CN-122022073-A - Charging station intelligent operation method and system based on charging user behavior characteristics

CN122022073ACN 122022073 ACN122022073 ACN 122022073ACN-122022073-A

Abstract

The invention relates to the field of electric vehicle charging facility operation and safety management, in particular to a charging station intelligent operation method and system based on charging user behavior characteristics. A charging station intelligent operation method based on charging user behavior characteristics comprises the following steps of S1, when a vehicle initiates a charging request, obtaining a vehicle identification code of the requesting vehicle, and analyzing and obtaining thermal management capability parameters of the vehicle based on the vehicle identification code, wherein the thermal management capability parameters comprise theoretical maximum continuous charging power and thermal management types. According to the invention, through merging vehicle hardware thermal management capability identification, user thermal behavior preference mining and multi-agent cooperative reinforcement learning, safety, efficiency and personalized experience cooperative optimization in a high-hot air risk scene of a charging network are realized, thermal safety management is converted into pre-prevention and global cooperation, and a complete closed loop from accurate perception, intelligent decision-making to personalized service is formed.

Inventors

ZHANG JUN
LU WEIMIN
LIU YANG
SHI XINLEI
Tang Baozheng
WANG SHUAI
WANG ZIYIN
RAN BIN
ZHAO BINGQI
SUN XIAOPENG
Sheng haigang

Assignees

国网(山东)电动汽车服务有限公司

Dates

Publication Date: 20260512
Application Date: 20260410

Claims (10)

