Electric vehicle charging optimization method and system based on synchronization coefficient

Abstract

The invention discloses an electric vehicle charging optimization method and system based on a synchronization coefficient, and belongs to the field of power systems. The electric vehicle charging optimization method based on the synchronization coefficient comprises the steps that data are collected and preprocessed, and an electric vehicle charging strategy is generated; the method comprises the following steps: classifying and parameterizing load units of a power distribution network, then calculating a synchronization coefficient and generating a four-dimensional simultaneous factor matrix, and then generating a load scene through a k-means clustering algorithm; and based on the generated load scene, combining topological parameters of the power distribution network and load flow calculation to evaluate the feasibility of an electric vehicle charging strategy. According to the method, typical load scenes are generated through clustering, influences of different strategies on the line load rate and the transformer utilization rate are evaluated in combination with load flow calculation, and finally dual optimization of efficient utilization of renewable energy sources and power grid safety is achieved.

Description

Technical Field

[0001] The present invention belongs to the field of power systems, and in particular relates to an electric vehicle charging optimization method and system based on a synchronization coefficient. Background Art

[0002] With the rapid adoption of electric vehicles (EVs), their charging behavior is increasingly impacting the power grid. The flexibility potential of EVs can be exploited to balance the temporal disparity between grid load and renewable energy generation. However, current research focuses primarily on the local impact of EV flexibility on the grid, lacking a holistic perspective on EV charging characteristics. In particular, increased synchronization in EV charging behavior can lead to concentrated grid load, placing additional pressure on the distribution network. Therefore, a method for optimizing EV charging based on a synchronization coefficient is proposed. Summary of the Invention

[0003] In view of the deficiencies in the prior art, the present invention aims to provide an electric vehicle charging optimization method and system based on a synchronization coefficient, which solves the problems in the prior art.

[0004] The purpose of the present invention can be achieved through the following technical solutions:

[0005] The electric vehicle charging optimization method based on synchronization coefficient includes the following steps:

[0006] Collect and pre-process data to generate electric vehicle charging strategies;

[0007] The load units of the distribution network are classified and parameterized, and then the synchronization coefficient is calculated to generate a four-dimensional simultaneity factor matrix. The load scenario is then generated using the k-means clustering algorithm.

[0008] Based on the generated load scenarios, the feasibility of the electric vehicle charging strategy is evaluated in combination with the distribution network topology parameters and power flow calculation.

[0009] Furthermore, the collected data includes:

[0010] Historical driving data: including daily mileage, parking time, parking location, and charging start and end times;

[0011] Grid data: including residual load curve, photovoltaic power generation forecast curve and heat pump load curve;

[0012] Vehicle parameters: including battery capacity, charging station rated power and charging efficiency;

[0013] Data preprocessing includes removing invalid records, standardizing timestamps, and filling missing data with linear interpolation or similar vehicle behavior patterns.

[0014] Furthermore, the electric vehicle charging strategy includes:

[0015] Uncontrolled charging strategy: The user connects the electric vehicle to the charging pile after arriving at the charging pile and fully charges it at the maximum charging power;

[0016] Delayed charging strategy: Calculate the time step required to charge the EV based on the EV’s charging demand, and then select the charging time step with the lowest remaining load based on the sorting;

[0017] Weighted charging strategy: adapts the charging curve to the inverted curve of the residual load, and uses weights to adjust the average charging power to adapt to the inverted residual load;

[0018] Smooth charging strategy: Smooth the remaining load, that is, bring the remaining load to the average value.

