A dynamic reward method and system for participants in a charging pile service chain

Abstract

This application provides a dynamic reward method and system for participants in the charging pile service chain, belonging to the field of charging operation management technology. This application synchronously acquires the total harmonic distortion (THD) data of the grid connection point and the inverter's switching status data. After short-time Fourier transform and state-space matrix construction, wavelet packet decomposition is used to accurately separate the compensation harmonic components actively emitted by the inverter from the inherent harmonic components of the power grid. By analyzing the phase amplitude relationship between the two, a purification contribution is generated, and this contribution is input into an adaptive PID neural network model to calculate a service performance score. Finally, a dynamic reward value is generated based on the score and a price tier. This application can accurately quantify the actual technical contribution of charging piles to power grid harmonic governance, achieving fair and dynamic reward allocation based on active filtering performance, and incentivizing operators to improve power quality governance.

Description

Technical Field

[0001] This application belongs to the field of charging operation and management technology, and in particular relates to a dynamic reward method and system for participants in the charging pile service chain. Background Technology

[0002] With the rapid development of the new energy vehicle industry, the large-scale integration of charging infrastructure has placed higher demands on the power quality of the power grid, making vehicle-grid interaction and power auxiliary services key areas for industry development. Establishing an effective dynamic incentive mechanism can guide participants in the charging pile service chain to actively utilize their equipment capabilities to participate in grid regulation, which has broad application prospects for improving grid stability and promoting energy consumption.

[0003] Existing charging pile service chain reward methods mainly rely on construction subsidies or simple metering and settlement based on charging volume and active power response. Although some solutions introduce demand response mechanisms, they usually only conduct extensive revenue distribution based on the completion of dispatch instructions or peak-valley electricity price differences, focusing on simple load shaving and valley filling.

[0004] However, existing methods often lack the ability to finely assess the effectiveness of power quality management, making it difficult to accurately identify and separate whether harmonics at the grid connection point are caused by inherent background pollution on the grid side or by active compensation or pollution from charging equipment. This lack of assessment dimensions makes it impossible to accurately quantify the actual technical contributions of operators in ancillary services such as active filtering, thus hindering the achievement of fair and effective incentives. Therefore, existing technologies suffer from insufficient dynamic rewards due to the inability to accurately and dynamically reward based on actual power quality management contributions. Summary of the Invention

[0005] The purpose of this application is to provide a dynamic reward method and system for participants in the charging pile service chain, so as to solve the problem of insufficient dynamic rewards in the prior art.

[0006] To address the aforementioned technical problems, in a first aspect, this application provides a dynamic reward method for participants in a charging pile service chain, comprising:

[0007] Simultaneously acquire the total harmonic distortion of the charging station and the public power grid connection point, as well as the pulse width modulation switching frequency and duty cycle status of the inverter inside the charging pile power module;

[0008] By performing a short-time Fourier transform on the total harmonic distortion, a harmonic characteristic spectrum is generated, and a switching state space matrix is ​​constructed based on the pulse width modulation switching frequency and duty cycle.

[0009] Based on the harmonic feature spectrum and the switch state space matrix, the first harmonic component and the second harmonic component are extracted by wavelet packet decomposition. The first harmonic component is the harmonic compensation component actively emitted by the charging pile inverter, and the second harmonic component is the inherent harmonic component of the public power grid.

[0010] The contribution is generated by superimposing the phase of the first harmonic component and comparing the amplitude of the second harmonic component;

[0011] The contribution is used as a feedforward variable input to a neural network model based on adaptive PID control. The preset power quality parameters are used as the objective function to calculate the service performance score, which represents the performance level of the charging pile operator in participating in the active filter auxiliary service.

[0012] Based on service performance scores, a target price tier is determined within a pre-defined correspondence between performance scores and price tiers. A target dynamic reward value is then generated based on the service performance scores and the target price tier.

[0013] Optionally, the method further includes:

[0014] The phase and amplitude of the voltage fundamental wave at the grid connection point of the public power grid are acquired synchronously, and the phase angle and amplitude modulation ratio of the inverter output voltage fundamental wave are extracted from the switch state space matrix.

[0015] Calculate the phase difference between the phase and the phase angle, and calculate the reactive power compensation based on the phase difference, amplitude modulation ratio, and amplitude.

[0016] The contribution is generated by superimposing the phase of the first harmonic component and comparing its amplitude with that of the second harmonic component, including:

[0017] An initial contribution is generated by superimposing the phases of the first harmonic component and comparing the amplitudes of the second harmonic component, and the contribution is determined based on the initial contribution and the reactive power compensation.

[0018] Optionally, an initial contribution is generated by superimposing the phase of the first harmonic component and comparing the amplitude of the second harmonic component, and the contribution is determined based on the initial contribution and the reactive power compensation, including:

[0019] Calculate the phase difference and amplitude ratio between the first harmonic component and the second harmonic component, and when the phase difference is within a preset difference range and the amplitude ratio is within a preset compensation range, use the amplitude ratio as the initial contribution.

[0020] Calculate the deviation ratio between the reactive power compensation amount and the preset reactive power demand value, and use the preset reactive power weighting coefficient to weight the deviation ratio to obtain the reactive power contribution.

[0021] The initial contribution and the reactive contribution are weighted and summed to obtain the contribution value.

[0022] Optionally, before inputting the contribution as a feedforward variable into the neural network model based on adaptive PID control, the method further includes:

[0023] Calculate the real-time phase difference between the first harmonic component and the second harmonic component;

[0024] A safe phase margin is generated by calculating the absolute difference between the real-time phase difference and the preset critical phase value.

[0025] The contribution is nonlinearly weighted and adjusted using the safety phase margin.

[0026] Optionally, a switching state space matrix is ​​constructed based on the pulse width modulation switching frequency and duty cycle state, including:

[0027] The number of pulse cycles per unit time is determined based on the pulse width modulation switching frequency, and the conduction time sequence within each pulse cycle number is extracted according to the duty cycle state.

[0028] The conduction time series is mapped to a discrete state series, and the discrete state series is continuously truncated using a preset sliding window to obtain multiple local state vectors.

[0029] Multiple local state vectors are stacked in a temporal order to construct a switch state space matrix.

[0030] Optionally, based on the harmonic feature spectrum and the switching state space matrix, wavelet packet decomposition is used to extract the first harmonic component and the second harmonic component, including:

[0031] The spectral distribution features are extracted by performing a frequency domain transformation on the switch state space matrix;

[0032] The harmonic characteristic spectrum is decomposed into sub-band sequences of multiple independent frequency bands, and the correlation coefficient between each sub-band sequence and the spectral distribution characteristics is calculated using covariance calculation.

[0033] The sub-band sequence with a correlation coefficient greater than a preset correlation threshold is determined as the first sub-band, and the sub-band sequence with a correlation coefficient less than or equal to the preset correlation threshold is determined as the second sub-band;

[0034] The first harmonic component and the second harmonic component are obtained by reconstructing the first sub-band and the second sub-band by performing inverse wavelet packet transform on them respectively.

[0035] Optionally, the contribution level is used as a feedforward variable input to a neural network model based on adaptive PID control, with preset power quality parameters as the objective function, to calculate a service performance score, including:

[0036] The contribution is input as a feedforward variable into a neural network model based on adaptive PID control, and the preset power quality parameters are set as the target reference trajectory.

[0037] The backpropagation algorithm is used to calculate the error between the feedforward variables and the target reference trajectory, and the weights of the neural network model are updated using the error to obtain the convergence residual value and the weight change rate.

[0038] The first performance score is obtained by weighting the convergence residual value using a preset accuracy weight, and the second performance score is obtained by weighting the weight change rate using a preset speed weight. The service performance score is obtained by weighted summing of the first and second performance scores.