1. Charging station intelligent operation method based on charging user behavior characteristics is characterized by comprising the following steps: The method comprises the following steps of S1, when a vehicle initiates a charging request, acquiring a vehicle identification code of the requesting vehicle, and analyzing and acquiring a thermal management capability parameter of the vehicle based on the vehicle identification code, wherein the thermal management capability parameter comprises theoretical maximum continuous charging power and a thermal management type; S2, acquiring historical charging behavior data of the request vehicle, constructing and training a hidden Markov model for deducing the thermal behavior preference of the request vehicle based on the historical charging behavior data, and decoding to obtain the type of the thermal behavior preference hidden in the request of the request vehicle; S3, constructing a multi-agent reinforcement learning environment taking a charging station in an area as an agent, wherein the state space of each agent comprises an operation state, a current environment temperature and a hot air risk situation vector formed by aggregation of thermal management capability parameters and hot air behavior preference types of vehicles to be charged in the area; And S4, aiming at a single request vehicle, fusing the thermal management capability parameters of the single request vehicle, the decoded thermal behavior preference type and the current environment temperature, calculating the thermal adaptation degree score of the charging station in the area, and generating and outputting a thermal adaptation charging guide sequence by combining a station-level guide strategy.
2. The charging station intelligent operation method based on the behavior characteristics of the charging user according to claim 1, wherein the analyzing and obtaining the thermal management capability parameters of the vehicle based on the vehicle identification code comprises the steps of obtaining battery type, battery rated capacity and thermal management configuration information of the vehicle according to the vehicle identification code through interaction with a manufacturer data platform, determining the thermal management type according to the thermal management configuration information, wherein the type comprises active liquid cooling, forced air cooling and natural cooling, and determining theoretical maximum continuous charging power of the vehicle under standard working conditions according to the battery type and the rated capacity and in combination with battery technical specifications.
3. The charging station intelligent operation method based on the behavior characteristics of the charging user according to claim 1, wherein the building and training of the hidden Markov model for deducing the thermal behavior preference of the request vehicle based on the historical charging behavior data comprises the environmental temperature when the historical charging event occurs, the ratio of the actual charging power to the theoretical maximum continuous charging power when the historical charging event occurs and the physical shielding attribute of the charging station of the historical charging event, defining the thermal behavior preference hidden state set of the user as Wherein As a total number of states, The representation is composed of A set of hidden states, each state representing a user's attitude and behavior pattern of charge and heat risks, including at least two of thermal risk desertification type, thermal risk evasion type and thermal sensitivity speed priority type, defining a user's observation sequence Wherein To observe the length of the sequence, representing the total number of user historical charging events, The representation is composed of A sequence of observations, each observation Comprises the first step of Features of the secondary historical charging record comprise ambient temperature during charging, the ratio of actual charging average power to theoretical maximum continuous charging power of a vehicle and shielding attribute identification of a charging station, wherein parameters of the hidden Markov model are as follows Wherein As a matrix of state transition probabilities, In order to observe the probability matrix, The initial state distribution is realized by adopting a forward-backward algorithm and a Baum-Welch algorithm, and the observation sequence is utilized For model parameters Training is carried out, and a user personalized thermal behavior preference model is obtained.
4. The charging station intelligent operation method based on the charging user behavior characteristics according to claim 1, wherein the decoding to obtain the thermal behavior preference type implicit in the request of the requesting vehicle comprises the following steps of using the user Secondary history charge observation sequence Calculating the most probable hidden state sequence by Viterbi algorithm as input, and taking the last state of the hidden state sequence as the current thermal behavior preference type of the user 。
5. The charging station intelligent operation method based on the charging user behavior characteristics according to claim 1, wherein the hot air risk situation vector aggregated by the thermal management capability parameters and the hot behavior preference types of the vehicles to be charged in the area comprises the following steps of Is an intelligent body Each dimension corresponds to statistics of a hot air risk related attribute, and the construction mode of the hot air risk situation vector is as follows: Wherein, the method comprises the steps of, Is expressed in the intelligent agent In the service range, the thermal management type is the proportion of the number of vehicles with natural cooling and forced air cooling to the total number of vehicles; Is expressed in the intelligent agent In the service range, the hot behavior preference type is the proportion of the number of users of the hot air risk desertification type to the total number of users; Representing an agent Average ambient temperature over the service range.
6. The charging station intelligent operation method based on the behavior characteristics of the charging user according to claim 1, wherein each intelligent agent learns to generate a station-level guiding strategy through collaborative training, the operation comprises guiding strategies directly related to hot air risk regulation, the operation comprises adjusting recommendation priority of charging piles, generating and sending instruction packets containing delay charging advice, transfer guiding information and excitation compensation to a user group, controlling the charging piles to actively cool and start and adjust operation power, and the station-level guiding strategy comprises adjusting recommendation priority of the charging piles generated based on hot air risk regulation, the instruction packets sent to the user group and controlling the charging piles.
7. The intelligent operation method of charging station based on behavior characteristics of charging user according to claim 1, wherein the construction of multi-agent reinforcement learning environment using charging stations in areas as agents comprises the steps of providing the areas as configurable operation management units, and each agent in the multi-agent reinforcement learning environment Is a reward function of (2) The design is as follows: Wherein, the method comprises the steps of, Is an intelligent body Earnings within the decision period; is an intelligent body The average utilization rate of the charging piles; 、 Respectively are intelligent agents 、 Is a load factor of (2); is an intelligent body Is a collection of adjacent agents; is an intelligent body The high-risk charging session is defined as a charging request of which the vehicle thermal management type is natural cooling and the user thermal behavior preference is hot air risk desertification and the actual request charging power is higher than the theoretical maximum continuous charging power by a certain proportion; is a preset positive weight coefficient.
8. The charging station intelligent operation method based on the charging user behavior characteristics according to claim 1, wherein the calculating the thermal fitness score of the charging station in the area by fusing the thermal management capability parameter of the single request vehicle, the decoded thermal behavior preference type and the current environment temperature for the single request vehicle comprises the following steps of The calculation formula of (2) is as follows: Wherein, the method comprises the steps of, As an efficiency factor, with the user Go to charging station Is used for estimating driving time and charging station Is inversely related to the current queuing time; is an economic factor, and is connected with a charging station Is inversely related to the service charge of (a); the thermal matching factor is calculated by a thermal matching function The calculation formula is calculated as follows: Wherein, the method comprises the steps of, Representing a user The thermal management capability score of the vehicle, Representing a user Is a hot performance preference score for (a), Indicating the temperature of the environment and, Indicating a charging station Is provided with a pile tip heat dissipation capacity score, Indicating a charging station Current load of (2); function The design is that when the vehicle heat-dissipation capability is weak, the user preference is aggressive and the environment temperature is high, a high matching score is given to the charging station with strong pile end heat-dissipation capability and low load.
9. The charging station intelligent operation method based on the charging user behavior characteristics according to claim 1, wherein the generating and outputting of the thermally adaptive charging guidance sequence by combining the station-level guidance strategy comprises the steps of sorting all candidate charging stations in descending order according to the thermal fitness score, selecting the K-top charging stations to form the guidance sequence, and for each charging station option in the sequence, converting the station-level guidance strategy into user-perceivable guidance information and outputting the guidance information together with basic information of the charging stations when the station-level guidance strategy is attached.
10. Charging station intelligent operation system based on charging user behavior characteristics, for implementing the charging station intelligent operation method based on charging user behavior characteristics according to any one of claims 1 to 9, characterized by comprising: The vehicle heat portrait construction module is used for acquiring a vehicle identification code of a requesting vehicle when the vehicle initiates a charging request, and analyzing and acquiring a thermal management capability parameter of the vehicle based on the vehicle identification code, wherein the thermal management capability parameter comprises theoretical maximum continuous charging power and thermal management type; the user thermal preference analysis module is used for acquiring historical charging behavior data of the request vehicle, constructing and training a hidden Markov model for deducing thermal behavior preference of the request vehicle based on the historical charging behavior data, and decoding to obtain a thermal behavior preference type implied by the request vehicle in the request; the hot air risk collaborative decision-making module is used for constructing a multi-agent reinforcement learning environment taking a charging station in an area as an agent, wherein the state space of each agent comprises an operation state, a current environment temperature and a hot air risk situation vector formed by aggregating a thermal management capability parameter and a hot air behavior preference type of a vehicle to be charged in the area; The personalized heat adaptation guiding module is used for calculating a heat adaptation degree score of a charging station in an area aiming at a single request vehicle, fusing heat management capability parameters of the single request vehicle, a decoded heat behavior preference type and a current environment temperature, and generating and outputting a heat adaptation charging guiding sequence by combining a station-level guiding strategy; the data communication and interface module is used for responding to a charging request initiated by a vehicle, acquiring a vehicle identification code and vehicle state information from a user terminal, acquiring operation state data from charging pile equipment, acquiring environment temperature data from a weather service provider, distributing the vehicle state information, the operation state data and the environment temperature data to corresponding processing processes, and simultaneously receiving the thermal adaptation charging guide sequence and outputting the thermal adaptation charging guide sequence to the user terminal.