[0019] Furthermore, in the weighted charging strategy, the following is determined for each electric vehicle and each charging process:

[0020]

[0021] in, Indicates the average charging power; SOC Arrival Indicates the SOC state of the electric vehicle at arrival, E Battery Indicates the battery capacity considering the building load demand or battery charging, T Parking Indicates parking charging time;

[0022] Weight The expression is:

[0023]

[0024] Among them, max t Represents the peak value of a certain load in the entire time series t; represents the residual load, represents the average residual load;

[0025] Set the maximum power p max and minimum power p min , the charging power during the entire parking time is determined as follows:

[0026]

[0027] in, represents the charging power vector; p Inatatt Indicates the installed power, which refers to the rated maximum output power of the charging pile; p Install It is the minimum installed power of the charging pile to which the electric vehicle is connected.

[0028] Furthermore, in the smooth charging strategy, a linear optimization problem is set for each charging process of each electric vehicle with load transfer potential, and its objective function is:

[0029]

[0030] Furthermore, the distribution network load units are classified into: household load, commercial load, heat pump load, photovoltaic load;

[0031] Among them, for household and commercial loads, standard load curves are used for parameterization; for heat pump and photovoltaic loads, dynamic curves generated based on meteorological data are used for parameterization.

[0032] Furthermore, the synchronization coefficient calculation expression is as follows:

[0033]

[0034] Where g(n,t) represents the synchronization coefficient, n is the total number of units, and P is the power value of the unit;

[0035] The four-dimensional simultaneity factor matrix G is:

[0036] G=[g HH ,g CTS ,g HP ,g PV ]

[0037] Among them, g HH 、g CTS 、g HP and g PV They represent the synchronization coefficients of household load, commercial load, heat pump load and photovoltaic load respectively.

[0038] Furthermore, the objective function J of the load scenario generated by the k-means clustering algorithm is to minimize the intra-cluster square error:

[0039]

[0040] Where, k represents the preset number of clusters; S c represents the sample set of the cth cluster; g i represents the i-th sample; μ c represents the center vector of the c-th cluster; ||.|| represents the Euclidean distance operation.

[0041] Furthermore, the indicators used to evaluate the electric vehicle charging strategy include: transformer load factor, line overload ratio and number of voltage-exceeding-limit nodes.

[0042] The electric vehicle charging optimization system based on synchronization coefficient includes:

[0043] Strategy generation module: collects and preprocesses data to generate electric vehicle charging strategies;

[0044] Scenario generation module: Classifies and parameterizes the distribution network load units, calculates the synchronization coefficient and generates a four-dimensional simultaneity factor matrix, and then generates load scenarios using the k-means clustering algorithm;

[0045] And, the strategy evaluation module: Based on the generated load scenarios, combined with the distribution network topology parameters and power flow calculation, it evaluates the feasibility of the electric vehicle charging strategy.

[0046] A computer storage medium stores a readable program, which can execute the above-mentioned electric vehicle charging optimization method based on synchronization coefficient when the program is running.

[0047] An electronic device comprising: a processor, a memory, a communication interface, and a communication bus, wherein the processor, the memory, and the communication interface communicate with each other via the communication bus;

[0048] The memory is used to store at least one executable instruction, and the executable instruction enables the processor to execute operations corresponding to the above-mentioned electric vehicle charging optimization method based on synchronization coefficient.

[0049] A computer program product includes computer instructions, wherein the computer instructions instruct a computing device to execute operations corresponding to the above-mentioned electric vehicle charging optimization method based on synchronization coefficient.

[0050] Beneficial effects of the present invention:

[0051] 1. This invention introduces a synchronization coefficient to quantify the synergistic characteristics between electric vehicle charging behavior and other loads, and designs multiple charging strategies to better match the charging load with the grid load characteristics, thereby reducing the operating pressure of the grid.

[0052] 2. The four charging strategies designed in this invention (no control, delay, weighted, and smoothing) not only meet the user's charging needs, but also achieve peak shaving and valley filling, reduce voltage fluctuations, and lower the risk of equipment overload by adjusting the charging timing and power.

[0053] 3. The present invention uses a four-dimensional synchronization coefficient matrix and k-means clustering method to construct a typical load scenario, which can effectively represent the load characteristics of different time periods throughout the year, avoid repeated calculations under large-scale time series data, and improve evaluation efficiency.