[0039] Secondly, this application provides a dynamic reward system for participants in a charging pile service chain, including:

[0040] The acquisition module is used to synchronously acquire the total harmonic distortion of the charging station and the public power grid connection point, as well as the pulse width modulation switching frequency and duty cycle status of the inverter inside the charging pile power module.

[0041] The generation module is used to generate a harmonic feature spectrum by performing a short-time Fourier transform on the total harmonic distortion, and to construct a switching state space matrix based on the pulse width modulation switching frequency and duty cycle state.

[0042] The extraction module is used to extract the first harmonic component and the second harmonic component based on the harmonic feature spectrum and the switch state space matrix using wavelet packet decomposition. The first harmonic component is the harmonic compensation component actively emitted by the charging pile inverter, and the second harmonic component is the inherent harmonic component of the public power grid.

[0043] The generation module is also used to generate a contribution by superimposing the phase of the first harmonic component and comparing the amplitude of the second harmonic component;

[0044] The calculation module is used to input the contribution as a feedforward variable into the neural network model based on adaptive PID control, and to calculate the service performance score with the preset power quality parameters as the objective function. The service performance score represents the performance level of the charging pile operator in participating in the active filter auxiliary service.

[0045] The generation module is also used to determine the target price tier based on the service performance score and the pre-defined correspondence between the performance score and the price tier, and to generate a target dynamic reward value based on the service performance score and the target price tier.

[0046] Thirdly, this application provides an electronic device, comprising:

[0047] Memory, used to store computer programs;

[0048] A processor is configured to execute the computer program to implement the steps of the dynamic reward method for participants in the charging pile service chain as described in the first aspect above.

[0049] Fourthly, this application provides a computer-readable storage medium storing a computer program that, when executed by a processor, can implement the steps of the dynamic reward method for participants in the charging pile service chain as described in the first aspect above.

[0050] The dynamic reward method for participants in the charging pile service chain provided in this application firstly combines the inverter switching status on the equipment side with the harmonic distortion characteristics on the grid side, and uses wavelet packet decomposition technology to achieve source-end decoupling and component separation of complex harmonic signals at the grid connection point, thereby accurately identifying the compensation component actively emitted by the charging equipment and the inherent component of the grid.

[0051] Secondly, based on this precise signal separation result, the actual governance contribution of the equipment was quantified through phase superposition analysis. An adaptive PID neural network model was then used to dynamically and intelligently score this contribution, transforming abstract technical indicators into measurable service performance. This effectively solves the problems of existing technologies being unable to distinguish harmonic sources and quantify governance effects, ensuring the scientific and fair allocation of rewards and significantly enhancing the operator's enthusiasm for participating in power grid power quality governance. Therefore, this application achieves precise and efficient dynamic rewards based on the actual power quality governance contribution of equipment.

[0052] Furthermore, this application analyzes and calculates the reactive power compensation of the charging pile at the fundamental level by deeply mining the fundamental control information in the switch state space matrix, and integrates this reactive power contribution with the harmonic mitigation contribution for evaluation. This design breaks through the limitation of focusing only on harmonic mitigation in a single dimension, makes full use of the inverter capability of the charging equipment, and realizes the comprehensive quantification of the dual auxiliary service value of "purifying the power grid" and "supporting voltage".

[0053] This multi-dimensional contribution calculation method not only more comprehensively reflects the actual regulatory role of charging pile service chain participants in the power grid, but also encourages operators to balance reactive power while controlling harmonics, further optimizing the overall power quality and stability of the power grid. Therefore, this application effectively expands the evaluation dimensions of dynamic rewards and enhances the comprehensiveness and systematic nature of the ancillary service incentive mechanism. Attached Figure Description

[0054] To more clearly illustrate the technical solutions of the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0055] Figure 1 A flowchart illustrating a dynamic reward method for participants in a charging pile service chain, provided as an embodiment of this application;

[0056] Figure 2 A flowchart illustrating a method for generating harmonic components provided in an embodiment of this application;

[0057] Figure 3 A flowchart illustrating a method for generating service performance scores provided in an embodiment of this application;

[0058] Figure 4 A schematic diagram of a dynamic reward system for participants in a charging pile service chain, provided in an embodiment of this application;

[0059] Figure 5 This is a schematic diagram of the hardware structure of an electronic device provided in one embodiment of this application. Detailed Implementation

[0060] In the dynamic reward mechanism of the charging pile service chain, existing technologies generally suffer from a key deficiency: a single evaluation dimension and insufficient granularity. Current incentive models are mostly based on simple measurement of electricity or power, making it difficult to distinguish the responsibility for power quality fluctuations at the grid connection point and to accurately identify whether harmonics originate from the inherent background of the power grid or the operation of the charging equipment. This "black box" evaluation status makes it impossible to effectively quantify the technical contributions of operators to harmonic mitigation, making it difficult for those who truly invest resources to improve power grid quality to obtain commensurate benefits.

[0061] To address the aforementioned issues, this application proposes a dynamic reward method for participants in the charging pile service chain. Its core lies in constructing a data fusion and analysis mechanism spanning the grid and equipment sides. First, the method simultaneously collects the harmonic distortion at the grid connection point and the inverter's switching status. Utilizing wavelet packet decomposition technology combined with time-frequency domain features, it accurately separates the compensation components actively emitted by the inverter from the inherent grid components. Second, through joint evaluation using phase superposition analysis and an adaptive PID neural network model, it dynamically quantifies the actual purification contribution of the equipment to the grid and generates a performance score. Finally, it matches a price tier to achieve precise rewards.

[0062] This application ensures that the reward distribution is strictly linked to the actual technical contribution by quantitatively calculating the source-end decoupling and governance effect of harmonic signals. This avoids erroneous rewards for non-governance behaviors and guarantees reasonable income for high-quality operators. It solves the technical problem in the prior art that the dynamic reward is insufficient due to the inability to accurately and dynamically reward based on the actual power quality governance contribution, and enhances the enthusiasm and technical efficiency of charging infrastructure to participate in grid ancillary services.

[0063] To enable those skilled in the art to better understand the present application, the present application will be further described in detail below with reference to the accompanying drawings and specific embodiments. Obviously, the described embodiments are merely some embodiments of the present application, and not all embodiments. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0064] To address the problems of existing technologies, embodiments of this application provide a method, apparatus, device, computer storage medium, and computer program product for dynamic rewards of participants in a charging pile service chain. The dynamic reward method for participants in a charging pile service chain provided in this application embodiment will be described first below.

[0065] Figure 1 This illustration shows a flowchart of a dynamic reward method for participants in a charging pile service chain according to an embodiment of this application. Figure 1 As shown, the method includes:

[0066] S101. Synchronously acquire the total harmonic distortion of the charging station and the public power grid connection point, as well as the pulse width modulation switching frequency and duty cycle status of the inverter inside the charging pile power module.

[0067] Total harmonic distortion (THD) is the ratio of the effective value of harmonic components to the effective value of the fundamental component in the grid-connected voltage waveform, used to quantify the degree of distortion in the grid voltage waveform. Pulse width modulation (PWM) switching frequency refers to the number of times the inverter power devices are turned on and off per unit time. Duty cycle status refers to the sequence of ratios between the on-time of the power devices within one switching cycle and the entire cycle.

[0068] First, a high-precision power quality analysis device is deployed at the grid connection point between the charging station and the public power grid. This device collects the three-phase voltage signals from the grid side in real time at a preset high-frequency sampling rate and outputs the total harmonic distortion (THD) data sequence in real time through a built-in fast calculation module. Simultaneously, it directly reads the register data from the digital controller inside the charging pile via the data bus, synchronously capturing the pulse-width modulation (PWM) switching frequency value of the inverter power module at the corresponding moment and the real-time duty cycle state sequence of each phase arm. Finally, these two sets of cross-domain data are timestamped to form a synchronized dataset that includes grid-side quality characteristics and equipment-side control characteristics.

