An evaluation device and a method for evaluating the current harmonic feed-in content of a charging station into a power grid

Abstract

The application discloses an evaluation device and method for evaluating current harmonic feed-in content of a power grid of a charging station, the evaluation device comprises a central controller CCU, a grid-side collector PQA, charging pile-side collectors Ax and charging piles Changx, the central controller CCU is the core of the evaluation device, the grid-side current collector PQA is installed at the output end of a transformer, and the charging pile-side collectors Ax are installed at the front end of each charging pile; the evaluation method establishes a functional relationship between each harmonic and the output power of the charging pile, evaluates the content of the charging station harmonic feed-in side of the distribution network, solves the quantitative evaluation problem of the current harmonic feed-in side of the power grid generated when the electric vehicle charging station operates, proposes an evaluation device and method for the current harmonic feed-in side of the power grid of the charging station, effectively evaluates the content of the charging station harmonic feed-in side of the distribution network, provides the user with a suggestion on whether it is necessary to install a filtering device, and guarantees the power supply quality of the power department.

Description

Technical Field

[0001] This invention relates to the field of electric vehicle charging station current harmonic estimation, and more specifically, to an evaluation device and method for assessing the content of charging station current harmonics fed into the power grid. Background Technology

[0002] Currently, "pure electric drive" electric vehicles are a top priority in my country's automotive development. As a crucial foundational support system for electric vehicles, the charging system primarily utilizes power electronics technology. As a non-linear electrical device, it generates harmonic pollution to the power distribution network. This is especially true in large charging stations where numerous charging devices operate simultaneously, causing harmonic currents that can impact the power grid and other electrical equipment.

[0003] Numerous studies have investigated the causes of harmonic generation from single electric vehicle charging piles and methods for harmonic suppression. Other studies have examined the impact of large-scale electric vehicle charging harmonics on the distribution network side, indicating that factors such as the charging scale of charging stations and the mixed charging methods of charging piles all affect the harmonics on the distribution network side. The main conclusions of these studies are: 1) The harmonics generated by a single charging pile in constant current or constant voltage charging mode are different, with harmonics increasing significantly during constant voltage charging; 2) When multiple charging piles are used for mixed charging, due to the time-varying nature of different charging piles, current harmonics will partially cancel each other out. As the number of charging piles gradually increases, the current harmonics on the distribution network side gradually decrease until the number of charging piles reaches a certain level, at which point the current harmonics on the distribution network side will gradually stabilize.

[0004] A Chinese patent document, CN110008644B, discloses a method for estimating multi-source harmonics in charging stations based on classification correction. This method involves a measuring device that automatically detects data such as the number of charging piles currently charging, the remaining capacity of the vehicle's battery, the charging method, the harmonics of the charging current and voltage of each charging pile, the charging stage, the charging time, and the total charging current and voltage harmonics of the charging station. It then classifies the harmonics based on their variation characteristics and automatically provides the variation law of multi-source harmonics in the charging station with the number, structure, and charging stage of the charging piles. Based on this, it provides a method for estimating multi-source harmonics in charging stations. However, this method does not consider the complexity of the charging station's operating conditions, making it difficult to achieve high accuracy in the estimation.

[0005] A method for harmonic modeling of electric vehicle charging stations based on neural networks, disclosed in Chinese patent literature (publication number CN109800520A), introduces an equivalent simulation model of the electric vehicle charging process based on the battery state of charge (SOC). It establishes a sample dataset, trains a neural network using the collected sample data, and tests and evaluates the trained RBF neural network. This method can evaluate at a certain point in time, but it fails to reflect the harmonic statistics of the charging station load change when the charging load changes randomly.

[0006] A Chinese patent document, "An Algorithm for Harmonic Current Detection of Electric Vehicle Charging Stations Based on Variational Bayesian Parameter Learning Method" (publication number CN109444505A), discloses an algorithm that uses a random number algorithm and a comparison module to determine whether to randomly connect the charging pile to the power distribution network system. Then, it performs ideal superposition calculation on the harmonic currents that meet the Gaussian normal distribution. This method is relatively accurate for calculating a single charging pile, but when multiple charging piles have harmonics that cancel each other out, their phases change over time and are difficult to estimate accurately. Therefore, it also has certain defects in superposition calculation.