[0054] 4. During the evaluation process, the present invention combines distribution network flow calculations, sets key operating indicators such as transformer load rate, line overload ratio, and voltage over-limit ratio, and quantifies the grid adaptability and risk level of the strategy in different scenarios.

[0055] 5. The method proposed in the present invention can be flexibly embedded in the energy management platform or charging management system of the power system, and has good engineering feasibility and promotion prospects. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0057] Figure 1 is a schematic diagram of load scenario generation of the present invention;

[0058] Figure 2 It is a flow chart of the electric vehicle charging optimization method of the present invention. DETAILED DESCRIPTION

[0059] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making any creative efforts shall fall within the scope of protection of the present invention.

[0060] Example 1

[0061] like Figure 2 As shown, the electric vehicle charging optimization method based on the synchronization coefficient includes the following steps:

[0062] S1, generates electric vehicle charging strategies based on the collected and pre-processed data;

[0063] 1) Data Type

[0064] The collected data includes: historical driving data, power grid data and vehicle parameters;

[0065] Historical driving data: including daily mileage, parking time (arrival and departure times), parking location (home / commercial charging station), and charging start and end times.

[0066] Grid data: including the target area's 15-minute resolution residual load curve, photovoltaic power generation forecast curve, and heat pump load curve.

[0067] Vehicle parameters: including battery capacity, charging station rated power and charging efficiency.

[0068] 2) Data preprocessing

[0069] Data preprocessing includes removing invalid records (such as charging interruptions and timestamp errors), standardizing timestamps to the UTC+0 time zone, and filling missing data with linear interpolation or similar vehicle behavior patterns; the final output is a cleaned dataset (CSV or database format).

[0070] 3) Electric vehicle charging strategy generation;

[0071] In this embodiment, distributed modeling is used to generate electric vehicle charging strategies to smooth excess load and maximize the use of excess energy from renewable energy sources. These strategies include uncontrolled charging, delayed charging, weighted charging, and smooth charging.

[0072] 3.1) Uncontrolled charging

[0073] With uncontrolled charging, it is assumed that when an electric vehicle reaches a suitable charging station, it will charge as quickly as possible, injecting as much energy as possible to provide maximum comfort. Therefore, upon reaching the charging station, the user will immediately connect the electric vehicle to the charging station and fully charge it at the maximum charging power.

[0074] 3.2) Delayed charging

[0075] In the delayed charging scenario, the expected residual load for the target year is first used as the flexibility demand. The strategy's philosophy is to charge EVs when the residual load is as low as possible (preferably negative). The required charging time steps (15 minutes each) are calculated based on the EV's charging demand. Then, based on the ranking, the charging time step with the lowest possible residual load is selected.

[0076] 3.3) Weighted Charging

[0077] Weighted charging is an inverted charging curve that adapts the charging curve to the residual load. This means that when the residual load is particularly low, more charging is performed, while when the residual load is high, less charging is performed. Therefore, the charging process is determined for each electric vehicle and each charging process as follows:

[0078]

[0079] in, Indicates the average charging power; SOC Arrival Indicates the SOC state of the electric vehicle at arrival, E Battery Indicates the battery capacity considering the building load demand or battery charging, T Parking Indicates parking charging time; SOC when arriving at the charging station Arrival Should be parked for charging time T Parking The voltage increases to 1 within 10 seconds to charge the full battery capacity of the electric vehicle.

[0080] Use the following weights The process of adjusting this average charging power to accommodate the reverse residual load: Weight The specific expression is as follows:

[0081]

[0082] Among them, max t Represents the peak value of a certain load in the entire time series t; represents the residual load, Indicates the average residual load.