[0069] For example, suppose that at the grid connection point of charging station A, the power quality analysis device collects total harmonic distortion data of H=[0.03, 0.032, 0.028] at a certain moment, which corresponds to the distortion of the three-phase voltage. At the same time, the main control unit of the charging pile obtains the pulse width modulation switching frequency of the inverter at the current moment as F=10000, and the corresponding duty cycle state sequence of phase A bridge arm in three consecutive cycles as D=[0.45, 0.48, 0.52].

[0070] S102. By performing a short-time Fourier transform on the total harmonic distortion, a harmonic characteristic spectrum is generated, and a switching state space matrix is ​​constructed based on the pulse width modulation switching frequency and duty cycle state.

[0071] A harmonic characteristic spectrum is a data matrix that displays the energy distribution characteristics of total harmonic distortion on a joint two-dimensional plane of time and frequency. It can include a set of time-frequency pixels reflecting the amplitude intensity of each harmonic at different times. A switching state-space matrix is ​​a structured numerical matrix formed by mapping the discrete switching action sequence of the inverter to a multi-dimensional state space, used to represent the logic action trajectory of the inverter within a continuous time window.

[0072] First, a time window function of fixed length is defined, which slides along the time axis with a preset step size to capture local segments from the total harmonic distortion (THD) data sequence. Next, a Fast Fourier Transform (FFT) is performed on each captured local segment to calculate the spectral distribution of the signal within that time window, obtaining the corresponding frequency amplitude vector. Finally, by arranging and concatenating the frequency amplitude vectors calculated from all sliding windows in chronological order along the time axis, a three-dimensional data volume including time, frequency, and amplitude dimensions is generated, which is the harmonic characteristic spectrum.

[0073] For example, for the input data H, after transformation, a spectrum matrix M can be obtained, where the element M(t, f) represents the amplitude of the harmonic component with frequency f at time t. Secondly, for the data on the inverter side, a basic time base is determined based on the acquired pulse width modulation switching frequency. Combined with the conduction duration information reflected by the duty cycle state, the continuous pulse waveform is abstracted into a discrete "0-1" level logic sequence. These sequences are then stacked in time sequence to construct a switching state space matrix S reflecting the device's operational logic.

[0074] Optionally, the process of constructing the switching state space matrix based on the pulse width modulation switching frequency and duty cycle state in step S102 may specifically include:

[0075] S1021. Determine the number of pulse cycles per unit time based on the pulse width modulation switching frequency, and extract the conduction time sequence within each pulse cycle based on the duty cycle state.

[0076] The number of pulse cycles refers to the number of complete switching cycles completed by the inverter within a selected reference time unit, reflecting the time-domain density of switching actions. It can be an integer count value calculated based on frequency. The conduction time sequence refers to the set of numerical values ​​of the specific duration for which the power switching device is in the conduction state within each consecutive pulse cycle.

[0077] First, the previously acquired pulse width modulation (PWM) switching frequency value of 10kHz is read. With a time unit of 1 second selected, the number of pulse cycles is calculated to be 10,000. This means there are 10,000 independent switching cycles within 1 second, and the theoretical duration of each cycle is 1 / 10,000 of a second, or 100 microseconds. Next, all duty cycle state data acquired within this second are iterated through. For each specific pulse cycle, the formula is used... Calculate the conduction time within this cycle, where For the duration of the cycle, This represents the duty cycle at that moment. Finally, all the calculated conduction times are arranged in chronological order to form a conduction time sequence.

[0078] For example, assuming the duty cycles of three consecutive periods are 0.5, 0.6, and 0.4, and the period duration is 100us, the calculated conduction time series segment is Tseq=[50us, 60us, 40us].

[0079] S1022. Map the conduction time sequence to a discrete state sequence, and use a preset sliding window to continuously extract the discrete state sequence to obtain multiple local state vectors.

[0080] A discrete state sequence refers to a sequence of high and low states converted from continuous analog time quantities into digital logic levels. It is used to simulate the physical switching actions of inverter switches and can be a long binary data stream consisting of 0s and 1s. A local state vector is a short sequence segment of fixed length extracted from a long sequence, representing the switching action pattern within a small time period. It can be a row vector containing hundreds of logic state bits.

[0081] First, set a sampling resolution much higher than the switching frequency, such as 1µs, and expand the conduction time sequence obtained in the previous step. Second, for each pulse cycle, mark the state as "1" during the corresponding conduction duration and as "0" during the remaining off duration. For example, for a 50µs conduction and 50µs off cycle, a sequence segment of 50 1s followed by 50 0s will be generated. Concatenate the segments of all cycles to obtain the complete discrete state sequence. Finally, set a length of... , with step size It slides across the discrete state sequence, extracting data from the window with each slide. Each slide outputs the data within the window as a local state vector.

[0082] For example, assuming the first 10 bits of the discrete state sequence are 1111100000, and the window length is set to 5 and the step size is 5, then the first local state vector extracted is V1=[1,1,1,1,1], and the second is V2=[0,0,0,0,0].

[0083] S1023. Stack multiple local state vectors according to their temporal order to construct a switch state space matrix.

[0084] Stacking refers to the operation of combining and arranging multiple vectors according to specific spatial geometric rules to increase the dimension of data or construct a matrix structure, which can include row stacking or column stacking.

[0085] First, the multiple local state vectors generated in the previous step are used as rows of a matrix and arranged in chronological order of generation. Assume a total of [number missing] local state vectors were generated. There are n local state vectors, each vector having a length of n. Then place the first vector in the first row of the matrix, the second vector in the second row, and so on, until the first vector is placed in the second row. A vector. Ultimately, a... A dimensional switch-state space matrix. For example, assuming three local state vectors V1, V2, and V3, the constructed switch-state space matrix S is:

[0086] ;

[0087] This matrix S completely preserves all the spatiotemporal action characteristics of the inverter over a period of time.

[0088] This embodiment achieves precise alignment between the grid-side spectral characteristics and the equipment-side control logic in the spatiotemporal dimensions by generating a time-frequency spectrum of total harmonic distortion and constructing an inverter switching state-space matrix. In particular, by converting the conduction time series into discrete state vectors and stacking them, the micro-temporal patterns of inverter switching actions are fully captured, improving the accuracy of signal tracing.

[0089] S103. Based on the harmonic feature spectrum and the switch state space matrix, the first harmonic component and the second harmonic component are extracted by wavelet packet decomposition. The first harmonic component is the harmonic compensation component actively emitted by the charging pile inverter, and the second harmonic component is the inherent harmonic component of the public power grid.

[0090] Optionally, step S103, which involves extracting the first harmonic component and the second harmonic component using wavelet packet decomposition based on the harmonic feature spectrum and the switching state space matrix, may specifically include:

[0091] Figure 2 A schematic flowchart of a method for generating harmonic components according to an embodiment of this application is shown. Figure 2 As shown, the method includes:

[0092] S1031. Extract the spectral distribution features by performing frequency domain transformation on the switch state space matrix.

[0093] The spectral distribution characteristic refers to the energy density distribution curve formed after mapping the switching operation logic in the time domain to the frequency domain. It is used to represent the inherent emission intensity of the inverter switching operation at different frequency points and can be an amplitude vector including the fundamental frequency, the switching frequency and its harmonic components.