[0007] A Chinese patent document, "A simulation system for simulating the harmonic characteristics of a charging station under different operating conditions," with publication number CN102496935A, simulates several parallel charging models to make their output characteristics match those of actual charging units. It studies the harmonic characteristics and their impact on the power grid, but does not perform statistical evaluation of the harmonics.

[0008] With the continuous increase in electric vehicles in my country, there are more and more charging stations with increasing capacity. The impact of current harmonics generated by charging stations on the power grid can no longer be ignored. How to reasonably and effectively assess the content of harmonics fed into the distribution network from charging stations, and then provide users with suggestions on whether it is necessary to install filtering devices based on these assessments, so as to ensure the power supply quality of the power sector. Summary of the Invention

[0009] The purpose of this invention is to scientifically and effectively solve the problem of quantitative assessment of current harmonics generated during the operation of electric vehicle charging stations and their feed into the power grid. This invention proposes an assessment device and method for assessing the feed of current harmonics from charging stations into the power grid. The assessment device of this invention acquires raw data of the electrical parameters of the charging piles by collecting electrical parameter signals from monitoring points, determines the charging status of the charging piles, establishes a functional relationship between each harmonic and the output power of the charging piles, and, based on this, comprehensively considers the assessment of the harmonic content fed into the power grid from multiple charging piles operating simultaneously under different charging modes.

[0010] To solve the above problems, the present invention adopts the following technical solution:

[0011] An evaluation device for assessing the harmonic content of charging station current fed into the power grid comprises a central controller (CCU), a grid-side data collector (PQA), a charging pile-side data collector (Ax), and charging piles (Changx) (x = 0, 1, 2, ..., n, representing the data collector or charging pile numbered x). The central controller (CCU) is the core of the evaluation device, employing an industrial control computer with a powerful processor and memory. It sends commands or receives signals from the grid-side data collector (PQA) and the charging pile-side data collector (Ax). The grid-side current collector (PQA) is installed at the transformer output end to collect the electrical parameter signals output from the transformer to the charging station and transmit the data to the central controller (CCU). The charging pile-side data collector (Ax) is installed at the front end of each charging pile to collect the electrical parameter signals of the charging pile and transmit the signals wirelessly to the central controller (CCU).

[0012] An assessment method for evaluating the content of harmonic current fed into the power grid from a charging station, based on an assessment device for evaluating the content of harmonic current fed into the power grid from a charging station, includes the following steps:

[0013] S1. The charging pile side data collector Ax collects the current and voltage passing through the charging pile x in real time, calculates the effective values ​​of the current and voltage, and compares them at regular intervals T.

[0014] S2. The central controller (CCU) sends a command to start sampling. After receiving the start sampling command, the grid-side collector (PQA) and the charging pile-side collector (Ax) record the current effective voltage value and the charging pile charging mode, and save the collected current data, the current effective voltage value, and the charging pile charging status in the local storage area.

[0015] S3. The central controller (CCU) sends a receive command to the grid-side data collector (PQA) and the charging pile-side data collector (Ax) to receive the data collected by each data collector.

[0016] S4. The central controller (CCU) establishes a fitting curve equation between the charging power of the charging pile and the generated harmonics, enabling the system to calculate the amount of harmonics generated by the charging pile through the charging power P and the charging method of the charging pile.

[0017] S5. A BP neural network is used for harmonic measurement. Each charging pile has three input data: charging pile current harmonics, charging power, and charging mode. Charging piles that are not in operation are filled with 0. If there are n charging piles in a charging station, the neural network has a total of 3n input data. The output of the neural network is the harmonic content.

[0018] S6. Based on the total number of electric vehicles registered in the area and the basic data measured by the evaluation device, the daily load curve of the charging station is predicted using the Monte Carlo method. Each harmonic is measured in the optimal operating mode with the minimum harmonics, and then in the most unfavorable condition with the maximum harmonics. Finally, each harmonic is measured in combination with other operating modes.

[0019] S7. Repeat step S6 to obtain the harmonic content for multiple time periods throughout the day, and display the harmonic situation for each time period in a table or graph.