[0083] In addition to the weights, set the maximum power p max and minimum power p min , the charging power during the electric vehicle parking time is determined by the following formula:

[0084]

[0085] in, represents the charging power vector; p Inatatt Indicates the installed power, which refers to the rated maximum output power of the charging pile; p Install It is the minimum installed power of the charging pile to which the electric vehicle is connected.

[0086] 3.4) Smooth charging

[0087] The main purpose of smooth charging is to smooth the residual load, that is, to bring the residual load to an average value. To this end, a linear optimization problem is set up and solved for each charging process of each electric vehicle with load transfer potential. The objective function is:

[0088]

[0089]

[0090] Among them, SOC Departure Indicates the battery status when leaving; p i represents the charging power at the i-th moment; η represents the efficiency coefficient, t Step Indicates the parking time interval, usually 15 minutes.

[0091] S2, classifies and parameterizes the load units of the distribution network, then calculates the synchronization coefficient and generates a four-dimensional simultaneity factor matrix, and then generates the load scenario through the k-means clustering algorithm;

[0092] like Figure 1 As shown in Figure 2, the process of generating a load scenario includes:

[0093] 1) Classification and parameterization of load units;

[0094] The distribution network load units are divided into the following four categories:

[0095] Home (HH): base load (including lighting and appliances);

[0096] Commercial (CTS): Commercial equipment load;

[0097] Heat pump (HP): heating / cooling load (depending on air temperature);

[0098] Photovoltaic (PV): Power generated (irradiance dependent).

[0099] The load units are then parameterized. For household and commercial loads, standard load curves (such as H0 / H1 type) are used for parameterization; for heat pump and photovoltaic loads, dynamic curves generated based on meteorological data are used for parameterization.

[0100] 2) Synchronization coefficient calculation

[0101] The synchronization factor plays multiple roles in power systems, including quantifying load synchronization, supporting distribution network planning, optimizing grid operation, assessing renewable energy absorption capacity, supporting power market design, and improving grid economics. The synchronization factor describes the probability that multiple devices or users will reach their maximum power demand at the same time. The calculation expression for the synchronization factor is as follows:

[0102]

[0103] Where g(n,t) represents the synchronization coefficient, n is the total number of units, and P is the power value of the unit. In this embodiment, the synchronization coefficient is used to determine the synchronization coefficient of each technology except electric vehicles and at each time point of the year.

[0104] For each type of power generation unit, g(n,t) is calculated at 8760 time points throughout the year with a resolution of 15 minutes, and a four-dimensional simultaneity factor matrix G is generated:

[0105] G=[g HH ,g CTS ,g HP ,g PV ] (10)

[0106] Among them, g HH 、g CTS 、g HP and g PV They represent the synchronization coefficients of household load, commercial load, heat pump load and photovoltaic load respectively.

[0107] 3) Construct a four-dimensional synchronization coefficient matrix

[0108] The four types of load synchronization coefficients corresponding to each time point are combined into a four-dimensional vector. A total of T four-dimensional vectors (T = 8760 or higher resolution) are obtained throughout the year, and finally a matrix G is formed, which is the four-dimensional synchronization coefficient matrix.

[0109] 4) k-means clustering algorithm generates load scenarios

[0110] The k-means clustering algorithm is an unsupervised learning method used to divide a dataset into k clusters. Its core goal is to minimize the sum of squared distances between samples within a cluster and the cluster center. It takes a four-dimensional synchronization coefficient matrix G as input and performs unsupervised clustering:

[0111] 4.1) Input data: standardized four-dimensional synchronization coefficient matrix Where T = 8760 (number of data points with 15-minute resolution throughout the year), each row corresponds to a synchronization coefficient combination at a time point [g HH ,g CTS ,g HP ,g PV ].

[0112] 4.2) Objective Function: The objective function of k-means is to minimize the within-cluster sum of squares (WCSS):

[0113]

[0114] Where k represents the preset number of clusters, which is determined by the elbow rule; S c represents the sample set of the cth cluster; g i represents the i-th sample; μ c represents the center vector of the c-th cluster; || .| | represents the Euclidean distance operation.