[0094] Regarding the previously constructed switch state space matrix This involves applying a two-dimensional Fast Fourier Transform (FFT) or a column-wise Fourier Transform along the time axis. Specifically, first, each column vector in the matrix is ​​treated as an independent discrete-time signal, and its spectrum is calculated. Then, a weighted average or maximum envelope extraction is performed on the spectrum results of all columns to obtain a one-dimensional spectrum vector that covers the entire frequency range, which is the spectral distribution characteristic.

[0095] For example, suppose the switching state space matrix of charging station A is: After transformation, the extracted spectral distribution features are represented as a vector of length 1024. In this vector, a significant amplitude spike is assumed to appear at the index positions corresponding to the pulse width modulation switching frequency of 10 kHz and its harmonic of 20 kHz, while the amplitude is close to zero in other frequency bands, thus characterizing the current transmit spectrum profile of the device.

[0096] S1032. Decompose the harmonic characteristic spectrum into multiple independent frequency band sub-band sequences, and use covariance calculation to calculate the correlation coefficient between each sub-band sequence and the spectral distribution characteristics.

[0097] A sub-band sequence refers to a set of narrowband components after a broadband harmonic signal is divided into high and low frequencies using signal decomposition techniques. It can include a low-frequency overview and high-frequency details at different levels.

[0098] First, the wavelet packet decomposition algorithm is used to analyze the previously generated harmonic feature map. Multi-level processing is performed. A suitable wavelet basis function for power signal analysis is selected, and the number of decomposition levels is set to [value missing]. The harmonic characteristic spectrum is decomposed into Each sub-band sequence is then transformed to the frequency domain and compared with the spectral distribution features extracted in step S1031. The comparison is performed within the corresponding frequency range. Finally, by calculating the covariance between the two, the degree of matching between each sub-band sequence and the inverter switching action characteristics is obtained, i.e., the correlation coefficient.

[0099] For example, suppose the harmonic feature map is decomposed into sub-band sequences. to .in, The frequency band covers a range around 10kHz, and calculations show that it corresponds to the spectral distribution characteristics. correlation coefficient A value of 0.85 indicates that the energy in this frequency band closely matches the inverter's operating characteristics; while the value covering low-frequency even harmonics... Correlation coefficient of frequency band The value of 0.12 indicates that this frequency band has a weak correlation with inverter operation.

[0100] S1033. The sub-band sequence with a correlation coefficient greater than the preset correlation threshold is determined as the first sub-band, and the sub-band sequence with a correlation coefficient less than or equal to the preset correlation threshold is determined as the second sub-band.

[0101] The first sub-band refers to the set of signal components determined to be actively emitted by the charging pile inverter, representing the active influence of the equipment on the grid connection point. The second sub-band refers to the set of signal components determined to be unrelated to the charging pile's operation, representing the inherent background of the grid side. The preset association threshold is a critical value used to distinguish signal sources. The preset association threshold is set according to the complexity of the grid environment, as shown in Table 1 below:

[0102] Table 1: Preset Association Threshold Comparison Table

[0103]

[0104] Table 1 lists the recommended threshold settings for different power grid ambient noise levels. In low-noise environments, the threshold is set lower to more sensitively detect the subtle emissions from the equipment; while in high-noise environments, the threshold is set higher to prevent background noise from being misinterpreted as equipment emissions.

[0105] First, by statistically analyzing the energy of the second harmonic component in the characteristic frequency band without active compensation, the background noise level of the current power grid is estimated, thereby determining the environmental noise level of the current grid connection point. Table 1 is then consulted to determine the applicable preset correlation threshold. Next, all correlation coefficients calculated in step S1032 are iterated and compared one by one with the threshold. If the correlation coefficient of a sub-frequency band is greater than the threshold, it is classified into the first sub-frequency band list; otherwise, it is classified into the second sub-frequency band list. For example, assuming the current environmental assessment is Level II, the preset correlation threshold is selected according to Table 1. For the previously calculated The coefficient is 0.85, because ,therefore It was designated as the first sub-band; for The coefficient is 0.12, because ,therefore It was designated as the second sub-band.

[0106] S1034. The first harmonic component and the second harmonic component are obtained by reconstructing the first sub-band and the second sub-band by performing wavelet packet inverse transform on them respectively.

[0107] First, a new wavelet packet coefficient node tree is constructed. All coefficients identified as belonging to the first sub-band retain their original values, while all coefficient nodes corresponding to the second sub-band are forcibly set to zero. Then, the inverse transform algorithm is executed, and the output time-domain waveform represents the first harmonic component. Similarly, the first sub-band coefficients are set to zero, only the second sub-band coefficients are retained, and an inverse transform is performed; the output time-domain waveform represents the second harmonic component.

[0108] For example, utilizing the preserved The first harmonic component is reconstructed from the coefficient tree of the equal correlation frequency band. This component clearly demonstrates the high-frequency ripple caused by the inverter switching action; utilizing the preserved The second harmonic component is reconstructed from the coefficient tree of the low correlation band. This component restores the original low-frequency distortion characteristics of the power grid side. Ultimately, it achieves the decomposition of the mixed total harmonic distortion signal into... and The physical process.

[0109] This embodiment extracts the spectral distribution characteristics of the inverter's switching state as a reference fingerprint, and combines the multi-resolution characteristics of wavelet packet decomposition to refine the complex grid harmonic signal into independent sub-frequency bands and perform correlation screening. It can accurately identify and separate the first harmonic component that highly matches the charging pile's switching action and the second harmonic component originating from the grid background, effectively overcoming the problem of unclear responsibility attribution under mixed signals.

[0110] S104. The contribution is generated by superimposing the phase of the first harmonic component and comparing the amplitude of the second harmonic component.

[0111] Contribution refers to a comprehensive evaluation index that quantifies, through numerical means, the positive or negative impact of a charging pile inverter on the power quality of the power grid at a given moment.

[0112] Optionally, the method further includes:

[0113] The phase and amplitude of the voltage fundamental wave at the grid connection point of the public power grid are acquired synchronously, and the phase angle and amplitude modulation ratio of the inverter output voltage fundamental wave are extracted from the switch state space matrix.

[0114] The amplitude modulation ratio refers to the proportionality coefficient between the fundamental voltage amplitude of the inverter output and the DC bus voltage utilization rate in pulse width modulation control. The fundamental frequency of the inverter output voltage refers to the sinusoidal component with a frequency consistent with the power grid frequency, which is hidden in the high-frequency pulse sequence of the switching state space matrix.

[0115] First, for the public grid side, phase-locked loop (PLL) technology or discrete Fourier transform (DFT) algorithm is used to analyze the collected grid connection point voltage signal in real time, extracting the phase and amplitude values ​​of the fundamental frequency. Second, for the charging pile side, the switch state space matrix constructed in step S102 is subjected to low-pass filtering or spectral analysis of the fundamental frequency point. By extracting the fundamental component, the phase angle and corresponding amplitude modulation ratio of the equivalent fundamental voltage actually output by the inverter at that moment are deduced.

[0116] For example, suppose that at a certain moment the fundamental amplitude of the grid voltage is detected to be 311V and the phase is 0 radians; at the same time, by analyzing the switch state space matrix, the fundamental phase angle of the inverter output voltage is extracted to be 0.05 radians and the amplitude modulation ratio is 0.95.

[0117] Calculate the phase difference between the phase and the phase angle, and calculate the reactive power compensation based on the phase difference, amplitude modulation ratio, and amplitude.

[0118] Phase difference refers to the angular deviation between the inverter's output voltage fundamental wave and the grid voltage fundamental wave on the time axis, determining the direction and nature of power flow. Reactive power compensation refers to the reactive power injected into or absorbed from the grid by the charging pile inverter, used to measure the equipment's support for grid connection voltage stability.