[0020] Preferably, in step S1, when the effective value of the current continues to increase or stabilizes at a certain threshold I... T If the current is within the specified range, it is determined that the current charging pile is in constant current charging mode; if the effective value of the current decreases continuously for five consecutive cycles, it can be determined that it is in constant voltage charging mode. The charging pile side collector Ax saves the current charging mode Chang_St_x, where x in Chang_St_x represents the charging mode that the xth charging pile is currently operating in.

[0021] Preferably, in step S2, after receiving the start sampling command, the grid-side collector PQA and the charging pile-side collector Ax start collecting current signals at the next voltage zero crossing point. They sample 256 points per power frequency cycle, continuously collect for 10 cycles, and collect 256*10 current data points, as well as the current effective voltage value and the charging pile charging status.

[0022] Preferably, step S3 further includes step S31: repeating steps S2 and S3 multiple times, with the central controller (CCU) acquiring data from the charging pile at different charging power and different charging methods.

[0023] Preferably, in step S4, the central controller (CCU) uses the data from the charging pile to determine the coefficients of the fitting curve by taking different power levels and corresponding calculated harmonics under constant current charging and constant voltage charging modes, respectively, and obtains the curve fitting equation between charging power and harmonics.

[0024] Preferably, in step S5, the large amount of experimental data collected by the central controller (CCU) is used for BP neural network training. Through training, the weights of various components in the neural network can be determined. After training is completed, the parameters of the charging pile in operation are input to the neural network input node, and the grid-side harmonic content can be obtained through the neural network.

[0025] Preferably, step S6 further includes the following steps:

[0026] S61. Divide the daily load of the charging station into 96 points, with each point spaced 15 minutes apart.

[0027] S62. Let the average total load power at the first time point of a certain day be P1, the average total load power at the second time point be P2, and so on, until the average total load power at the 96th time point is P... 96 .

[0028] S63. If the rated charging power of the charging pile is known to be P, then at the first time point, each charging pile in the charging station needs to be put into operation at its rated power, where n = P1 / P rounded down. If each charging pile is charged at half its rated power, then 2n charging piles need to be put into operation.

[0029] S64. The charging stations are operated under the following four charging methods under the condition of known power demand P: 1) n charging stations are put into operation, with a charging power of P, and all are charged in constant current mode. The harmonic content Hi1 of each charging station is obtained according to the fitting curve, where i represents the charging station number; 2) n charging stations are put into operation, with a charging power of P, and all are charged in constant voltage mode. The harmonic content Hi2 of each charging station is obtained according to the fitting curve, where i represents the charging station number; 3) n charging stations are put into operation, n / 2 are charged in constant current mode, and n / 2 are charged in constant voltage mode. When n / 2 is not an integer, it is rounded down. The harmonic content Hi3 of each charging station is obtained according to the fitting curve, where i represents the charging station number; 4) 2n charging stations are put into operation, with a charging power of P / 2 5) 2n charging piles are put into operation, with a charging power of P / 2, and all charging piles are charged with constant current. The harmonic content Hi4 of each charging pile is obtained according to the fitting curve, where i represents the number of the charging pile; 6) 2n charging piles are put into operation, with n charging piles charged with constant current and n charging piles charged with constant voltage. The harmonic content Hi6 of each charging pile is obtained according to the fitting curve, where i represents the number of the charging pile.

[0030] S65. Input the harmonics, charging mode, and charging power of the operating charging pile into the input node of the BP neural network to obtain the harmonic content of the grid side for that time period, and then take the average value of these six working modes as the harmonic content of that time period.

[0031] Preferably, in step S7, if the period of time during which the cumulative harmonic threshold exceeds the specified threshold is relatively long, the harmonics of the charging station are determined to be substandard.

[0032] Therefore, the present invention has the following beneficial effects: (1) it proposes an effective evaluation method for the content of current harmonics fed into the power grid; (2) it reasonably and effectively evaluates the content of harmonics fed into the distribution network side of the charging station, provides users with suggestions on whether it is necessary to install filter devices, and ensures the power supply quality of the power sector. Attached Figure Description

[0033] Figure 1This is a schematic diagram of the device for evaluating the content of charging station current harmonics fed into the power grid according to the present invention.

[0034] Figure 2 This is a structural model diagram of the BP neural network of the present invention.

[0035] Figure 3A This is a graph showing the changes in charging load at charging stations during working days according to the present invention.

