Vehicle identification method based on electric vehicle charging pile data

Abstract

The invention discloses a vehicle identification method based on electric vehicle charging pile data, and belongs to the technical field of electric vehicles, and the method comprises the following steps: S1, obtaining historical electric vehicle charging data collected by a charging pile; s2, constructing a vehicle identification model based on the multi-stage electric vehicle charging data difference degree, performing electric vehicle type grouping on the historical electric vehicle charging data through the vehicle identification model, and establishing a grouping database; and S3, collecting charging data of the charged electric vehicle in real time through the charging pile, extracting the charging data under at least three charge states, inputting the charging data into a vehicle identification module to obtain a to-be-classified feature vector, and comparing the feature vector with data in the grouping database to identify the type of the electric vehicle. By adopting the vehicle identification method based on the electric vehicle charging pile data, vehicle type identification is performed through the charging data of the charging pile, and a foundation is laid for providing a customized charging scheme for each vehicle subsequently.

Description

Technical Field

[0001] The present invention relates to the technical field of electric vehicles, and particularly to a vehicle identification method based on electric vehicle charging pile data. Background Art

[0002] As an important part of new energy vehicles, electric vehicles are gradually increasing their market share globally. One difference between electric vehicles and traditional fuel vehicles is the battery charging problem, which promotes the growth of the demand for charging infrastructure, making the construction and operation of charging piles one of the key links in the development of the electric vehicle industry. The technology of charging piles is constantly progressing, including fast charging technology, wireless charging technology, and the research and development and application of high-power charging facilities, all of which are improving charging efficiency and user experience. Although significant progress has been made in the construction of charging infrastructure, there are still some challenges, such as the problem that charging piles cannot identify vehicle types. Charging piles are the bridge connecting the power grid and electric vehicles and are important infrastructure for promoting the utilization of renewable energy and driving the electrification transformation of urban transportation. However, in order to protect user privacy, some charging piles cannot obtain the vehicle identification number (VIN), thus unable to effectively estimate the battery state in a timely manner. This will not only affect the charging efficiency of the charging pile, but also unable to handle potential hidden dangers and faults in a timely manner, affecting the life and safety of the battery. Summary of the Invention

[0003] The purpose of the present invention is to provide a vehicle identification method based on electric vehicle charging pile data to solve the above technical problems.

[0004] To achieve the above purpose, the present invention provides a vehicle identification method based on electric vehicle charging pile data, and the specific steps are as follows:

[0005] Step S1: Obtain the historical electric vehicle charging data collected by the charging pile and process it;

[0006] Step S2: Construct a vehicle identification model based on the difference degree of multi-level electric vehicle charging data, group the historical electric vehicle charging data by vehicle type through the vehicle identification model, and establish a grouped database based on the characteristic values of the electric vehicle charging difference degree;

[0007] Step S3: Real-time collect the charging data of the electric vehicle being charged by the charging pile, extract the charging data under at least three charge states, and input it into the vehicle identification module to obtain the feature vector to be classified. Compare the feature vector with the data in the grouped database for electric vehicle type identification. When the difference between the feature vector to be classified and the characteristic value of the electric vehicle charging difference degree in the grouped database is within the set threshold, it is determined that the electric vehicle being charged is the vehicle type corresponding to the characteristic value of the electric vehicle charging difference degree in the database.

[0008] Preferably, in step S1, the specific process of preprocessing the historical electric vehicle charging data is as follows:

[0009] Step S11: Perform operations of removing abnormal data and filling missing data on the historical electric vehicle charging data;

[0010] Step S12: Extract the charging data under different charge states, including the charging data at 10%, 20%, 30%, 40%, 50%, 60%, 70%, 80%, 90%, and 99% charge states.

[0011] Preferably, in step S2, the multi-level electric vehicle charging data difference includes voltage difference, voltage change rate difference, current difference, and current change rate difference, and the comprehensive difference under the corresponding charge state is calculated according to the voltage difference, voltage change rate difference, current difference, and current change rate difference.

[0012] Preferably, the voltage difference calculation formula is as follows:

[0013]

[0014] Among them, D U is the voltage difference, U A and U B are the voltage values of group A and group B under the same charge state respectively, and max() is the maximum value function.

