Charging pile metering method, system, device and storage medium

Abstract

This invention discloses a charging pile metering method, system, device, and storage medium. The key technical points are: calculating the acquired power signal based on the square root extended Kalman filter algorithm to obtain a state metering result; filtering the power signal based on the normalized least mean square filter algorithm to obtain a signal filtering result; calculating the filtered metering result based on the signal filtering result; and dynamically weighting and fusing the state metering result and the filtered metering result to obtain the metering output result. This invention achieves high-precision energy measurement and is suitable for high-power charging and bidirectional energy flow charging scenarios.

Description

Technical Field

[0001] This invention belongs to the field of electric vehicle charging technology, specifically relating to a charging pile metering method, system, equipment, and storage medium. Background Technology

[0002] With the rapid development of electric vehicle (EV) and vehicle-to-grid (V2G) technologies, high-power supercharging and bidirectional energy flow scenarios have placed higher demands on electricity metering.

[0003] Traditional charging pile metering methods are mainly based on fixed bandwidth filtering and static calibration technology, which are difficult to cope with high-frequency power transients (such as switching ripple above 20kHz) and complex harmonic environments in V2G systems during supercharging. Summary of the Invention

[0004] The purpose of this invention is to provide a charging pile metering method, system, device and storage medium to achieve high-precision energy measurement, which is suitable for high-power charging and bidirectional energy flow charging scenarios.

[0005] The first aspect of this invention provides a charging pile metering method, comprising:

[0006] The acquired power signal is calculated based on the square root extended Kalman filter algorithm to obtain the state measurement results;

[0007] The power signal is filtered using the normalized least mean square filtering algorithm to obtain the signal filtering result, and the filtered metering result is calculated based on the signal filtering result.

[0008] The state measurement results and the filtered measurement results are dynamically weighted and fused to obtain the measurement output results.

[0009] In some implementations, the calculation of the power signal based on the square root extended Kalman filter algorithm to obtain the state measurement result includes:

[0010] State prediction is performed on instantaneous power based on power signal and square root extended Kalman filter algorithm to obtain state estimation results;

[0011] Update the square root factor of the covariance matrix in the square root extended Kalman filter algorithm based on the state estimation results;

[0012] Update the filter gain of the square root extended Kalman filter algorithm based on the updated square root factor;

[0013] The state estimation results are obtained by updating the state estimation results based on the updated filter gain.

[0014] In some implementations, the equation for calculating the filter gain is constructed based on the square root factor, a preset observation matrix, and a preset measurement noise covariance matrix;

[0015] In the vehicle-to-everything (V2X) interactive system, the method for updating the measurement noise covariance matrix includes:

[0016] Calculate the power change rate based on the state estimation results;

[0017] The adjustment term is obtained by multiplying the power change rate and the preset adjustment coefficient, and the adjustment factor is obtained based on the adjustment term and the benchmark term;

[0018] Update the measurement noise covariance matrix based on the adjustment factor.

[0019] In some implementations, the filtering of the power signal based on the normalized least mean square filtering algorithm to obtain a signal filtering result, and the calculation of the filtered measurement result based on the signal filtering result, includes:

[0020] A phase-locked loop is constructed based on the power signal, and an ideal fundamental component is generated based on the phase-locked loop.

[0021] A window function is used to extract the noisy input vector from the power signal;

[0022] The error vector is calculated based on the ideal fundamental component and the noisy input vector.

[0023] Update the fractional gradient based on the error vector and the noisy input vector, and update the weight vector of the normalized minimum mean square filter algorithm based on the updated fractional gradient.

[0024] The updated normalized least mean square filtering algorithm is used to filter the power signal to obtain the signal filtering result.

[0025] In some implementations, the dynamic weighted fusion of the state measurement results and the filtered measurement results to obtain the measurement output results includes:

[0026] The first dynamic weight is generated based on signal-to-noise ratio estimation and power differentiation;

[0027] The second dynamic weight is determined based on the first dynamic weight;

[0028] The state measurement result and the filtered measurement result are weighted and summed according to the first dynamic weight and the second dynamic weight to obtain the measurement output result.

[0029] In some implementations, generating the first dynamic weight based on signal-to-noise ratio estimation and power differentiation includes:

[0030] The signal-to-noise ratio (SNR) is estimated using the Welch periodogram method to obtain the current SNR. The ratio of the current SNR to a preset benchmark value is then calculated to obtain the first influence term.