[0036] Figure 3B This is a diagram showing the charging load changes at charging stations during holidays, as per the present invention.

[0037] In the diagram: 1. Central Controller (CCU) 2. Network-side Data Collector (PQA) 3. Charging Pile-side Data Collector (Ax) 4. Charging Pile (Changx). Detailed Implementation

[0038] The present invention will be further described below with reference to the accompanying drawings and specific embodiments:

[0039] like Figure 1 As shown, the power grid connects to the charging station via transformer TF. The charging station comprises several branches, each with a number of charging piles. For example, branch one has m charging piles, and other branches each contain a number of charging piles. The charging station is composed of n charging piles. An evaluation device for assessing the harmonic content of charging station current fed into the power grid comprises a central controller unit (CCU) 1, a grid-side data collector (PQA2), a charging pile-side data collector (Ax3), and charging piles (Changx4) (x = 0, 1, 2, ..., n, representing the data collector or charging pile numbered x-th). The device is characterized by a central controller (CCU1) serving as the core of the evaluation system. It employs an industrial control computer with a powerful processor and memory, which sends commands or receives signals from the grid-side current collector PQA2 and the charging pile-side current collector Ax3. The grid-side current collector PQA is installed at the transformer output end to collect the electrical parameter signals output from the transformer to the charging station and send the data to the central controller CCU1. The charging pile-side current collector Ax3 is installed at the front end of each charging pile to collect the electrical parameter signals of the charging pile and send the data to the central controller CCU1 via wireless transmission technology.

[0040] An evaluation device for assessing the content of harmonic current fed into the power grid from a charging station includes the following steps in its evaluation method:

[0041] S1. The charging pile-side data collector Ax3 collects the current and voltage passing through the charging pile x in real time, calculates the effective values ​​of the current and voltage, and compares them at regular intervals T. When the effective value of the current continues to increase or stabilizes at a certain threshold I, the data collection is successful. TIf the current is within the specified range, it is determined that the current charging pile is in constant current charging mode; if the effective value of the current decreases continuously for five consecutive cycles, it can be determined that it is in constant voltage charging mode. The charging pile side collector Ax3 saves the current charging mode Chang_St_x, where x in Chang_St_x represents the charging mode that the xth charging pile is currently operating in.

[0042] S2. The central controller CCU1 sends a command to start sampling. After receiving the start sampling command, the grid-side collector PQA2 and the charging pile-side collector Ax3 record the current effective voltage value and the charging pile charging mode. Then, they start collecting current signals at the next voltage zero-crossing point, sampling 256 points per power frequency cycle, and continuously collecting for 10 cycles. The collected 256*10 current data points, along with the current effective voltage value and the charging pile charging status, are saved in the local storage area. Sampling 256 points per power frequency cycle is used for current Fourier transform to obtain the harmonic content of the current of this charging pile.

[0043] Here, only the instantaneous current signal is collected and transmitted to the central controller CCU1, while only the effective value of the voltage for that period is uploaded. This is mainly because a large amount of data cannot be transmitted to the central controller CCU1 in real time during data transmission. After the electrical parameters are collected, they need to be stored in the local memory and wait for instructions from the central controller CCU1 before they can be uploaded. In order to maintain a high sampling rate but limited by the memory capacity, this invention only uploads 256*10 data points of the current sampling signal to the central controller CCU1 for harmonic measurement. Since the voltage does not change suddenly during the charging process, and the charging pile's charging status does not change suddenly, the effective value of the voltage and the charging method are transmitted at the end of the data transmission.

[0044] S3. The central controller CCU1 sends a receive command to the network-side data collector PQA2 and the charging pile-side data collector Ax3 to receive the data collected by each data collector. Steps S2 and S3 are repeated multiple times, and the central controller CCU1 acquires data on different charging powers and charging methods of the charging piles.

[0045] S4. Throughout the charging cycle, the harmonic current of the charging pile will change with the charging power of the charger. Using the curve fitting method, the amplitude and phase of the harmonic current are regarded as functions of the charging power. A fourth-order polynomial model is used to measure the correspondence between each harmonic coefficient and the power.