[0119] First, the phase difference is obtained by calculating the difference between the fundamental phase angle of the inverter output voltage and the fundamental phase of the grid voltage through subtraction. Next, using the power transmission principle of the power system, combined with the grid-connected inductance parameters of the charging pile, the DC bus voltage value, and the obtained amplitude modulation ratio, the equivalent fundamental voltage amplitude at the inverter output port is estimated. Finally, based on the reactive power calculation formula for parallel operation of voltage sources, the current reactive power compensation is calculated by combining the grid voltage amplitude, the inverter's equivalent voltage amplitude, and the phase difference. For example, assuming the phase difference is positive and small, and the inverter's equivalent voltage amplitude is higher than the grid voltage amplitude, the calculated reactive power compensation is +5kVar, indicating that the charging pile is sending inductive reactive power to the grid to support the voltage.

[0120] Step S104 generates a contribution by superimposing the phase of the first harmonic component and comparing the amplitude of the second harmonic component, including:

[0121] S1041. By superimposing the phase of the first harmonic component and comparing the amplitude of the second harmonic component, an initial contribution is generated, and the contribution is determined based on the initial contribution and the reactive power compensation.

[0122] Initial contribution refers to a single technical performance indicator that only considers the harmonic control dimension, representing the degree to which the harmonic compensation component emitted by the charging pile cancels or suppresses the inherent harmonic components of the power grid.

[0123] First, the waveform relationship between the first and second harmonic components is compared. If their superposition reduces the overall harmonic amplitude, it is considered a positive contribution and quantified as the initial contribution level; otherwise, it is considered a negative contribution. Next, the calculated reactive power compensation is taken into account and combined with the initial contribution level to arrive at a final contribution level that encompasses both harmonic mitigation and reactive power support. For example, assuming the initial contribution level is calculated to be 0.8, and the equipment provides 5kVar of reactive power compensation, the final contribution level is calculated by combining both.

[0124] This embodiment analyzes the reactive power compensation of the charging pile by deeply mining the fundamental control information in the switch state space matrix, and then integrates this reactive power contribution with the initial contribution of harmonic mitigation for evaluation. This overcomes the limitation of solely focusing on harmonic mitigation and achieves a comprehensive quantification of the dual auxiliary service value of harmonic elimination and voltage stabilization provided by the charging pile.

[0125] Optionally, step S1041, which generates an initial contribution by superimposing the phase of the first harmonic component and comparing the amplitude of the second harmonic component, and then determines the contribution based on the initial contribution and the reactive power compensation, may specifically include:

[0126] Calculate the phase difference and amplitude ratio between the first harmonic component and the second harmonic component, and when the phase difference is within a preset difference range and the amplitude ratio is within a preset compensation range, use the amplitude ratio as the initial contribution.

[0127] The preset difference range refers to the phase angle range that is determined to be effectively canceled out. The preset compensation range refers to the amplitude matching range that is determined to be effective and safe. The preset difference range is set according to different power grid management models, as shown in Table 2 below, and the preset compensation range is shown in Table 3 below.

[0128] Table 2: Preset Difference Interval Comparison Table

[0129]

[0130] Table 2 defines the required angular range of the phase difference between the first and second harmonic components under different power grid management modes. Ideally, the phase difference should be 180 degrees to achieve maximum cancellation, but a certain deviation is permissible in practical engineering.

[0131] Table 3: Preset Compensation Interval Comparison Table

[0132]

[0133] Table 3 defines the reasonable range within which the amplitude ratio of the first harmonic component to the second harmonic component should be controlled to avoid overcompensation or undercompensation.

[0134] First, extract the phase and amplitude of the first harmonic component and the second harmonic component, and calculate the phase difference between them. and amplitude ratio Next, refer to Table 2 based on the currently set governance mode, and refer to Table 3 based on the compensation strategy to obtain the corresponding preset difference range and preset compensation range. Finally, determine the calculated... and Do they both fall within their respective intervals? If the condition is met, it means that the current compensation is valid and compliant, and the amplitude ratio can be directly calculated. Assign the value as the initial contribution level If the condition is not met, further judgment is made: if the phase difference value... If the value is within the range of [0°, 30°] or [330°, 360°], it is considered harmful compensation, and the initial contribution is determined accordingly. If the initial contribution is -K, then it is considered invalid compensation. It is 0.

[0135] For example, suppose the system is set to precise harmonic cancellation mode and full compensation strategy. If the measured phase difference is 175° and the amplitude ratio is 0.95, then the compensation is deemed effective, and the initial contribution is... .

[0136] Calculate the deviation ratio between the reactive power compensation amount and the preset reactive power demand value, and use the preset reactive power weighting coefficient to weight the deviation ratio to obtain the reactive power contribution.

[0137] The preset reactive power demand value refers to the theoretical value of reactive power urgently needed by the power grid at the current moment in order to maintain the voltage at the grid connection point within the standard range. The deviation ratio refers to the degree of difference between the actual reactive power compensation and the reactive power demand value; the smaller the deviation, the more accurate the response. Reactive power contribution refers to the quantified reactive power service performance score. The preset reactive power demand values ​​are shown in Table 4 below:

[0138] Table 4: Preset Reactive Power Demand Values ​​Comparison Table

[0139]

[0140] Table 4 lists the graded requirements of the power grid for reactive power based on the degree of deviation between the measured voltage and the nominal voltage at the grid connection point.

[0141] First, refer to Table 4 based on the real-time voltage deviation at the grid connection point to obtain the current preset reactive power demand value. Next, using the formula Calculate the deviation ratio; assign full marks when the denominator is 0. Finally, obtain the preset reactive power weighting coefficient. =0.3, calculate reactive power contribution. For example, the measured voltage was 5% lower than expected. Referring to Table 4, the required value was found. kVar. Assuming the actual calculated reactive power compensation... kVar, then the deviation ratio is The reactive power contribution was calculated. .

[0142] The initial contribution and the reactive contribution are weighted and summed to obtain the contribution value.

[0143] The initial contribution obtained from the preceding steps With reactive power contribution Add them together to get the final contribution. For example, combining the aforementioned data, .

[0144] This embodiment achieves a refined and standardized evaluation of harmonic mitigation and reactive power support, ensuring that only high-quality anti-phase cancellation behavior is included in the initial contribution, effectively eliminating ineffective or harmful regulation. Simultaneously, it precisely quantifies the response accuracy of reactive power support, generating a comprehensive contribution index that balances power quality purification and voltage stability support.

[0145] Optionally, before inputting the contribution as a feedforward variable into the neural network model based on adaptive PID control in step S105, the method further includes:

[0146] Calculate the real-time phase difference between the first harmonic component and the second harmonic component.

[0147] Real-time phase difference refers to the difference in instantaneous phase angle between the first harmonic component and the second harmonic component at a specific frequency point at any sampling time.

[0148] First, the time-domain waveforms of the first and second harmonic components are analyzed using Hilbert transform or zero-crossing detection algorithm. Second, the analytic signals of the two components are calculated separately, and their instantaneous phase functions are extracted. and Then, through subtraction... The real-time phase difference sequence is obtained. For example, assuming that for the 5th harmonic component, at t=0.01s, the phase of the compensation wave emitted by the charging pile is detected to be 175 degrees, and the phase of the background wave of the power grid is 0 degrees, then the calculated real-time phase difference is 175 degrees.

[0149] A safe phase margin is generated by calculating the absolute difference between the real-time phase difference and the preset critical phase value.

[0150] The preset critical phase value refers to the dangerous phase boundary value that leads to parallel resonance or voltage instability, usually set in the range of 90 degrees or 270 degrees. The safety phase margin refers to the degree to which the current real-time phase difference deviates from these dangerous boundaries, used to quantify the safety of operation. The larger the value, the farther away from the resonance region, and the safer.