[0031] The power change rate is obtained based on the power differential calculation, and the ratio of the hyperbolic tangent function of the power change rate to the rated power is calculated to obtain the second influence term;

[0032] A first dynamic weight is generated based on the first and second influence terms.

[0033] In some embodiments, the method for acquiring the power signal includes:

[0034] Acquire the original signal from synchronous sampling;

[0035] The original signal is subjected to anti-aliasing filtering to obtain the power signal.

[0036] A second aspect of the present invention provides a charging pile metering system, comprising:

[0037] The Kalman filter module is used to calculate the acquired power signal based on the square root extended Kalman filter algorithm to obtain the state measurement results;

[0038] The minimum mean square filtering module is used to filter the power signal based on the normalized minimum mean square filtering algorithm to obtain the signal filtering result, and to calculate the filtered measurement result based on the signal filtering result.

[0039] The dynamic fusion module is used to dynamically weight and fuse the state measurement results and the filtered measurement results to obtain the measurement output results.

[0040] A third aspect of the present invention provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the method described above.

[0041] A fourth aspect of the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the steps of the method described above.

[0042] The technical solution provided by this invention has the following advantages and effects: Attached Figure Description

[0043] Figure 1 This is a flowchart illustrating the charging pile metering method provided by the present invention;

[0044] Figure 2 This is a structural block diagram of the charging pile provided by the present invention;

[0045] Figure 3This is a structural block diagram of the charging pile metering system provided by the present invention;

[0046] Figure 4 This is an internal structural diagram of a computer device provided in an embodiment of the present invention. Detailed Implementation

[0047] To facilitate understanding of the present invention, specific embodiments of the present invention will be described in more detail below with reference to the accompanying drawings.

[0048] Unless otherwise specified or defined, the terms "first," "second," etc., used in this document are for distinguishing names only and do not represent a specific number or order.

[0049] Unless otherwise stated or defined, the term “and / or” as used herein includes any and all combinations of one or more of the associated listed items.

[0050] It should be noted that in this article, "fixed to" or "connected to" can mean directly fixed to or connected to a component, or indirectly fixed to or connected to a component.

[0051] like Figure 1 As shown, this embodiment provides a charging pile metering method, including the following steps S1 to S3:

[0052] Step S1: Calculate the acquired power signal based on the square root extended Kalman filter algorithm to obtain the state measurement result.

[0053] In practical applications, the method of this application is mainly used in supercharging and vehicle-to-grid interaction scenarios. It requires first acquiring the power signal, and then calculating the state measurement result and signal filtering result based on the power signal. Specifically, the method for acquiring the power signal includes:

[0054] Acquire the original signal from synchronous sampling;

[0055] The original signal is subjected to anti-aliasing filtering to obtain the power signal.

[0056] In practical applications, the TIADS8588S ADC chip can be used for synchronous sampling, with a voltage / current channel isolation bandwidth of 500kHz, meeting the 20kHz sampling requirement for DC charging. The original signal is obtained through synchronous sampling, and the ADC chip incorporates a programmable anti-aliasing filter. This filter filters the original signal to obtain the power signal, with the cutoff frequency dynamically adjusted according to the power level (20Hz-500Hz). The anti-aliasing filter, along with its programmable cutoff frequency linked to the switching frequency, avoids the performance degradation of traditional fixed-bandwidth designs over a wide power range.

[0057] Specifically, the calculation of the power signal based on the square root extended Kalman filter algorithm to obtain the state measurement result includes:

[0058] State prediction is performed on instantaneous power based on power signal and square root extended Kalman filter algorithm to obtain state estimation results;

[0059] Update the square root factor of the covariance matrix in the square root extended Kalman filter algorithm based on the state estimation results;

[0060] Update the filter gain of the square root extended Kalman filter algorithm based on the updated square root factor;

[0061] The state estimation results are obtained by updating the state estimation results based on the updated filter gain.