[0046]

[0047]

[0048]

[0049] In the formula, i(t) represents the AC signal of the charging pile, which is composed of the superposition of various harmonic currents, and a m and b m Let K be the coefficients of each degree of the Fourier series expansion. m-a0 and K m-b0 , ..., K m-a4 and K m-b4 The coefficients of the fourth-order polynomial fitting curve model are given by equations (2) and (3). Obviously, if the coefficients of each order of the fitting curve can be determined, the charging power P0 is known, and a can be calculated at this power according to equations (2) and (3). m and b m This allows us to determine the quantities of each harmonic of the current.

[0050] The central controller CCU1, based on the instantaneous current data collected by the grid-side data acquisition unit PQA2, can obtain the harmonics and total distortion rate of the grid-side current through Fourier transform, denoted as H0_1_1, H0_1_3, H0_1_5, H0_1_7, H0_1_9, ..., H0_1_21 and H0_1_all, respectively. In H0_x_y, H0 represents the grid-side harmonics, the first number x represents the data from the x-th measurement, and the second number y represents the y-th harmonic. For example, H0_1_3 represents the third harmonic in the first grid-side measurement, and H0_x_all represents the total distortion rate in the x-th grid-side measurement. Let the data collected by the central controller CCU1 from n measurements be as follows:

[0051] H0_1_1, H0_1_3, H0_1_5, H0_1_7, H0_1_9,…, H0_1_21, H0_1_all,

[0052] H0_n_1, H0_2_3, H0_2_5, H0_2_7, H0_2_9,…, H0_2_21, H0_2_all,

[0053] ………………;

[0054] H0_n_1, H0_n_3, H0_n_5, H0_n_7, H0_n_9,…, H0_n_21, H0_n_all.

[0055] Similarly, the central controller CCU1 calculates the harmonics, total distortion rate, and power of the charging pile side current based on the data collected by the charging pile side data collector Ax3. Let the data collected by the central controller CCU1 from charging pile 1 for n times be as follows:

[0056] H1_1_1, H1_1_3, H1_1_5, H1_1_7,…, H1_1_21, H1_1_all, P1_1;

[0057] H1_n_1, H1_2_3, H1_2_5, H1_2_7,…, H1_2_21, H1_2_all, P1_2;

[0058] …………;

[0059] H1_n_1, H1_n_3, H1_n_5, H1_n_7,…, H1_n_21, H1_n_all, P1_n.

[0060] Where P1_x represents the corresponding charging power of the x-th charging pile being measured.

[0061] Obviously, let the data collected by the central controller CCU1n from charging pile m be:

[0062] Hm_1_1, Hm_1_3, Hm_1_5, Hm_1_7,…, Hm_1_21, Hm_1_all, Pm_1;

[0063] Hm_2_1, Hm_2_3, Hm_2_5, Hm_2_7,…, Hm_2_21, Hm_2_all, Pm_2;

[0064] …………;

[0065] Hm_n_1, Hm_n_3, Hm_n_5, Hm_n_7,…, Hm_n_21, Hm_n_all, Pm_n.

[0066] Where Pm_x corresponds to the power measured at the m-th charging pile in the x-th measurement.

[0067] Based on the data from each charging pile, the central controller CCU1 takes different power values ​​and corresponding harmonics under constant current charging and constant voltage charging modes, substitutes them into equations (2) and (3), and calculates K. m-a0 and K m-b0 , ..., K m-a4 and K m-b4 The coefficients of the fitting curve are determined, and finally the curve fitting equation of power and harmonics is obtained, so that the system can calculate the amount of harmonics generated by the charging pile by using the curve fitting equation as long as the charging power P and the charging method of the charging pile are known.

[0068] S5. For example Figure 2 As shown, a BP neural network consists of an input layer, hidden layers, and an output layer. It is often referred to as a three-layer perceptron. The three-layer structure has n input nodes (x1, x2, ..., xn). n m hidden nodes (y1, y2, ..., y m ) and i output nodes (o1, o2, ..., o iThe weight matrix from the input layer to the hidden layer is composed of V = (v1, v2, ..., v). m ) represents the column vector v j The weight vector corresponding to the j-th neuron in the hidden layer, and the weights from the hidden layer to the output layer are represented by W = (w1, w2, ..., wj). i ) represents the column vector w k It is the weight vector corresponding to the k-th neuron in the output layer.