[0015] Preferably, the voltage change rate difference calculation formula is as follows:

[0016]

[0017] Among them, D ΔU is the voltage change rate difference, U ΔA and U ΔB are the voltage change rates of group A and group B under the same charge state respectively.

[0018] Preferably, the current difference calculation formula is as follows:

[0019]

[0020] Among them, D I is the current difference, I A and I B are the current values of group A and group B under the same charge state respectively.

[0021] Preferably, the current change rate difference calculation formula is as follows:

[0022]

[0023] Among them, D ΔI is the difference degree of the current change rate, I ΔA and I ΔB are respectively the current change rate of Group A and the current change rate of Group B under the same charge state.

[0024] Preferably, the comprehensive difference degree calculation formula is as follows:

[0025] D total = αD U + βD I + γD ΔU + δD ΔI ;

[0026] Among them, D total is the comprehensive difference degree, α, β, γ, and δ are weight coefficients, and α + β + γ + δ = 1.

[0027] Preferably, when the comprehensive difference degrees under at least three charge states in the charging data of two chargings are within the set threshold, the two chargings belong to the same group and correspond to the same vehicle type.

[0028] Therefore, the vehicle identification method based on electric vehicle charging pile data of the present invention has the following beneficial effects: By analyzing the difference degrees of voltage data, current data, voltage change rate, and current change rate parameters, when charging two vehicle charging piles, the vehicle can be identified according to its charging curve, and then a customized charging plan can be provided for each vehicle, which also enhances the user experience and lays a foundation for improving user satisfaction with personalized services.

[0029] Next, through the drawings and embodiments, the technical solutions of the present invention will be further described in detail. Description of the Drawings

[0030] Figure 1 is a flowchart of a vehicle identification method based on electric vehicle charging pile data of the present invention;

[0031] Figure 2 is a graph of real electric vehicle charging data of the present invention. Detailed Embodiments

[0032] In the description of the present invention, it should be noted that the orientation or positional relationship indicated by the terms "upper", "lower", "inner", "outer", etc. is based on the orientation or positional relationship shown in the drawings, or the orientation or positional relationship in which the product of the present invention is usually placed during use. It is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation. Therefore, it should not be construed as a limitation to the present invention. In the description of the present invention, it should also be noted that unless otherwise clearly specified and defined, the terms "set", "installed", "connected" should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be directly connected, or indirectly connected through an intermediate medium, and it can be the communication inside two elements. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to specific situations.

[0033] The following will describe the embodiments of the present invention in detail with reference to the accompanying drawings.

[0034] As Figure 1 shown, a vehicle identification method based on electric vehicle charging pile data is as follows:

[0035] Step S1: Obtain and process the historical electric vehicle charging data collected by the charging pile. The specific process of preprocessing the historical electric vehicle charging data is as follows:

[0036] Step S11: Perform operations of removing abnormal data and filling missing data on the historical electric vehicle charging data.

[0037] Step S12: Extract the charging data under different state of charge (SOC), including the charging data at 10%, 20%, 30%, 40%, 50%, 60%, 70%, 80%, 90% and 99% state of charge. Here, choosing the SOC as an integer multiple of 10% is on the one hand because it can reduce the calculation amount, and on the other hand because in the actual charging activities of charging piles, the starting SOC of electric vehicle charging is not fixed, and the cut-off SOC is also not fixed. Selecting the charging data under different state of charge can reflect the voltage and current characteristics at the corresponding moment. The reason for choosing 99% instead of 100% is that after analyzing the charging pile data, it is found that there are a certain number of charging processes that stop charging when the SOC is 99%, and the difference in voltage and current between 99% and 100% of the SOC is not significant.

[0038] Step S2: Construct a vehicle identification model based on the difference degree of multi-level electric vehicle charging data, group the historical electric vehicle charging data by the vehicle identification model, and establish a grouped database based on the characteristic values of the electric vehicle charging difference degree.

[0039] The multi-level electric vehicle charging data difference includes voltage difference, voltage change rate difference, current difference, and current change rate difference, and the comprehensive difference under the corresponding charge state is calculated based on the voltage difference, voltage change rate difference, current difference, and current change rate difference.