[0062] In practical applications, a state equation based on the Kalman filter algorithm is pre-constructed. The instantaneous power is then predicted using this state equation to obtain the state estimation result. The state equation can be expressed as:

[0063]

[0064] Where f(·) is the nonlinear state transition function, u k For control parameters, This is the state measurement result for the (k-1)th sampling point. This represents the state estimation result for the k-th sampling point. The control parameters include: grid frequency and switch duty cycle. The state estimation result for the k-th sampling point is an instantaneous power state vector, which can be expressed as: This represents the instantaneous active power at the k-th sampling point. This represents the instantaneous reactive power at the k-th sampling point. This represents the phase difference between the voltage and current at the k-th sampling point. For example, in a Boost PFC circuit, the power signal includes the voltage signal, current signal, and the switching transistor's percentage. The formula for calculating instantaneous active power is:

[0065]

[0066] Among them, f g Let N represent the grid frequency, V[n] represent the pulse count, v[n] represent the voltage value at the nth sampling point, ν[n] represent the current value at the nth sampling point, and D[n] represent the duty cycle of the switching transistor at the nth sampling point. The formulas for calculating instantaneous active power, instantaneous reactive power, and the duty cycle of the switching transistor can be adjusted according to the actual circuit conditions.

[0067] After obtaining the state estimation result, it is necessary to update the square root factor of the covariance matrix in the square root extended Kalman filter algorithm, then correct and update the filter gain to obtain the updated filter gain, and finally correct and update the state estimation result to obtain the state measurement result. Specifically, the update formula for the square root factor of the covariance matrix is:

[0068]

[0069] in, The Jacobian matrix of the state transition function is in The value at the specified position is used by `cholupdate()`, which is a rank-1 Cholesky update operation. This avoids the problem of non-positive definite matrices caused by truncation errors in traditional EKF.

[0070] After updating the square root factor of the covariance matrix, the filter gain of the square root extended Kalman filter algorithm is updated using the updated square root factor. Specifically, the calculation equation of the filter gain is constructed based on the square root factor, the preset observation matrix, and the preset measurement noise covariance matrix.

[0071] In the vehicle-to-everything (V2X) interactive system, the method for updating the measurement noise covariance matrix includes:

[0072] Calculate the power change rate based on the state estimation results;

[0073] The adjustment term is obtained by multiplying the power change rate and the preset adjustment coefficient, and the adjustment factor is obtained based on the adjustment term and the benchmark term;

[0074] Update the measurement noise covariance matrix based on the adjustment factor.

[0075] In practical applications, the equation for calculating the filter gain is:

[0076]

[0077] Among them, H k Let R be the observation matrix. k To measure the noise covariance matrix in a vehicle-to-grid (V2G) system scenario, the state estimation results include the calculation of instantaneous active power. The power change rate is obtained by calculating the rate of change of instantaneous active power. The update formula for the noise covariance matrix is ​​as follows:

[0078]

[0079] Wherein, ΔP k The rate of change of power, The base noise covariance matrix is ​​defined as follows: 0.5 is a preset adjustment coefficient, and 1 is a baseline term. In other embodiments, the preset adjustment coefficient and the baseline term can be adjusted according to actual conditions. By calculating the power change rate and updating the measurement noise covariance matrix based on the power change rate, the constraints of the measurement noise covariance matrix are automatically relaxed when there are sudden power changes.

[0080] Step S2: Filter the power signal based on the normalized least mean square filtering algorithm to obtain the signal filtering result, and calculate the filtered measurement result based on the signal filtering result.

[0081] In practical applications, a fractional-order gradient is introduced into the normalized least mean square (NMS) filtering algorithm to update the weight coefficients of the filtering algorithm, thereby suppressing specific spectral harmonics (such as 150Hz-2kHz). Specifically, the filtering of the power signal based on the NMS filtering algorithm to obtain the signal filtering result includes:

[0082] A phase-locked loop is constructed based on the power signal, and an ideal fundamental component is generated based on the phase-locked loop.

[0083] A window function is used to extract the noisy input vector from the power signal;

[0084] The error vector is calculated based on the ideal fundamental component and the noisy input vector.

[0085] Update the fractional gradient based on the error vector and the noisy input vector, and update the weight vector of the normalized minimum mean square filter algorithm based on the updated fractional gradient.

[0086] The updated normalized least mean square filtering algorithm is used to filter the power signal to obtain the signal filtering result.

