle battery at a specific time point. Indicates the first The voltage value of the electric vehicle battery at a specific time point. This represents a fixed time interval between two adjacent time points;

[0123] The surface temperature of the electric vehicle battery is collected in real time by a temperature sensor and plotted as a temperature rise curve over time.

[0124] While adjusting the power of new energy vehicles, the surface temperature of electric vehicle batteries needs to be collected in real time by temperature sensors to monitor the impact of power adjustment on the battery's thermal state. The temperature sensors continuously collect battery surface temperature data at a high sampling rate, with the sampling frequency consistent with that of the voltage sensors, to ensure the consistency of time-series data. During the collection process, the temperature sensors correct for ambient temperature fluctuations, eliminate external thermal interference, and generate temperature data that reflects the true thermal state of the battery. The collected temperature data is stored in time-series format, recording the battery surface temperature value at each time point. The temperature rise curve plotting process arranges the temperature data sequence in chronological order and generates a continuous temperature curve through linear interpolation, reflecting the trend of battery surface temperature change over time. The temperature rise curve is plotted with time as the horizontal axis and temperature value as the vertical axis, preserving the dynamic characteristics of the temperature data sequence and providing thermal state data for subsequent synchronization with the voltage change rate.

[0125] Synchronize and align the voltage change rate with the temperature rise curve to generate temperature-voltage data pairs;

[0126] Voltage change rate and temperature rise curve, as dynamic data generated by voltage and temperature sensors respectively, need to be synchronized and aligned to generate temperature-voltage data pairs for comprehensive evaluation of battery status. The synchronization and alignment process first ensures that the time series of voltage change rate and temperature rise curve have the same time axis, which is based on a unified high-precision clock signal at the time of acquisition. During alignment, for each time point, the voltage change rate in the voltage change rate series and the temperature value in the temperature rise curve are extracted and paired to form temperature-voltage data pairs. The pairing process ensures that the voltage change rate and temperature value at each time point correspond strictly to avoid data misalignment caused by time deviation. If there is a slight deviation in the time series, the voltage change rate or temperature rise curve is fine-tuned by linear interpolation to fill in the missing time point data. After synchronization and alignment are completed, the temperature-voltage data pairs are stored in the form of time series, with each data pair containing a time point, voltage change rate, and temperature value, forming an ordered temperature-voltage data pair.

[0127] By comparing the power parameters and temperature and voltage data of charging piles on the same time axis, the dynamic relationship between power adjustment and battery state changes is determined, and the power distribution effect of new energy charging piles is verified.

[0128] Specifically, the charging power value and temperature / voltage data in the charging pile power parameters are first aligned with time points to ensure that the charging power value at each time point is correlated with the corresponding voltage change rate and temperature value. The synchronous trend of the charging power value change with the voltage change rate and temperature rise curve is statistically analyzed. For example, whether the voltage change rate slows down when the power is reduced, and whether the temperature rise curve tends to be stable, is used to determine whether the power adjustment effectively controls the electrochemical and thermal state of the battery. The dynamic relationship is determined by statistically analyzing the correlation between the change amplitude of the voltage change rate and temperature rise curve and the charging power value. If the voltage change rate and temperature rise curve remain stable after a high-risk power adjustment, it indicates that the power allocation effect is good. The verification results are stored in the form of time series comparative data, recording the relationship between the charging power value, voltage change rate, and temperature value at each time point, and generating an evaluation report on the power allocation effect of new energy charging piles to provide a basis for optimizing charging strategies.

[0129] This embodiment also provides a computer device applicable to the power allocation control method for new energy charging piles based on intelligent algorithms, including: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to implement the power allocation control method for new energy charging piles based on intelligent algorithms as proposed in the above embodiment.

[0130] The computer device can be a terminal, comprising a processor, memory, communication interface, display screen, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, carrier networks, NFC (Near Field Communication), or other technologies. The display screen can be an LCD screen or an e-ink screen. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad on the computer device's casing, or an external keyboard, touchpad, or mouse.

[0131] This embodiment also provides a storage medium storing a computer program. When executed by a processor, the program implements the power allocation control method for new energy charging piles based on intelligent algorithms as proposed in the above embodiments. The storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read Only Memory (EPROM), Programmable Red-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0132] In summary, this invention achieves multimodal comprehensive assessment of charging risks and precise dynamic control of power allocation for new energy charging piles by fusing risk identification codes and energy integral values. High-frequency acoustic signals are acquired using an ultrasonic sensor array, from which energy integral values ​​are extracted. These values ​​are then combined with weighted processing and energy accumulation quantification of the physical dimension signal, and normalized and dynamically weighted with the risk identification code to output a unified battery risk level. This achieves complementary enhancement of electrochemical and acoustic data, reduces single-factor misjudgments, and forms a comprehensive risk diagnosis basis. The verification mechanism monitors the voltage change rate and aligns the power adjustment effect with the temperature rise curve, optimizing the power strategy in a closed loop to adapt to actual charging conditions, ultimately ensuring the safe and reliable operation of electric vehicle batteries under dynamic conditions.