[0040] The formula for calculating the voltage difference is as follows:

[0041]

[0042] Where D U is the voltage difference, U A and U B are the voltage values of Group A and Group B under the same charge state respectively, and max() is the maximum value function.

[0043] The formula for calculating the voltage change rate difference is as follows:

[0044]

[0045] Where D ΔU is the voltage change rate difference, U ΔA and U ΔB are the voltage change rates of Group A and Group B under the same charge state respectively.

[0046] The formula for calculating the current difference is as follows:

[0047]

[0048] Where D I is the current difference, I A and I B are the current values of Group A and Group B under the same charge state respectively.

[0049] The formula for calculating the current change rate difference is as follows:

[0050]

[0051] Where D ΔI is the current change rate difference, I ΔA and I ΔB are the current change rates of Group A and Group B under the same charge state respectively.

[0052] The formula for calculating the comprehensive difference is as follows:

[0053] D total = αD U + βD I + γD ΔU + δD ΔI ;

[0054] Among them, D total is the comprehensive difference degree, α, β, γ, and δ are weight coefficients, and α + β + γ + δ = 1. The specific values of the weight coefficients in the comprehensive difference degree calculation formula need to be determined through experimental optimization. Because for different data sets, the electric vehicle charging data they store is different, the weights of different parameters are also different.

[0055] Calculate the comprehensive difference degree at each SOC for different charging processes. For one charging process, up to 10 comprehensive difference degrees at SOC of 10%, 20%, 30%, 40%, 50%, 60%, 70%, 80%, 90%, and 99% can be obtained. When the comprehensive difference degrees in at least three charge states in the charging data of two chargings are within the set threshold, the two chargings belong to the same group and correspond to the same vehicle type. Group the vehicles with similar charging characteristics together. At the same time, because the charging curves of electric vehicles of different brands are different, and even for vehicles of the same brand, different usage years will also result in different charging curves, so it can not only identify and group different brand models, but also group vehicles with different health states.

[0056] Step S3: Real-time collect the charging data of the electric vehicle being charged through the charging pile, extract the charging data in at least three charge states, and input it into the vehicle recognition module to obtain the feature vector to be classified. Compare the feature vector with the data in the grouping database for electric vehicle type recognition. When the difference between the feature vector to be classified and the electric vehicle charging difference degree eigenvalue in the grouping database is within the set threshold, it is determined that the electric vehicle being charged is the vehicle type corresponding to the electric vehicle charging difference degree eigenvalue in the database. When a new vehicle comes to this charging pile for charging, its charging data will be recorded in the database, and the database will be updated in real time. At the same time, the vehicles need to be regrouped regularly every once in a while to update the vehicles included in each group in a timely manner and gradually include more vehicle types.

[0057] After determining the vehicle type, the charged electric vehicle can be compared with the charging curves of the vehicles in the same group to analyze abnormal changes in the charging curves, the deviation degree of battery parameters, etc., and timely detect whether there are abnormal situations such as overcharging, over-discharging, and aging of the vehicle battery. It can also provide customized charging schemes for different types of vehicles according to the classification results of the vehicles and the charging data in the same group. For vehicles with a relatively high degree of battery aging, appropriately reduce the charging power and extend the charging time to protect the battery; for high-performance electric vehicles, optimize the charging strategy according to their battery characteristics to improve the charging efficiency. In addition, it can also provide personalized services such as reservation charging and off-peak charging according to the user's charging habits and needs to enhance the user experience.

[0058] Such as Figure 2As shown, in this embodiment, a dataset of 181 sets of real electric vehicle charging data is adopted. In Figure 2 Figures (a) and (b) show the original charging curves. It can be seen that the voltage and current curves are disorderly, and the charging characteristics of different vehicles are mixed together, making it difficult to distinguish. Figures (c) and (d) show the charging curves after voltage difference screening. By comparing the voltage differences of different charging processes at the same SOC and excluding the charging processes with a difference greater than the threshold, it can be seen that the disorderliness of the curves has decreased, and similar charging processes begin to cluster together. Figures (e) and (f) show the results after voltage change rate screening. By further excluding the charging processes with large voltage change rate differences, the curves become more regular and have higher similarity. Figures (g) and (h) show the charging curves after current difference screening. After adding the screening of current differences, the clustering of the curves is further improved, and the charging characteristics of different vehicle types begin to be clearly distinguished. Figures (i) and (j) show the results after current change rate screening. After four-level comprehensive difference screening, the curves become very regular, and similar charging processes clearly cluster together. The differences between different groups are obvious. As can be seen from Figures (i) and (j), the charging curves of the charging processes in the same group are closely clustered after screening, indicating that these vehicles have similar charging characteristics and belong to the same brand or model, realizing the matching of the charging curve and the vehicle type, which is convenient for subsequent identification of the type of electric vehicle being charged.