[0087] In practical applications, phase tracking can be achieved through a closed loop using a phase detector, loop filter, and voltage-controlled oscillator. Then, the phase signal output from the phase-locked loop is used to separate the fundamental component via Parker transform to obtain an ideal fundamental component. This provides a distortion-free reference for subsequent filtering, avoiding filtering deviations caused by reference signal distortion. The expression for the noisy input vector is:

[0088] u k [i] = v[ki]·rect(i / M)

[0089] Among them, u kThe input vector is noisy, rect() is a rectangular window function, and M is the memory length, typically 16. The rectangular window function is used to extract the noisy input vector of the current signal and the noisy input vector of the voltage signal, preserving local signal features and providing real-time input for adaptive filtering. Then, the error vector is calculated based on the ideal fundamental component and the noisy input vector. The formula for calculating the error vector is:

[0090]

[0091] Among them, e k Let d be the error vector. k For the ideal fundamental component, This is the transpose of the weight vector for the normalized least mean square filter algorithm. Substituting the error vector and the noisy input vector into the fractional gradient update formula, we obtain the updated fractional gradient, which is:

[0092]

[0093] Where μ is the step size factor, used to control the convergence speed and steady-state error; δ is the regularization constant, used to prevent the denominator from being zero; α∈(0,1) is the fractional order, and the non-integer derivative of α=0.5 enhances the ability to capture intermittent harmonics. The updated weight vector of the normalized least mean square filter algorithm is as follows:

[0094]

[0095] in, The sgn() function preserves harmonic phase information for the Hadamard product. This application utilizes fractional gradient calculation to optimize convergence speed. Then, a normalized least mean square filtering algorithm with updated weight vectors is used to filter the power signal, yielding the signal filtering results, namely the current filtering result and the voltage filtering result. Finally, the power filtering result, i.e., the filtered metering result, is calculated based on the voltage and current filtering results.

[0096] Step S3: Dynamically weight and fuse the state measurement results and the filtered measurement results to obtain the measurement output results.

[0097] In practical applications, optimal filtering results can be achieved by dynamically adjusting the weights of state measurement results and filter measurement results. In this application, the filtering strategy is dynamically adjusted based on the signal-to-noise ratio and power flow direction, solving the performance degradation problem of traditional filtering algorithms during mode switching. Fractional gradient calculation further improves the harmonic suppression effect.

[0098] Specifically, the dynamic weighted fusion of the state measurement results and the filtered measurement results to obtain the measurement output results includes:

[0099] The first dynamic weight is generated based on signal-to-noise ratio estimation and power differentiation;

[0100] The second dynamic weight is determined based on the first dynamic weight;

[0101] The state measurement result and the filter measurement result are weighted and summed according to the first dynamic weight and the second dynamic weight to obtain the target output result.

[0102] In practical applications, the current signal-to-noise ratio (SNR) is calculated based on SNR estimation. A first dynamic weight is generated based on the current SNR and power derivative. A second dynamic weight is then determined based on the first dynamic weight, achieving a convex combination of the SREKF and NLMS two-channel outputs.

[0103] y k =λ k y SREKF,k +(1-λ k )y NLMS,k

[0104] Where, λ k As the first dynamic weight, y SREKF,k For the output of the square root extended Kalman filter algorithm, y NLMS,k As the output of the normalized least mean square filter algorithm, the first dynamic weight approaches 1 in the supercharging transient phase and approaches 0.3 in the V2G steady-state phase.

[0105] Specifically, the generation of the first dynamic weight based on signal-to-noise ratio estimation and power differentiation includes:

[0106] The signal-to-noise ratio (SNR) is estimated using the Welch periodogram method to obtain the current SNR. The ratio of the current SNR to a preset benchmark value is then calculated to obtain the first influence term.

[0107] The power change rate is obtained based on the power differential calculation, and the ratio of the hyperbolic tangent function of the power change rate to the rated power is calculated to obtain the second influence term;

[0108] A first dynamic weight is generated based on the first and second influence terms.

[0109] In practical applications, the Welch periodogram method is used to estimate the signal-to-noise ratio (SNR) by calculating the power spectral density (PSD) frame by frame, separating the signal and noise frequency bands, and achieving dynamic SNR tracking. After calculating the current SNR using the Welch periodogram method and obtaining the power change rate through power differentiation, the first dynamic weight is calculated using the following formula:

[0110]

[0111] Among them, SNR kP represents the current signal-to-noise ratio. rated For rated power, when SNR k >20dB and ΔP k >0.3P rated Time (supercharging transient), λ k >0.9 SREKF will be used preferentially. When SNR k When λ is <10dB and the power is stable (V2G steady state), k <0.4 to activate harmonic suppression of NLMS. Specifically, taking a 350kW supercharging station as an example:

[0112] Input conditions:

[0113] Voltage: 400V DC ±5% ripple (including 1kHz / 3kHz harmonics)

[0114] Current: 875A, rise time: 2ms

[0115] Temperature: 45℃ causes sensor drift of ±0.5%.