[0069] This invention uses a BP neural network for harmonic measurement. Each charging pile has three input data points: charging pile current harmonics, charging power, and charging mode. For charging piles that are not in operation, these data points are padded with 0. If there are n charging piles in a charging station, the neural network has a total of 3n input data points, and the input is the measured harmonics. Suppose that at a certain moment, there are m charging piles operating. The third harmonic of the first charging pile is H1_1_3, the charging power is P1_1, and the charging method is Chang_St_1. The third harmonic of the second charging pile is H2_1_3, the charging power is P2_1, and the charging method is Chang_St_2. The third harmonic of the third charging pile is H3_1_3, the charging power is P3_1, and the charging method is Chang_St_3. Then the input data of the neural network are x1=H1_1_3, x2=P1_1, x3=Chang_St_1, x4=H2_1_3, x5=P2_1, x6=Chang_St_2, x7=H3_1_3, x8=P3_1, x9=Chang_St_3, ..., x 3m-2 =Hm_1_3,x 3m-1 =Pm_1,x 3m =Chang_St_m, ...,x 3n-2 =0, x 3n-1 =0, x 3n =0. Output a data point o = H_W_3, representing the 3rd harmonic output content from the network side. Similarly, BP neural networks are established for the 5th, 9th, ..., 21 harmonics, with each neural network acquiring the corresponding harmonic content.

[0070] The aforementioned measured data, such as the third harmonic measured at a certain moment as H0_1_3 on the grid side, corresponds to the data for m operating charging piles as H1_1_3, P1_1, Chang_St_1, H2_1_3, P2_1, Chang_St_2, ..., Hm_1_3, Pm_1, Chang_St_m. The large amount of experimental data collected by the central controller CCU1 is used for BP neural network training. Through training, the weights of various components within the neural network can be determined. After training, the parameters of the operating charging piles are input to the neural network input node, and the grid-side harmonics can then be measured through the neural network.

[0071] S6. Obtain the total number of registered electric vehicles in the region from the management department. Combined with the evaluation device of this invention, the charging time, charging method, charging power, number of charging stations within a certain period of time, etc. of a single electric vehicle can be obtained. The central controller CCU1 counts the electric vehicles in the station in different time periods during the evaluation period. Then, the Monte Carlo method is used to predict the daily load curve of the charging station. Each harmonic is measured in the optimal working mode with the minimum harmonics. Then, each harmonic is measured in the most unfavorable condition with the maximum harmonics. Then, each harmonic is measured in combination with other working modes. The harmonic values ​​and operating parameters under each condition are input into the neural network input node. The network-side harmonic value under the condition is obtained according to the neural network. Finally, the average value of the network-side harmonics under various conditions is obtained as the harmonic quantity for that period of time.

[0072] The specific method is as follows: Divide the daily load of the charging station into 96 points, with each point spaced 15 minutes apart. Let the average total load power at the first time point of a day be P1, the average total load power at the second time point be P2, and so on, until the average total load power at the 96th time point is P1. 96 If the rated charging power of each charging pile is known to be P, then at the first time point, each charging pile in the charging station needs to be put into operation at its rated power, where n = P1 / P (rounded down). Therefore, if each charging pile charges at half its rated power, then 2n charging piles need to be put into operation. Based on the previously stated functional relationship between harmonics and charging power, and the fact that harmonics differ depending on the charging method, this invention operates the charging station under the known power demand P using the following four charging methods:

[0073] (1) n units are put into operation, with a charging power of P, and all are charged in constant current mode. The harmonic content Hi1 of each charging pile is obtained according to the fitting curve, where i represents the number of the charging pile.

[0074] (2) n units are put into operation, with a charging power of P, and all are charged in constant voltage mode. The harmonic content Hi2 of each charging pile is obtained according to the fitting curve, where i represents the number of the charging pile.

[0075] (3) n units are put into operation, n / 2 units are charged with constant current, and n / 2 units are charged with constant voltage. When n / 2 is not an integer, it is rounded down. The harmonic content Hi3 of each charging pile is obtained according to the fitting curve, where i represents the number of the charging pile.

[0076] (4) 2n units are put into operation, with a charging power of P / 2, and all are charged in constant current mode. The harmonic content Hi4 of each charging pile is obtained according to the fitting curve, where i represents the number of the charging pile.