[0059] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit them. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that they can still modify or equivalently replace the technical solutions of the present invention, and these modifications or equivalent replacements cannot make the modified technical solutions deviate from the spirit and scope of the technical solutions of the present invention.

Claims

1. A vehicle recognition method based on electric vehicle charging pile data, characterized in that, The specific steps are as follows: Step S1: Obtain and process historical electric vehicle charging data collected by charging piles; Step S2: constructing a vehicle identification model based on multi-level electric vehicle charging data differences, grouping historical electric vehicle charging data by electric vehicle type using the vehicle identification model, and establishing a grouping database based on electric vehicle charging difference characteristic values; Step S3: Charging data of the electric vehicle being charged is collected in real time through the charging pile, charging data under at least three charge states is extracted, and the data is input into the vehicle identification module to obtain a feature vector to be classified. The feature vector is compared with the data in the grouping database to identify the type of electric vehicle. When the difference between the feature vector to be classified and the electric vehicle charging difference characteristic value in the grouping database is within a set threshold, the charged electric vehicle is determined to be the vehicle type corresponding to the electric vehicle charging difference characteristic value in the database.

2. The vehicle recognition method based on electric vehicle charging pile data according to claim 1, wherein: In step S1, the specific process of preprocessing historical electric vehicle charging data is as follows: Step S11: removing abnormal data and filling missing data from historical electric vehicle charging data; Step S12: extracting charging data at different states of charge, including charging data at states of charge of 10%, 20%, 30%, 40%, 50%, 60%, 70%, 80%, 90%, and 99%.

3. The vehicle recognition method based on electric vehicle charging pile data according to claim 1, characterized in that: In step S2, the multi-level electric vehicle charging data difference includes voltage difference, voltage change rate difference, current difference and current change rate difference, and the comprehensive difference under the corresponding charge state is calculated based on the voltage difference, voltage change rate difference, current difference and current change rate difference.

4. The vehicle identification method based on electric vehicle charging pile data according to claim 3, characterized in that: The voltage difference calculation formula is as follows: Among them, D U is the voltage difference degree, U A and U B are the voltage values of Group A and Group B under the same charge state respectively, and max() is the maximum value function.

5. The vehicle identification method based on electric vehicle charging pile data according to claim 4, characterized in that: The voltage change rate difference calculation formula is as follows: Among them, D ΔU is the voltage change rate difference, U ΔA and U ΔB are the voltage change rates of group A and group B under the same charge state respectively.

6. The vehicle identification method based on electric vehicle charging pile data according to claim 5, characterized in that: The current difference calculation formula is as follows: Among them, D I is the current difference degree, I A and I B are the current values of Group A and Group B under the same charge state respectively.

7. A vehicle recognition method based on electric vehicle charging pile data according to claim 6, characterized in that: The current change rate difference calculation formula is as follows: Among them, D ΔI is the difference degree of current change rate, I ΔA and I ΔB are the current change rates of Group A and Group B under the same charge state respectively.

8. The vehicle identification method based on electric vehicle charging pile data according to claim 7, wherein: The formula for calculating the comprehensive difference is as follows: D total = αD U + βD I + γD ΔU + δD ΔI ; Among them, D total is the comprehensive difference degree, α, β, γ and δ are weight coefficients, and α + β + γ + δ = 1.

9. A vehicle recognition method based on electric vehicle charging pile data according to claim 1, characterized in that: When the comprehensive difference in at least three charge states in the charging data of two charges is within a set threshold, the two charges belong to the same group and correspond to the same vehicle type.

Images