[0116] Processing procedure:

[0117] 1. SREKF tracks current jumps within 1ms, and the prediction error covariance maintains a condition number of less than 10^3.

[0118] 2. NLMS improves the 1kHz harmonic rejection ratio from -25dB to -42dB.

[0119] 3. The dynamic weight transitions from an initial value of 0.95 (transient) to 0.6 (steady-state).

[0120] To implement this charging pile metering method, a distributed processing architecture based on dual DSPs can be adopted. Specifically, the charging pile metering method is implemented in hardware using TI TMS320F28379D dual DSP chips. The master DSP is responsible for SREKF calculations, while the slave DSP is dedicated to NLMS filtering. Both synchronize data via a high-speed SPI interface (50MHz). The SNR estimation logic is deployed in an FPGA (Lattice MachXO3), and the weighting coefficients are updated in real time triggered by hardware interrupts. The ADC sampling rate (20kHz / 100kHz / 500kHz) is automatically selected according to the current power level, and aliasing-free sampling is achieved in conjunction with a programmable analog filter (MAX7420).

[0121] In one embodiment, specific optimizations are performed for high-frequency harmonic interference in vehicle-to-grid (V2G) scenarios, primarily including NLMS channel expansion and SREKF parameter adjustment. NLMS channel expansion involves adding parallel harmonic suppression branches to independently filter the 150Hz-1kHz (inverter switching harmonics) and 1kHz-5kHz (high-frequency noise) bands. SREKF parameter adjustment automatically increases the process noise covariance value when reverse power flow is detected, improving adaptability to inverter nonlinear characteristics. Additionally, the CORDIC algorithm is deployed in the PL section of the Xilinx Zynq to achieve hardware-accelerated calculation of fractional-order derivatives.

[0122] In one embodiment, for charging piles with power ratings below 100kW, a hybrid filter is constructed using an ST STM32H743ZI microcontroller, utilizing its double-precision FPU and ART accelerator to implement lightweight SREKF computation. The filter order of the normalized least mean square filtering algorithm is reduced to the 8th order, and the real-time computation load is reduced by pre-setting typical harmonic templates (such as the 5th and 7th harmonics). The optimal values ​​under different combinations of SNR and power change rate are pre-calculated and stored in Flash to reduce online computation overhead.

[0123] like Figure 2 and Figure 3 As shown, this embodiment of the invention also provides a charging pile metering system, including:

[0124] Kalman filter module 10 is used to calculate the acquired power signal based on the square root extended Kalman filter algorithm to obtain the state measurement result;

[0125] The minimum mean square filtering module 20 is used to filter the power signal based on the normalized minimum mean square filtering algorithm to obtain the signal filtering result, and to calculate the filtered measurement result based on the signal filtering result.

[0126] The dynamic fusion module 30 is used to dynamically weight and fuse the state measurement results and the filtered measurement results to obtain the measurement output results.

[0127] In practical applications, the charging pile metering system also includes: a power sensing unit for acquiring raw voltage and current signals and inputting them into a hybrid adaptive filter; and a billing and grid management module for generating billing information and managing grid interaction based on the filtered power data. The Kalman filter module 10, the least mean square filter module 20, and the dynamic fusion module 30 constitute the hybrid adaptive filter. The charging pile metering system is used in charging piles and can receive information transmitted from the communication module, user interface, and safety protection system within the charging pile. The charging pile metering system can also interact with the data storage and management system.

[0128] Each module of the aforementioned charging pile metering system can be implemented entirely or partially through software, hardware, or a combination thereof. These modules and units can be embedded in or independent of the processor in a computer device, or stored in the memory of a computer device as software, so that the processor can call and execute the corresponding operations of each module.

[0129] like Figure 4 As shown, an embodiment of the present invention discloses a computer device, including a memory and a processor, wherein the memory stores a computer program;

[0130] The computer device can be a server, and its internal structure diagram can be as follows: Figure 4 As shown, the computer device includes a processor, memory, network interface, and database connected via a system bus. The processor provides computing and control capabilities. The memory includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores the operating system, computer programs, and database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The network interface is used to communicate with external terminals via a network connection. When the computer program is executed by the processor, it implements the charging pile metering method described in the above embodiments.