[0077] (5) 2n units are put into operation, with a charging power of P / 2, and all are charged in constant voltage mode. The harmonic content Hi5 of each charging pile is obtained according to the fitting curve, where i represents the number of the charging pile.

[0078] (6) 2n units are put into operation, n units are charged with constant current, and n units are charged with constant voltage. The harmonic content Hi6 of each charging pile is obtained according to the fitting curve, where i represents the number of the charging pile.

[0079] Under the six operating conditions mentioned above, the harmonics, charging mode, and charging power of the operating charging pile are input to the input node of the BP neural network to obtain the grid-side harmonic content. The average value of the grid-side harmonic content measured under the various operating modes is taken as the harmonic content for that time period.

[0080] S7. Repeat step S6 to calculate the harmonic content for 96 time periods. Display the harmonic content for each time period in a table or graph, such as... Figure 3A and Figure 3B As shown.

[0081] If the period of time during which the cumulative harmonics exceed the prescribed threshold is long within a day, the harmonics of the charging station are deemed to be substandard.

[0082] It should be understood that any changes or improvements made by those skilled in the art to the technical solutions of this invention, without departing from the scope of the claims of this invention, are covered within the patent protection scope of this invention.

Claims

1. An evaluation device for assessing the content of harmonic current fed into the power grid from a charging station, comprising a central controller CCU (1), a grid-side data collector PQA (2), a charging pile-side data collector Ax (3), and a charging pile Changx (4), characterized in that, The central controller CCU(1) serves as the core of the evaluation device. The grid-side current collector PQA is installed at the transformer output end, and the charging pile-side collector Ax(3) is installed at the front end of each charging pile. The central controller CCU(1) is used to send instructions or receive signals from the network-side collector PQA(2) and the charging pile-side collector Ax(3). Based on the data from the charging pile, the central controller CCU takes different power and corresponding calculated harmonics under constant current charging mode and constant voltage charging mode respectively, determines the coefficients of the fitting curve, and obtains the curve fitting equation of charging power and harmonics. The grid-side current acquisition unit PQA is used to acquire the electrical parameter signals output from the transformer to the charging station and send the data to the central controller CCU (1); The charging pile side collector Ax(3) is used to collect the electrical parameter signals of the charging pile and send the data to the central controller CCU(1) through wireless transmission technology.

2. An assessment method for evaluating the content of harmonic current fed into the power grid from a charging station, based on an assessment device for evaluating the content of harmonic current fed into the power grid from a charging station, comprising the following steps: S1. The charging pile side collector Ax(3) collects the current and voltage passing through the charging pile x in real time, calculates the effective values ​​of the current and voltage, and compares them at regular intervals T. S2. The central controller CCU (1) sends a command to start sampling. After receiving the start sampling command, the network-side collector PQA (2) and the charging pile-side collector Ax (3) record the current effective voltage value and the charging pile charging mode, and save the collected current data, the current effective voltage value and the charging pile charging status in the local storage area. S3. The central controller CCU(1) sends a receiving instruction to the network-side collector PQA(2) and the charging pile-side collector Ax(3) to receive the data collected by each collector; S4. The central controller CCU(1) establishes the fitting curve equation between the charging power of the charging pile and the generated harmonics, so that the system can realize the method of calculating the amount of harmonics generated by the charging pile through the charging power P and the charging method of the charging pile. S5. Harmonic measurement is performed using a BP neural network. Each charging pile has three input data points: charging pile current harmonics, charging power, and charging method. Charging piles that are not in operation are filled with 0. If there are n charging piles in a charging station, the neural network has a total of 3n input data points. The output of the neural network is the harmonic content. S6. Based on the total number of electric vehicles registered in the area and the basic data measured by the evaluation device, the daily load curve of the charging station is predicted using the Monte Carlo method. Each harmonic is measured in the optimal working mode with the minimum harmonics, and then each harmonic is measured in the most unfavorable condition with the maximum harmonics. Then, each harmonic is measured in combination with other working modes. S7. Repeat step S6 to obtain the harmonic content for multiple time periods throughout the day, and display the harmonic situation for each time period in a table or graph.