[0133] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A power distribution control method for new energy charging piles based on intelligent algorithms, characterized in that: include, Apply multi-frequency sinusoidal perturbation current to an electric vehicle battery and obtain the voltage response signal; The voltage response signal is decoupled by Fourier transform, the voltage frequency component is output and converted into battery electrochemical impedance parameters, and the risk identification code is marked according to the phase shift angle in the battery electrochemical impedance parameters; High-frequency sound wave signals are acquired through an ultrasonic sensor array, and energy integral values ​​are extracted from the high-frequency sound wave signals. The risk identification code and the energy integral value are fused together to determine the battery risk level of the electric vehicle battery. The battery risk level is used to perform hierarchical power control based on the preset risk power adjustment table, and the power parameters of the charging pile are output. The charging pile power regulator adjusts the new energy power according to the charging pile power parameters, and at the same time monitors the voltage change rate and temperature rise curve of the electric vehicle battery to verify the power distribution effect of the new energy charging pile. The specific steps for acquiring high-frequency sound wave signals through an ultrasonic sensor array and extracting energy integral values ​​from the high-frequency sound wave signals are as follows: High-frequency acoustic signals from new energy charging piles and electric vehicle batteries are acquired using an ultrasonic sensor array. The high-frequency acoustic signal is smoothed and weighted point-by-point using the Hamming window function to generate a weighted coefficient sequence. The weighted coefficient sequence is combined point by point with the amplitude of the high-frequency acoustic signal to output a weighted high-frequency acoustic signal. The weighted high-frequency acoustic signal is subjected to amplitude squaring, and the results of amplitude squaring are integrated to form an instantaneous energy sequence; Accumulate the instantaneous energy sequence over a fixed time window and obtain the energy integral value for the corresponding fixed time window; The specific steps for fusing the risk identification code and energy score to determine and output the battery risk level of the electric vehicle battery are as follows: The risk identification codes are numerically mapped to generate an electrochemical risk score. The electrochemical risk score and energy integral value are normalized respectively, and the standard risk score and standard energy integral are output. Based on the statistical results of historical failures, dynamic weights are assigned to the standard risk score and standard energy integral, and then weighted and fused to generate a comprehensive risk index. The K-means clustering algorithm is used to group the comprehensive risk index, and corresponding level labels are assigned to each group to output the battery risk level. The specific steps for implementing tiered power control based on battery risk levels and a preset risk power adjustment table, and outputting charging pile power parameters, are as follows: Extract correlation data between battery risk and actual power adjustment from the historical operation logs of new energy charging piles; Based on the power adjustment magnitude in the associated data, a risk-power mapping is performed on the temperature and charging efficiency changes of the electric vehicle battery to generate a preset risk power adjustment table. According to the battery risk level, the power adjustment value of each new energy charging pile is obtained by searching in the preset risk power adjustment table in layers. The various power adjustment values ​​are integrated to output the power parameters of the charging pile.

2. The power allocation control method for new energy charging piles based on intelligent algorithms according to claim 1, characterized in that: The specific steps for applying a multi-frequency sinusoidal perturbation current to the electric vehicle battery and obtaining the voltage response signal are as follows: The multi-frequency sinusoidal disturbance current is input into the power amplifier for amplitude adjustment, and the output is the adjusted disturbance current; The disturbance current is injected into the electric vehicle battery to monitor the dynamic response of the internal voltage of the electric vehicle battery in real time and obtain the voltage waveform. The voltage waveform is sampled synchronously in time to generate segmented voltage data; The segmented voltage data is interpolated to generate a voltage response signal.

3. The power allocation control method for new energy charging piles based on intelligent algorithms according to claim 2, characterized in that: The specific steps for decoupling the voltage response signal using Fourier transform, outputting the voltage frequency component, and converting it into battery electrochemical impedance parameters are as follows: The voltage response signal is divided into multiple overlapping time windows, and segmented signals are output. Perform a Fast Fourier Transform on the segmented signal to output component amplitude and component phase information; The component amplitude is compared with the disturbance current amplitude in the adjustment disturbance current at each frequency point to generate the impedance amplitude; The phase information of the component is compared with the phase information of the disturbance current in the regulating disturbance current to generate a phase shift angle; The impedance magnitude and phase shift angle are combined to output the battery electrochemical impedance parameters.

4. The power allocation control method for new energy charging piles based on intelligent algorithms according to claim 3, characterized in that: The specific steps for labeling risk identification codes based on the phase shift angle in the battery's electrochemical impedance parameters are as follows: The battery electrochemical impedance parameters are classified according to the phase shift angle, and the risk classification parameters are output. Based on the risk classification parameters, the corresponding risk identifier is matched to generate a risk identifier code.

5. The power allocation control method for new energy charging piles based on intelligent algorithms according to claim 4, characterized in that: The charging pile power regulator adjusts the new energy power according to the charging pile power parameters, and simultaneously monitors the voltage change rate and temperature rise curve of the electric vehicle battery to verify the power distribution effect of the new energy charging pile. The specific steps are as follows: Obtain the real-time charging power of the electric vehicle from the charging pile interface, and adjust the real-time power to new energy power according to the charging pile power parameters; The voltage value of the vehicle after the new energy power is adjusted is obtained in real time by a voltage sensor, and differential calculation is performed to generate the voltage change rate. The surface temperature of the electric vehicle battery is collected in real time by a temperature sensor and plotted as a temperature rise curve over time. Synchronize and align the voltage change rate with the temperature rise curve to generate temperature-voltage data pairs; By comparing the power parameters and temperature and voltage data of charging piles on the same time axis, the dynamic relationship between power adjustment and battery state changes can be determined, and the power distribution effect of new energy charging piles can be verified.

6. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that: When the processor executes the computer program, it implements the steps of the power distribution control method for new energy charging piles based on intelligent algorithms as described in any one of claims 1 to 5.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by the processor, it implements the steps of the power distribution control method for new energy charging piles based on intelligent algorithms as described in any one of claims 1 to 5.

Images