[0115] 4.3) Euclidean distance: For two time points, the Euclidean distance d(g i ,μ c )for:

[0116]

[0117] Among them, g i,m represents the m-th dimension eigenvalue of the ith sample (such as the household load simultaneity factor); μ c,m Represents the m-th dimension eigenvalue of the c-th cluster center.

[0118] 4.4) Cluster center update: In each iteration, the cluster center μ c Update to the mean of all samples in the cluster:

[0119]

[0120] Among them, S c represents the number of samples in the cth cluster; represents the center vector of the c-th cluster in the t+1th iteration;

[0121] c i =argmin c ||g i -μ c || 2 (14)

[0122] Among them, c i Represents the cluster category label described by the i samples.

[0123] 4.5) Algorithm Flow

[0124] Step 1, initialization: randomly select k samples as the initial cluster centers

[0125] Step 2, distribute samples: assign each sample g i Assign to the nearest cluster center:

[0126] Step 3, update cluster centers: Update the center of each cluster based on the allocation results:

[0127] Step 4, iterative termination: repeat steps 2-3 until one of the following conditions is met: the change in cluster center is less than the threshold ∈=10 -5 ; Reach the maximum number of iterations T max =300.

[0128] 4.6) Elbow Rule (Determining the Number of Clusters k)

[0129] The elbow rule calculates the objective function J(k) for different k values ​​and selects the k value that significantly slows down the rate of decrease of J(k):

[0130]

[0131] Where J(k) represents the objective function value when the number of clusters is k; ΔJ(k) represents the rate of change of the objective function. And when ΔJ(k) < α (usually α = 0.1), select k * =k as the optimal number of clusters.

[0132] K cluster centers are obtained, each representing a "load scenario." Each load scenario represents a typical state of simultaneous characteristics of various load types within a specific time period, facilitating subsequent simulation evaluation of electric vehicle charging strategies and grid adaptability analysis.

[0133] S3, based on the load scenario generated by S2, combined with the distribution network topology parameters and power flow calculation, when evaluating the feasibility of the charging strategy, it is necessary to use the typical load scenario as the basic load input, and dynamically superimpose the electric vehicle charging power according to the charging strategy, and finally analyze the power grid operation indicators through power flow calculation.

[0134] 1) Distribution network modeling

[0135] The node admittance matrix Y is the core matrix that describes the electrical connection relationship of the distribution network. Its elements are determined by the line admittance and transformer parameters. ii and Y ij is the element of matrix Y, and the specific construction method is as follows:

[0136]

[0137] Y ij =-y ij (17)

[0138] Among them, Y ii represents the self-admittance matrix of node i; Ω i represents the set of nodes directly connected to node i; Y ij represents the mutual admittance matrix between node i and node j; y ij represents the elements in the admittance matrix, that is, the admittance between nodes i and j. If there is a transformer between nodes i and j with a transformation ratio of a:1, the admittance matrix needs to be modified to:

[0139]

[0140] Among them, y trans Represents the transformer admittance.

[0141] 2) Dynamic power flow calculation;

[0142] The power flow equation describes the node power balance relationship, and its complex form is:

[0143] S=V·(Y·V) * (19)

[0144] Where S represents the node complex power vector; V represents the node voltage vector; Y represents the network node admittance matrix; (·) * Represents the conjugate operation.

[0145] Expanding the real and imaginary parts of the equation is as follows:

[0146]

[0147] Among them, P i and Q i Represent the real and imaginary parts of power respectively; N represents the total number of nodes; θij =θ i -θ j Represents the node voltage phase difference; G ij and B ij denote the real and imaginary parts of the admittance matrix respectively.

[0148] 3) Evaluation Metrics

[0149] In this embodiment, the indicators related to the impact of electric vehicles on the power grid include: transformer load rate, line overload ratio and the number of voltage-limit-exceeding nodes.