3. The evaluation method for assessing the content of harmonic current fed into the power grid from a charging station, as described in claim 2, is characterized in that... In step S1, when the effective value of the current continues to increase or stabilizes at a certain threshold I... T If the current is within the range, it is determined that the current charging pile is in constant current charging mode; if the effective value of the current decreases continuously for five consecutive cycles, it can be determined that it is in constant voltage charging mode. The charging pile side collector Ax(3) saves the current charging mode Chang_St_x, where x in Chang_St_x represents the charging mode that the xth charging pile is currently operating.

4. The evaluation method for assessing the content of harmonic current fed into the power grid from a charging station according to claim 2, characterized in that, In step S2, after receiving the start sampling command, the grid-side collector PQA (2) and the charging pile-side collector Ax (3) start to collect current signals at the next voltage zero crossing point. They sample 256 points per power frequency cycle, continuously collect for 10 cycles, and collect 256*10 current data, the current effective voltage value, and the charging pile charging status.

5. The evaluation method for assessing the content of harmonic current fed into the power grid from a charging station according to claim 2, characterized in that, The step S3 further includes step S31: repeating steps S2 and S3 multiple times, the central controller CCU(1) acquires data of the charging pile under different charging power and different charging methods.

6. The evaluation method for assessing the content of harmonic current fed into the power grid from a charging station according to claim 2, characterized in that, In step S4, the central controller CCU(1) uses the data from the charging pile to determine the coefficients of the fitting curve by taking different power and corresponding calculated harmonics under constant current charging mode and constant voltage charging mode, respectively, and obtains the curve fitting equation of charging power and harmonics.

7. The evaluation method for assessing the content of harmonic current fed into the power grid from a charging station according to claim 2, characterized in that, In step S5, the large amount of experimental data collected by the central controller CCU(1) is used for BP neural network training. Through training, the weights of each item in the neural network can be determined. After the training is completed, the parameters of the charging pile in operation are input to the neural network input node, and the grid-side harmonic content can be obtained through the neural network.

8. The evaluation method for assessing the content of harmonic current fed into the power grid from a charging station according to claim 2, characterized in that, Step S6 further includes the following steps: S61. Divide the daily load of the charging station into 96 points, with each point spaced 15 minutes apart; S62. Let the average total load power at the first time point of a certain day be P1, the average total load power at the second time point be P2, and so on, until the average total load power at the 96th time point is P... 96 ; S63. If the rated charging power of the charging pile is known to be P, then at the first time point, each charging pile of the charging station needs to be put into operation at the rated power, n = P1 / P rounded down. If each charging pile is charged at half the rated power, then 2n charging piles need to be put into operation. S64. The charging stations are operated under the following four charging methods under the known power demand P: 1) n charging stations are in operation, with a charging power of P, all using constant current charging. The harmonic content Hi1 of each charging station is determined based on the fitted curve, where i represents the charging station number; 2) n charging stations are in operation, with a charging power of P, all using constant voltage charging. The harmonic content Hi2 of each charging station is determined based on the fitted curve, where i represents the charging station number; 3) n charging stations are in operation, n / 2 charging stations use constant current charging, and n / 2 charging stations use constant voltage charging. When n / 2 is not an integer, it is rounded down. The harmonic content Hi2 of each charging station is determined based on the fitted curve. 4) 2n charging piles are put into operation, with a charging power of P / 2, and all are charged in constant current mode. The harmonic content Hi4 of each charging pile is obtained according to the fitting curve, where i represents the charging pile number; 5) 2n charging piles are put into operation, with a charging power of P / 2, and all are charged in constant voltage mode. The harmonic content Hi5 of each charging pile is obtained according to the fitting curve, where i represents the charging pile number; 6) 2n charging piles are put into operation, with n charging in constant current mode and n charging in constant voltage mode. The harmonic content Hi6 of each charging pile is obtained according to the fitting curve, where i represents the charging pile number. S65. Input the harmonics, charging mode, and charging power of the operating charging pile into the input node of the BP neural network, obtain the grid-side harmonic content for each mode, and then take the average value of the grid-side harmonic content of these six working modes as the harmonic content for that time period.

9. The evaluation method for assessing the content of harmonic current fed into the power grid from a charging station according to claim 2, characterized in that, In step S7, if the period of time during which the cumulative harmonic threshold exceeds the specified threshold is relatively long, the harmonics of the charging station are determined to be substandard.

Images