[0151] First, set the critical phase value. Secondly, calculate the real-time phase difference. The absolute difference between the current phase value and the most recent critical phase value is the safety phase margin. For example, assuming the real-time phase difference is 175 degrees, the distance from the 90-degree boundary is 85 degrees, and the distance from the 270-degree boundary is 95 degrees, then the minimum value of 85 degrees is taken as the safe phase margin.

[0152] The contribution is nonlinearly weighted and adjusted using the safety phase margin.

[0153] First, a penalty function based on the safety phase margin is constructed. The function outputs a coefficient close to 1 when the margin is greater than the safety threshold, but the coefficient rapidly decays to 0 when the margin is less than the threshold. Secondly, using the formula... The contribution obtained in the previous step S104 is adjusted. For example, if the safety phase margin is 85 degrees, the penalty coefficient is 1.0, and the contribution remains unchanged; if the safety phase margin is only 5 degrees, the penalty coefficient drops sharply to 0.1. In this case, even if the amplitude cancellation effect is good, the final contribution will be greatly reduced, thus penalizing this risky adjustment behavior.

[0154] This embodiment achieves automatic identification and punishment of high-risk adjustment behaviors, ensuring that only effective mitigation performed within a safe range far from the resonance critical region receives full rewards. This not only incentivizes harmonic suppression but also proactively avoids the risk of grid connection point resonance.

[0155] S105. The contribution is used as a feedforward variable input to the neural network model based on adaptive PID control. The preset power quality parameters are used as the objective function to calculate the service performance score. The service performance score represents the performance level of the charging pile operator in participating in the active filter auxiliary service.

[0156] Optionally, step S105, which uses the contribution as a feedforward variable to input into the neural network model based on adaptive PID control, and calculates the service performance score using preset power quality parameters as the objective function, may specifically include:

[0157] Figure 3 A flowchart illustrating a method for generating service performance scores according to an embodiment of this application is shown. Figure 3 As shown, the method includes:

[0158] S1051. Input the contribution as a feedforward variable into the neural network model based on adaptive PID control, and set the preset power quality parameters as the target reference trajectory.

[0159] Feedforward variables refer to external input signals directly input into the control system to compensate for disturbances or guide the system response in advance; in this case, they are the quantified contribution values. The target reference trajectory refers to the curve or set of setpoints showing the desired ideal state changing over time, serving as the standard that the control algorithm strives to approximate. Preset power quality parameters refer to the qualified limits or ideal values ​​of key indicators at the grid connection point, set according to national standards or power grid operation specifications. The preset power quality parameters are shown in Table 5 below.

[0160] Table 5: Preset Power Quality Parameter Comparison Table

[0161]

[0162] Table 5 lists the target control values ​​for total harmonic distortion and power factor at different voltage levels and grid connection points. These values ​​serve as the full score standard or zero-error benchmark for the neural network model.

[0163] First, the contribution value generated in step S104 is formatted as an input vector. Then, based on the actual voltage level connected to the charging station, the corresponding target reference value vector is extracted from Table 5. These two vectors are then connected to the input layer and reference layer of the adaptive PID neural network model, respectively. For example, assuming the total harmonic distortion calculated from the contribution value at the current moment is 0.06 and the power factor is 0.92, the input vector is [0.06, 0.92]; the corresponding target reference trajectory is set as a constant vector [0.04, 0.95].

[0164] S1052. Calculate the error between the feedforward variable and the target reference trajectory using the backpropagation algorithm, and obtain the convergence residual value and weight change rate by updating the weights of the neural network model using the error.

[0165] The convergence residual refers to the steady-state deviation between the model output and the target reference trajectory after a certain number of iterations of training or online adjustment. The weight change rate refers to the magnitude of the update of the connection weights in a neural network per unit time step.

[0166] First, the backpropagation algorithm is used to calculate the error function between the input feedforward variables and the set target. Then, based on the gradient descent principle, the partial derivatives of the error with respect to the network weights are calculated, and the weighting coefficients of the proportional, integral, and differential units are updated accordingly. Finally, during model execution, the error value after each iteration is recorded in real time. When the error change tends to level off, the absolute error at this point is recorded as the convergence residual value. Simultaneously, the Euclidean distance between the weight vectors of two adjacent iterations is calculated and divided by the time step to obtain the weight change rate.

[0167] For example, suppose that after 50 iterations, the total harmonic distortion of the model output is 0.042, and compared with the target of 0.04, the convergence residual is . During this process, the proportional gain in the PID parameters... The value can be rapidly increased from 0.5 to 0.8 by calculating the proportional gain. The average weight change rate is obtained by taking the ratio of the change to the total number of iterations.

[0168] S1053. The convergence residual value is weighted using a preset accuracy weight to obtain a first performance score, and the weight change rate is weighted using a preset speed weight to obtain a second performance score. The service performance score is obtained by weighted summing of the first performance score and the second performance score.

[0169] The first performance score refers to the accuracy dimension score calculated based on the steady-state error of the control; the smaller the residual, the higher the score. The second performance score refers to the response dimension score calculated based on the parameter adjustment rate; the faster the rate of change within a reasonable range, the higher the score.

[0170] First, define a service performance scoring function, which can be an inverse proportional function or an exponential decay function to map physical indicators to a score of 0-100. Second, for the convergent residual value, use the formula... Calculate the first performance score; for the rate of change of weight, use the formula... Calculate the second performance score. Finally, obtain the preset accuracy weight of 0.6 and the preset speed weight of 0.4, and sum them up. For example, assume the first performance score is calculated. Second performance rating Points. Final calculation of service performance score. point.

[0171] This embodiment quantifies the steady-state accuracy and dynamic response speed of equipment governance. This method of evaluating control by controlling avoids the one-sidedness of a single result-oriented approach and achieves a deep profile and accurate scoring of the comprehensive capabilities of operator equipment hardware and software.

[0172] S106. Based on the service performance score, determine the target price tier in the pre-defined correspondence between the performance score and the price tier, and generate a target dynamic reward value based on the service performance score and the target price tier.

[0173] The pre-defined correspondence between performance scores and price tiers refers to a pre-established lookup table or piecewise function rule that maps technical evaluation scores to economic incentive standards. The target dynamic reward value refers to the actual economic benefit ultimately settled with the charging station operator. The pre-defined correspondence between performance scores and price tiers is shown in Table 6 below:

[0174] Table 6: Preset Performance Ratings and Price Tiers Comparison Table

[0175]

[0176] Table 6 divides the service performance score from 0 to 100 into several grade intervals, and sets a corresponding basic reward unit price and adjustment coefficient for each interval. The higher the score, the higher the unit price, and the greater the gain of the adjustment coefficient, thus non-linearly amplifying the reward for high-level governance services.

[0177] First, read the service performance score output in step S105. Next, match this score with the performance score ranges in Table 6 to determine its level, thereby locking in the corresponding target price tier and adjustment coefficient. The final reward can be generated using interpolation or a linear mapping formula. The specific calculation formula can be: ,in The equivalent service capacity provided by the equipment during this period.

[0178] For example, suppose the service performance score calculated in the previous step is 89 points. Referring to Table 6, this score falls within the range of [80, 90), belonging to grade A, with a corresponding base unit price. Yuan / kVarh, adjustment coefficient Assume the cumulative governance capacity provided by the equipment during this period. kVarh. Then calculate the target dynamic reward value. Yuan. Ultimately, this amount will be directly deposited into the operator's account as a dynamic reward, completing the closed-loop transformation from technical contribution to economic benefit.