Description

Charging station intelligent operation method and system based on charging user behavior characteristics Technical Field The invention relates to the field of electric vehicle charging facility operation and safety management, in particular to a charging station intelligent operation method and system based on charging user behavior characteristics. Background The method has the following limitations that firstly, the inherent differences of different vehicles on battery thermal management hardware capability such as heat dissipation type and maximum continuous charging power are not fully considered, so that the thermal safety risk of a charging process is lack of differential sensing and preventing capability, secondly, analysis on a charging behavior mode of a user is concentrated on macroscopic features such as time, location and electric quantity, personalized decision preference of the user when the user faces to thermal risks such as high temperature is not deeply mined, such as selection tendency of charging speed and charging station shielding attribute, so that the matching degree of an operation strategy and the user real demand is insufficient, thirdly, the traditional multi-cooperative optimization model is aimed at improving the overall utilization rate or income of the charging station, and an explicit modeling and cooperative regulation mechanism of the key constraint of thermal safety is lacked, so that the dynamic balance of the charging network between safety, efficiency and user experience is difficult to realize under severe environments such as high temperature, and the dynamic safety risk and the risk of the thermal safety risk and the mismatch of the resource are difficult to realize in the prior art. Disclosure of Invention The invention provides a charging station intelligent operation method and system based on charging user behavior characteristics, aiming at overcoming the defects of insufficient operation safety and inaccurate resource allocation in a high-heat risk scene caused by ignoring vehicle hardware thermal management differences, user thermal behavior preferences and multi-station thermal safety cooperation in the existing charging operation technology. The intelligent operation method of the charging station based on the behavior characteristics of the charging user comprises the following steps: The method comprises the following steps of S1, when a vehicle initiates a charging request, acquiring a vehicle identification code of the requesting vehicle, and analyzing and acquiring a thermal management capability parameter of the vehicle based on the vehicle identification code, wherein the thermal management capability parameter comprises theoretical maximum continuous charging power and a thermal management type; S2, acquiring historical charging behavior data of the request vehicle, constructing and training a hidden Markov model for deducing the thermal behavior preference of the request vehicle based on the historical charging behavior data, and decoding to obtain the type of the thermal behavior preference hidden in the request of the request vehicle; S3, constructing a multi-agent reinforcement learning environment taking a charging station in an area as an agent, wherein the state space of each agent comprises an operation state, a current environment temperature and a hot air risk situation vector formed by aggregation of thermal management capability parameters and hot air behavior preference types of vehicles to be charged in the area; And S4, aiming at a single request vehicle, fusing the thermal management capability parameters of the single request vehicle, the decoded thermal behavior preference type and the current environment temperature, calculating the thermal adaptation degree score of the charging station in the area, and generating and outputting a thermal adaptation charging guide sequence by combining a station-level guide strategy. The method comprises the steps of analyzing and obtaining thermal management capability parameters of a vehicle based on the vehicle identification code, wherein the thermal management capability parameters comprise battery type, battery rated capacity and thermal management configuration information of the vehicle are obtained according to the vehicle identification code through interaction with a manufacturer data platform, the thermal management type is determined according to the thermal management configuration information, the type comprises active liquid cooling, forced air cooling and natural cooling, and theoretical maximum continuous charging power of the vehicle under standard working conditions is determined according to the battery type and the rated capacity and in combination with battery technical specifications. Preferably, the hidden Markov model for deducing the thermal behavior preference of the request vehicle is constructed and trained based on the historical charging behavior data, and comprise