[0131] Those skilled in the art will understand that Figure 4 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0132] This invention also discloses a computer-readable storage medium storing a computer program that causes a computer to execute the charging pile metering method described in the above embodiments.

[0133] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0134] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

Claims

1. A method of charging pile metering, characterized in that, include: The acquired power signal is calculated based on the square root extended Kalman filter algorithm to obtain the state measurement results; The power signals include voltage signals, current signals, and the duty cycle of the switching transistor; The power signal is filtered using the normalized least mean square filtering algorithm to obtain the signal filtering result, and the filtered metering result is calculated based on the signal filtering result. The state measurement results and the filtered measurement results are dynamically weighted and fused to obtain the measurement output results; The square root extended Kalman filter algorithm is used to calculate the power signal to obtain state measurement results, including: State prediction is performed on instantaneous power based on power signal and square root extended Kalman filter algorithm to obtain state estimation results; Update the square root factor of the covariance matrix in the square root extended Kalman filter algorithm based on the state estimation results; Update the filter gain of the square root extended Kalman filter algorithm based on the updated square root factor; The state estimation results are obtained by updating the state estimation results based on the updated filter gain; The calculation equation for the filter gain is constructed based on the square root factor, the preset observation matrix, and the preset measurement noise covariance matrix; In the vehicle-to-everything (V2X) interactive system, the method for updating the measurement noise covariance matrix includes: Calculate the power change rate based on the state estimation results; The adjustment term is obtained by multiplying the power change rate and the preset adjustment coefficient, and the adjustment factor is obtained based on the adjustment term and the benchmark term; Update the measurement noise covariance matrix according to the adjustment factor; The filtering of the power signal based on the normalized least mean square filtering algorithm to obtain the signal filtering result includes: A phase-locked loop is constructed based on the power signal, and an ideal fundamental component is generated based on the phase-locked loop. A window function is used to extract the noisy input vector from the power signal; The error vector is calculated based on the ideal fundamental component and the noisy input vector. Update the fractional gradient based on the error vector and the noisy input vector, and update the weight vector of the normalized minimum mean square filter algorithm based on the updated fractional gradient. The updated normalized least mean square filtering algorithm is used to filter the power signal to obtain the signal filtering result. The dynamic weighted fusion of the state measurement results and the filtered measurement results to obtain the measurement output results includes: The first dynamic weight is generated based on signal-to-noise ratio estimation and power differentiation; The second dynamic weight is determined based on the first dynamic weight; The state measurement result and the filtered measurement result are weighted and summed according to the first dynamic weight and the second dynamic weight to obtain the measurement output result. The formula for calculating the measurement output results is as follows: in, As the first dynamic weight, This is the output of the square root extended Kalman filter algorithm. For the output of the normalized least mean square filter algorithm, the first dynamic weight approaches 1 in the supercharging transient phase and approaches 0.3 in the V2G steady-state phase; The generation of the first dynamic weight based on signal-to-noise ratio estimation and power differentiation includes: The signal-to-noise ratio (SNR) is estimated using the Welch periodogram method to obtain the current SNR. The ratio of the current SNR to a preset benchmark value is then calculated to obtain the first influence term. The power change rate is obtained based on the power differential calculation, and the ratio of the hyperbolic tangent function of the power change rate to the rated power is calculated to obtain the second influence term; A first dynamic weight is generated based on the first and second influence terms.

2. The charging pile metering method as described in claim 1, characterized in that, The method for acquiring the power signal includes: Acquire the original signal from synchronous sampling; The original signal is subjected to anti-aliasing filtering to obtain the power signal.

3. A charging pile metering system for implementing the charging pile metering method according to claim 1 or 2, characterized in that, include: The Kalman filter module is used to calculate the acquired power signal based on the square root extended Kalman filter algorithm to obtain the state measurement results; The minimum mean square filtering module is used to filter the power signal based on the normalized minimum mean square filtering algorithm to obtain the signal filtering result, and to calculate the filtered measurement result based on the signal filtering result. The dynamic fusion module is used to dynamically weight and fuse the state measurement results and the filtered measurement results to obtain the measurement output results.

4. Computer equipment comprising a memory and a processor, said memory storing a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method of claim 1 or 2.

5. A computer readable storage medium having stored thereon a computer program, characterized in that When the computer program is executed by a processor, it implements the steps of the method of claim 1 or 2.

Images