[0179] This embodiment realizes the marketization and refinement of ancillary service pricing, significantly widening the revenue gap between high-quality services and ordinary services, avoiding egalitarianism, and effectively guiding charging pile operators to continuously optimize equipment control strategies and proactively pursue higher governance performance to obtain excess revenue, thereby effectively promoting the macro-level improvement of power grid quality at the microeconomic level.

[0180] Figure 4 This is a schematic diagram illustrating a specific implementation of a dynamic reward system for participants in a charging pile service chain, as provided in this application embodiment. (Refer to...) Figure 4 The system may include:

[0181] The 410 acquisition module is used to synchronously acquire the total harmonic distortion of the charging station and the public power grid connection point, as well as the pulse width modulation switching frequency and duty cycle status of the inverter inside the charging pile power module.

[0182] The 420 generation module is used to generate a harmonic feature spectrum by performing a short-time Fourier transform on the total harmonic distortion, and to construct a switching state space matrix based on the pulse width modulation switching frequency and duty cycle.

[0183] The 430 extraction module is used to extract the first harmonic component and the second harmonic component based on the harmonic feature spectrum and the switch state space matrix using wavelet packet decomposition. The first harmonic component is the harmonic compensation component actively emitted by the charging pile inverter, and the second harmonic component is the inherent harmonic component of the public power grid.

[0184] The 420 generation module is also used to generate contribution by superimposing the phase of the first harmonic component and comparing the amplitude of the second harmonic component;

[0185] The 440 calculation module is used to input the contribution as a feedforward variable into a neural network model based on adaptive PID control, and to calculate the service performance score with preset power quality parameters as the objective function. The service performance score represents the performance level of the charging pile operator in participating in the active filter auxiliary service.

[0186] The 420 generation module is also used to determine the target price tier based on the service performance score and the pre-defined correspondence between the performance score and the price tier, and to generate a target dynamic reward value based on the service performance score and the target price tier.

[0187] The dynamic reward system for charging pile service chain participants in this application embodiment is used to implement the aforementioned dynamic reward method for charging pile service chain participants. Therefore, the specific implementation of the dynamic reward system for charging pile service chain participants can be found in the embodiment section of the dynamic reward method for charging pile service chain participants above. The specific implementation can be referred to the description of the corresponding embodiments, which will not be repeated here.

[0188] Figure 5 A schematic diagram of the hardware structure of an electronic device provided in one embodiment of this application is shown.

[0189] The electronic device may include a processor 510 and a memory 520 storing computer program instructions.

[0190] Specifically, the processor 510 may include a central processing unit (CPU), an application-specific integrated circuit (ASIC), or one or more integrated circuits that can be configured to implement the embodiments of this application.

[0191] Memory 520 may include mass storage for data or instructions. For example, and not limitingly, memory 520 may include a hard disk drive (HDD), floppy disk drive, flash memory, optical disk, magneto-optical disk, magnetic tape, or Universal Serial Bus (USB) drive, or a combination of two or more of these. Where appropriate, memory 520 may include removable or non-removable (or fixed) media. Where appropriate, memory 520 may be internal or external to the integrated gateway disaster recovery device. In a particular embodiment, memory 520 is non-volatile solid-state memory.

[0192] Memory may include read-only memory (ROM), random access memory (RAM), disk storage media devices, optical storage media devices, flash memory devices, and electrical, optical, or other physical / tangible memory storage devices. Therefore, typically, memory includes one or more tangible (non-transitory) computer-readable storage media (e.g., memory devices) encoded with software including computer-executable instructions, and when the software is executed (e.g., by one or more processors), it is operable to perform the operations described with reference to the method according to the first aspect of this disclosure.

[0193] The processor 510 reads and executes computer program instructions stored in the memory 520 to implement any of the dynamic reward methods for participants in the charging pile service chain in the above embodiments.

[0194] In one example, the electronic device may also include a communication interface 530 and a bus 540. Wherein, such as Figure 5 As shown, the processor 510, memory 520, and communication interface 530 are connected through bus 540 and complete communication with each other.

[0195] The communication interface 530 is mainly used to realize communication between various modules, devices, units and / or equipment in the embodiments of this application.

[0196] Bus 540 includes hardware, software, or both, that couples components of an online data traffic metering device together. For example, and not limitingly, the bus may include an Accelerated Graphics Port (AGP) or other graphics bus, an Enhanced Industry Standard Architecture (EISA) bus, a Front Side Bus (FSB), HyperTransport (HT) interconnect, an Industry Standard Architecture (ISA) bus, an Infinite Bandwidth Interconnect, a Low Pin Count (LPC) bus, a memory bus, a Microchannel Architecture (MCA) bus, a Peripheral Component Interconnect (PCI) bus, a PCI-Express (PCI-X) bus, a Serial Advanced Technology Attachment (SATA) bus, a Video Electronics Standards Association Local (VLB) bus, or other suitable buses, or combinations of two or more of these. Where appropriate, bus 540 may include one or more buses. Although specific buses are described and illustrated in embodiments of this application, any suitable bus or interconnect is contemplated herein.

[0197] The electronic device can execute the dynamic reward method for charging pile service chain participants in the embodiments of this application, thereby realizing the dynamic reward method for charging pile service chain participants described in conjunction with the accompanying drawings.

[0198] Furthermore, in conjunction with the dynamic reward method for charging pile service chain participants in the above embodiments, this application embodiment can provide a computer-readable storage medium for implementation. This computer-readable storage medium stores computer program instructions; when executed by a processor, these computer program instructions implement any of the dynamic reward methods for charging pile service chain participants in the above embodiments.

[0199] It should be clarified that this application is not limited to the specific configurations and processes described above and shown in the figures. For the sake of brevity, detailed descriptions of known methods are omitted here. In the above embodiments, several specific steps are described and shown as examples. However, the method process of this application is not limited to the specific steps described and shown. Those skilled in the art can make various changes, modifications, and additions, or change the order of steps, after understanding the spirit of this application.

[0200] The functional blocks shown in the above-described structural diagram can be implemented as hardware, software, firmware, or a combination thereof. When implemented in hardware, they can be, for example, electronic circuits, application-specific integrated circuits (ASICs), appropriate firmware, plug-ins, function cards, etc. When implemented in software, the elements of this application are programs or code segments used to perform the required tasks. Programs or code segments can be stored on a machine-readable medium or transmitted over a transmission medium or communication link via data signals carried on a carrier wave. "Machine-readable medium" can include any medium capable of storing or transmitting information. Examples of machine-readable media include electronic circuits, semiconductor memory devices, ROM, flash memory, erasable ROM (EROM), floppy disks, CD-ROMs, optical disks, hard disks, fiber optic media, radio frequency (RF) links, etc. Code segments can be downloaded via computer networks such as the Internet, intranets, etc.

[0201] It should also be noted that the exemplary embodiments mentioned in this application describe methods or systems based on a series of steps or apparatus. However, this application is not limited to the order of the above steps; that is, the steps can be performed in the order mentioned in the embodiments, or in a different order, or several steps can be performed simultaneously.

[0202] The aspects of this application have been described above with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It should be understood that each block in the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing apparatus to produce a machine such that these instructions, executable via the processor of the computer or other programmable data processing apparatus, enable the implementation of the functions / actions specified in one or more blocks of the flowchart illustrations and / or block diagrams. Such a processor can be, but is not limited to, a general-purpose processor, a special-purpose processor, a special application processor, or a field-programmable logic circuit. It is also understood that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can also be implemented by dedicated hardware performing the specified functions or actions, or can be implemented by a combination of dedicated hardware and computer instructions.