[0150] 3.1) Transformer Loading Rate measures the ratio of the actual load of the transformer to its rated capacity. The calculation formula is as follows:

[0151]

[0152] Among them, L Transformer Indicates transformer load rate (unit: %); P Load Indicates the current load power of the transformer (unit: kW or MW); P Rated Indicates the rated capacity of the transformer (unit: kVA or MVA).

[0153] 3.2) Line Overload Ratio measures whether the load on the grid lines exceeds their rated capacity and calculates the proportion of lines that exceed a certain load threshold. The formula is as follows:

[0154]

[0155] Among them, L Line,i represents the load rate of the i-th line (%); P Load,i represents the actual load power of the i-th line (kW or MW); P Rated,i Represents the rated power of the i-th line (kW or MW).

[0156] 3.3) Number of Voltage Violation Nodes is used to measure the number of nodes in the power grid where the voltage exceeds the allowable range (usually ±5% or ±10%).

[0157] ΔV i =V i -V nominal (twenty three)

[0158] If|V i -V nominal |>V threshold , then i is an out-of-limit node (24)

[0159]

[0160] Where, ΔV i Indicates the voltage deviation of the i-th node (V or %); V i represents the actual voltage of the i-th node (V or pu); V nominal Indicates the rated voltage (usually 1.0pu or 230V); V threshold Indicates the allowable voltage deviation range, N V-Over Indicates the number of voltage-exceeding-limit nodes; N Total Indicates the total number of nodes in the system; R V-Over Indicates the voltage over-limit ratio, which is used to measure the number of voltage over-limit nodes.

[0161] The specific steps for evaluating an electric vehicle charging strategy are:

[0162] Step 1: Build a distribution network model

[0163] By collecting the structural parameters of the target distribution network, including nodes, lines, and transformers, the node admittance matrix Y is constructed, providing the electrical topology foundation for power flow calculations. Transformer admittance and ratio are corrected to ensure calculation accuracy.

[0164] Step 2: Import load scenario

[0165] Based on the typical load scenarios generated in step S2 (derived from the four-dimensional simultaneity factor matrix and k-means clustering results), the power distribution characteristics of various loads (household, commercial, heat pump, photovoltaic) in each scenario are extracted as static load input.

[0166] Step 3: Load the charging strategy

[0167] The different charging strategies generated in S1 (such as no control strategy, delay strategy, weighted strategy, and smoothing strategy) are superimposed on the load scenario, and the charging power value of the electric vehicle at each time step is updated accordingly.

[0168] Step 4: Perform power flow calculation

[0169] Under the combined effect of load scenarios and charging strategies, the power flow calculation program (such as the Newton-Raphson algorithm) is called to solve the voltage, current and power distribution of each node in the system.

[0170] Step 5: Calculate the evaluation index

[0171] Based on the power flow calculation results, the following three key indicators are calculated to measure the impact of the charging strategy on grid operation:

[0172] Transformer load rate: the ratio of the load power to the rated capacity of each transformer;

[0173] Line overload ratio: the proportion of lines that exceed the rated load capacity;

[0174] Voltage out-of-limit node ratio: The percentage of nodes whose voltage values ​​exceed the set range (such as ±5%).

[0175] Step 6: Compare and analyze the results

[0176] By comparing the changing trends of the above indicators under different strategy and scenario combinations, we can identify charging solutions that are more friendly to the power grid, thereby providing a decision-making basis for strategy optimization and actual deployment.

[0177] Based on similar inventive concepts, an embodiment of the present invention further provides a computer storage medium storing a readable program, which can execute the above-mentioned electric vehicle charging optimization method based on synchronization coefficient when the program is run.

[0178] Based on similar inventive concepts, an embodiment of the present invention provides an electronic device, comprising: a processor, a memory, a communication interface, and a communication bus, wherein the processor, the memory, and the communication interface communicate with each other via the communication bus;

[0179] The memory is used to store at least one executable instruction, and the executable instruction enables the processor to execute operations corresponding to the above-mentioned electric vehicle charging optimization method based on synchronization coefficient.