[0203] The above provides a detailed description of a dynamic reward method and system for participants in a charging pile service chain, as provided in this application. Specific examples have been used to illustrate the principles and implementation methods of this application. The descriptions of the embodiments above are merely for the purpose of helping to understand the method and its core ideas. It should be noted that those skilled in the art can make various improvements and modifications to this application without departing from its principles, and these improvements and modifications also fall within the protection scope of this application.

Claims

1. A dynamic reward method for participants in a charging pile service chain, characterized in that, The method includes: Simultaneously acquire the total harmonic distortion of the charging station and the public power grid connection point, as well as the pulse width modulation switching frequency and duty cycle status of the inverter inside the charging pile power module; By performing a short-time Fourier transform on the total harmonic distortion, a harmonic feature spectrum is generated, and a switch state space matrix is ​​constructed based on the pulse width modulation switching frequency and the duty cycle state. Based on the harmonic feature spectrum and the switch state space matrix, the first harmonic component and the second harmonic component are extracted by wavelet packet decomposition. The first harmonic component is the harmonic compensation component actively emitted by the charging pile inverter, and the second harmonic component is the inherent harmonic component of the public power grid. The contribution is generated by superimposing the phase of the first harmonic component and comparing the amplitude of the second harmonic component; The contribution is used as a feedforward variable input to a neural network model based on adaptive PID control. The service performance score is calculated using preset power quality parameters as the objective function. The service performance score represents the performance level of the charging pile operator in participating in the active filter auxiliary service. The power quality parameters are the target control values ​​of total harmonic distortion and power factor at different voltage level grid connection points. Based on the service performance score, a target price tier is determined in the preset correspondence between performance scores and price tiers, and a target dynamic reward value is generated based on the service performance score and the target price tier. The step of using the contribution as a feedforward variable to input into a neural network model based on adaptive PID control, and using preset power quality parameters as the objective function to calculate a service performance score, includes: The contribution is input as a feedforward variable into the neural network model based on adaptive PID control, and the preset power quality parameters are set as the target reference trajectory. The error between the feedforward variable and the target reference trajectory is calculated using the backpropagation algorithm, and the weights of the neural network model are updated using the error to obtain the convergence residual value and the weight change rate. A first performance score is obtained by weighting the convergence residual value using a preset accuracy weight, and a second performance score is obtained by weighting the weight change rate using a preset speed weight. The service performance score is obtained by weighted summing of the first performance score and the second performance score.

2. The method according to claim 1, characterized in that, The method further includes: The phase and amplitude of the voltage fundamental wave at the grid connection point of the public power grid are acquired synchronously, and the phase angle and amplitude modulation ratio of the inverter output voltage fundamental wave are extracted from the switch state space matrix. Calculate the phase difference between the phase and the phase angle, and calculate the reactive power compensation based on the phase difference, the amplitude modulation ratio, and the amplitude. The step of generating a contribution by superimposing the phase of the first harmonic component and comparing the amplitude of the second harmonic component includes: An initial contribution is generated by superimposing the phase of the first harmonic component and comparing the amplitude of the second harmonic component, and the contribution is determined based on the initial contribution and the reactive power compensation.

3. The method according to claim 2, characterized in that, The step of generating an initial contribution by superimposing the phase of the first harmonic component and the second harmonic component and comparing their amplitudes, and determining the contribution based on the initial contribution and the reactive power compensation amount, includes: Calculate the phase difference and amplitude ratio between the first harmonic component and the second harmonic component, and when the phase difference is within a preset difference range and the amplitude ratio is within a preset compensation range, use the amplitude ratio as the initial contribution. Calculate the deviation ratio between the reactive power compensation amount and the preset reactive power demand value, and use the preset reactive power weighting coefficient to weight the deviation ratio to obtain the reactive power contribution. The initial contribution and the reactive power contribution are weighted and summed to obtain the contribution value.

4. The method according to claim 3, characterized in that, Before inputting the contribution as a feedforward variable into the neural network model based on adaptive PID control, the method further includes: Calculate the real-time phase difference between the first harmonic component and the second harmonic component; A safety phase margin is generated by calculating the absolute difference between the real-time phase difference and the preset critical phase value. The contribution is adjusted nonlinearly using the safety phase margin.

5. The method according to claim 1, characterized in that, The step of constructing a switching state space matrix based on the pulse width modulation switching frequency and the duty cycle state includes: The number of pulse cycles per unit time is determined based on the pulse width modulation switching frequency, and the conduction time sequence within each number of pulse cycles is extracted according to the duty cycle state. The conduction time sequence is mapped to a discrete state sequence, and the discrete state sequence is continuously truncated using a preset sliding window to obtain multiple local state vectors. The multiple local state vectors are stacked in a temporal order to construct the switch state space matrix.

6. The method according to claim 1, characterized in that, The step of extracting the first harmonic component and the second harmonic component using wavelet packet decomposition based on the harmonic feature spectrum and the switching state space matrix includes: The spectral distribution features are extracted by performing a frequency domain transformation on the switch state space matrix. The harmonic feature spectrum is decomposed into sub-band sequences of multiple independent frequency bands, and the correlation coefficient between each sub-band sequence and the spectral distribution feature is calculated using covariance calculation. The sub-frequency band sequence with a correlation coefficient greater than a preset correlation threshold is determined as the first sub-frequency band, and the sub-frequency band sequence with a correlation coefficient less than or equal to the preset correlation threshold is determined as the second sub-frequency band; The first harmonic component and the second harmonic component are obtained by reconstructing the first sub-band and the second sub-band by performing inverse wavelet packet transform on them respectively.

7. A dynamic reward system for participants in a charging pile service chain, characterized in that, include: The acquisition module is used to synchronously acquire the total harmonic distortion of the charging station and the public power grid connection point, as well as the pulse width modulation switching frequency and duty cycle status of the inverter inside the charging pile power module. The generation module is used to generate a harmonic feature spectrum by performing a short-time Fourier transform on the total harmonic distortion, and to construct a switch state space matrix based on the pulse width modulation switching frequency and the duty cycle state. The extraction module is used to extract the first harmonic component and the second harmonic component based on the harmonic feature spectrum and the switch state space matrix using wavelet packet decomposition. The first harmonic component is the harmonic compensation component actively emitted by the charging pile inverter, and the second harmonic component is the inherent harmonic component of the public power grid. The generation module is also used to generate a contribution by superimposing the phase of the first harmonic component and comparing the amplitude of the second harmonic component; The calculation module is used to input the contribution as a feedforward variable into a neural network model based on adaptive PID control, and to calculate a service performance score using preset power quality parameters as the objective function. The service performance score represents the performance level of the charging pile operator in participating in the active filter auxiliary service. The power quality parameters are the target control values ​​of total harmonic distortion and power factor at different voltage level grid connection points. Specifically, the calculation module is used to input the contribution as a feedforward variable into the neural network model based on adaptive PID control, and to set the preset power quality parameters as the target reference trajectory; to calculate the error between the feedforward variable and the target reference trajectory using the backpropagation algorithm, and to update the weights of the neural network model using the error to obtain the convergence residual value and the weight change rate; A first performance score is obtained by weighting the convergence residual value using a preset accuracy weight, and a second performance score is obtained by weighting the weight change rate using a preset speed weight. The service performance score is obtained by weighted summing of the first performance score and the second performance score. The generation module is also used to determine a target price tier based on the service performance score and in a preset correspondence between performance scores and price tiers, and to generate a target dynamic reward value based on the service performance score and the target price tier.

8. An electronic device, characterized in that, include: Memory, used to store computer programs; A processor, configured to implement the dynamic reward method for participants in the charging pile service chain as described in any one of claims 1 to 6 when executing the computer program.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, enables the implementation of a dynamic reward method for participants in the charging pile service chain as described in any one of claims 1 to 6.

Images