[0180] Based on similar inventive concepts, an embodiment of the present invention further provides a computer program product, including computer instructions, which instruct a computing device to execute operations corresponding to the above-mentioned electric vehicle charging optimization method based on synchronization coefficient.

[0181] Example 2

[0182] Based on the electric vehicle charging optimization method based on synchronization coefficient proposed in Example 1, this embodiment proposes an electric vehicle charging optimization system based on synchronization coefficient, including:

[0183] Strategy generation module: collects and preprocesses data to generate electric vehicle charging strategies;

[0184] Scenario generation module: Classifies and parameterizes the distribution network load units, calculates the synchronization coefficient and generates a four-dimensional simultaneity factor matrix, and then generates load scenarios using the k-means clustering algorithm;

[0185] And, the strategy evaluation module: Based on the generated load scenarios, combined with the distribution network topology parameters and power flow calculation, it evaluates the feasibility of the electric vehicle charging strategy.

[0186] The method of the present invention can be implemented in hardware, firmware, or as software or computer code that can be stored in a recording medium (such as a CDROM, RAM, floppy disk, hard disk or magneto-optical disk), or as computer code that is originally stored in a remote recording medium or a non-temporary machine-readable medium downloaded over a network and will be stored in a local recording medium, so that the method described herein can be stored in such software processing on a recording medium using a general-purpose computer, a special-purpose processor or programmable or special-purpose hardware (such as an ASIC or FPGA). It will be understood that a computer, a processor, a microprocessor controller or programmable hardware includes a storage component (e.g., RAM, ROM, flash memory, etc.) that can store or receive software or computer code, and when the software or computer code is accessed and executed by a computer, a processor or hardware, the method described herein is implemented. In addition, when a general-purpose computer accesses the code for implementing the method shown here, the execution of the code converts the general-purpose computer into a special-purpose computer for executing the method shown here.

[0187] The basic principles, main features, and advantages of the present invention are shown and described above. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The above embodiments and descriptions are merely illustrative of the principles of the present invention. Various changes and modifications may be made to the present invention without departing from the spirit and scope of the present invention, and such changes and modifications fall within the scope of the invention as claimed.

Claims

1. An electric vehicle charging optimization method based on synchronization coefficient, characterized in that: The following steps are involved: Collect and pre-process data to generate electric vehicle charging strategies; The load units of the distribution network are classified and parameterized, and then the synchronization coefficient is calculated to generate a four-dimensional simultaneity factor matrix. The load scenario is then generated using the k-means clustering algorithm. Based on the generated load scenarios, the feasibility of the electric vehicle charging strategy is evaluated in combination with the distribution network topology parameters and power flow calculation.

2. The electric vehicle charging optimization method based on synchronization coefficient according to claim 1 is characterized in that: The collected data includes: Historical driving data: including daily mileage, parking time, parking location, and charging start and end times; Grid data: including residual load curve, photovoltaic power generation forecast curve and heat pump load curve; Vehicle parameters: including battery capacity, charging station rated power and charging efficiency; Data preprocessing includes removing invalid records, standardizing timestamps, and filling missing data with linear interpolation or similar vehicle behavior patterns.

3. The electric vehicle charging optimization method based on synchronization coefficient according to claim 1, characterized in that: The electric vehicle charging strategy includes: Uncontrolled charging strategy: The user connects the electric vehicle to the charging pile after arriving at the charging pile and fully charges it at the maximum charging power; Delayed charging strategy: Calculate the time step required to charge the EV based on the EV’s charging demand, and then select the charging time step with the lowest remaining load based on the sorting; Weighted charging strategy: adapts the charging curve to the inverted curve of the residual load, and uses weights to adjust the average charging power to adapt to the inverted residual load; Smooth charging strategy: Smooth the remaining load, that is, bring the remaining load to the average value.

4. The electric vehicle charging optimization method based on synchronization coefficient according to claim 3 is characterized in that: In the weighted charging strategy, the following is determined for each electric vehicle and each charging process: in, Indicates the average charging power; SOC Arrival Indicates the SOC state of the electric vehicle at arrival, E Battery Indicates the battery capacity considering the building load demand or battery charging, T Parking Indicates parking charging time; Weight The expression is: Among them, max t Represents the peak value of a certain load in the entire time series t; represents the residual load, represents the average residual load; Set the maximum power p max and minimum power p min , the charging power during the entire parking time is determined as follows: in, represents the charging power vector; p Inatatt Indicates the installed power, which refers to the rated maximum output power of the charging pile; p Install It is the minimum installed power of the charging pile to which the electric vehicle is connected.

5. The electric vehicle charging optimization method based on synchronization coefficient according to claim 4 is characterized in that: In the smooth charging strategy, a linear optimization problem is set for each charging process of each electric vehicle with load transfer potential, and its objective function is:

6. The electric vehicle charging optimization method based on synchronization coefficient according to claim 1, characterized in that: The distribution network load units are classified into: household load, commercial load, heat pump load, and photovoltaic load; Among them, for household and commercial loads, standard load curves are used for parameterization; for heat pump and photovoltaic loads, dynamic curves generated based on meteorological data are used for parameterization.

7. The electric vehicle charging optimization method based on synchronization coefficient according to claim 6 is characterized in that: The synchronization coefficient calculation expression is as follows: Where g(n,t) represents the synchronization coefficient, n is the total number of units, and P is the power value of the unit; The four-dimensional simultaneity factor matrix G is: G=[g HH ,g CTS ,g HP ,g PV ] Among them, g HH 、g CTS 、g HP and g PV They represent the synchronization coefficients of household load, commercial load, heat pump load and photovoltaic load respectively.

8. The electric vehicle charging optimization method based on synchronization coefficient according to claim 7 is characterized in that: The objective function J of the load scenario generated by the k-means clustering algorithm is to minimize the intra-cluster square error: Where k represents the preset number of clusters; S c represents the sample set of the cth cluster; g i represents the i-th sample; μ c represents the center vector of the c-th cluster; |||.||| represents the Euclidean distance operation.

9. The electric vehicle charging optimization method based on synchronization coefficient according to claim 1, characterized in that: The indicators used to evaluate electric vehicle charging strategies include transformer load factor, line overload ratio and number of voltage-exceeding-limit nodes.

10. Electric vehicle charging optimization system based on synchronization coefficient, characterized in that: include: Strategy generation module: collects and preprocesses data to generate electric vehicle charging strategies; Scenario generation module: Classifies and parameterizes the distribution network load units, calculates the synchronization coefficient and generates a four-dimensional simultaneity factor matrix, and then generates load scenarios using the k-means clustering algorithm; And, the strategy evaluation module: Based on the generated load scenarios, combined with the distribution network topology parameters and power flow calculation, it evaluates the feasibility of the electric vehicle charging strategy.

11. A computer storage medium storing a readable program, characterized in that: When the program is running, the electric vehicle charging optimization method based on synchronization coefficient described in any one of claims 1 to 9 can be executed.

12. An electronic device, characterized in that: include: A processor, a memory, a communication interface and a communication bus, wherein the processor, the memory and the communication interface communicate with each other via the communication bus; The memory is used to store at least one executable instruction, and the executable instruction enables the processor to perform operations corresponding to the electric vehicle charging optimization method based on synchronization coefficient according to any one of claims 1 to 9.

13. A computer program product comprising computer instructions, characterized in that The computer instructions instruct the computing device to execute operations corresponding to the electric vehicle charging optimization method based on synchronization coefficient as described in any one of claims 1 to 9.

Images