A parking lot intelligent identification power supply method, system and terminal device

Abstract

The application provides a parking lot intelligent identification power supply method, system and terminal device, the method comprises the following steps: collecting license plate information in real time and inputting the license plate information into a user database constructed in advance to perform matching query and obtaining corresponding contact information of a car owner; monitoring charging pile usage information, and sending charging recommendation information according to the charging pile usage information and the contact information of the car owner; selecting a target charging pile based on charging locking response information, locking the target charging pile and sending a navigation route from a parking lot entrance position to the target charging pile; when the target charging pile enters a charging state, generating a preliminary charging strategy according to real-time power of a charging vehicle and historical charging habits, and controlling output power of the target charging pile based on the preliminary charging strategy; generating a charging adjustment strategy according to a charging demand prediction result output by a charging resource and user demand prediction model, and switching the output power of the target charging pile. The application can improve flexibility of a charging strategy and charging experience of a user.

Description

Technical Field

[0001] This invention relates to the field of smart charging technology, and in particular to a parking lot intelligent identification and charging method, system and terminal equipment. Background Technology

[0002] In recent years, the market share of new energy vehicles has experienced explosive growth, with more and more consumers choosing them as their daily mode of transportation. However, this rapid popularization has also brought a series of new problems, among which the conflict between the construction and use of charging infrastructure is becoming increasingly prominent. Parking lots, as the primary locations for vehicle parking, are one of the important scenarios for charging new energy vehicles. However, the construction and management of charging stations in parking lots are currently lagging far behind the growth rate of new energy vehicles.

[0003] Existing parking lot charging systems lack effective intelligent identification and management mechanisms. Traditional parking lots lack effective mechanisms to allow car owners to promptly understand the availability of charging stations. Car owners often only discover that charging stations are full after entering the parking lot, or they have to spend a significant amount of time searching for an available station. This not only wastes the car owner's time and energy but may also lead to excessive battery drain. Furthermore, charging stations typically use a fixed output power, without dynamically adjusting based on the vehicle's real-time battery level and the car owner's historical charging habits. This can result in charging times that are too long or too short, failing to meet the actual needs of car owners. Moreover, the charging process does not adequately consider grid load conditions and electricity pricing policies. When the grid is under peak load, a large number of charging stations simultaneously charging at high power puts enormous pressure on the grid, increasing its operating costs. Additionally, car owners may charge when electricity prices are high, further increasing charging costs. Summary of the Invention

[0004] The present invention aims to provide a parking lot intelligent identification and charging method, system and terminal equipment to solve the above-mentioned technical problems, improve the flexibility of charging strategy and the user charging experience.

[0005] To address the aforementioned technical problems, this invention provides a smart identification and charging method for parking lots, comprising: The system collects license plate information of vehicles within the parking lot entrance area in real time, and inputs the collected license plate information into a pre-built user database for segmented matching and querying to obtain the contact information of the vehicle owners corresponding to the license plate information; wherein, the user database stores license plate information and corresponding user information; The system monitors the usage information of charging piles in the parking lot in real time, and sends charging recommendation information to the vehicle owner based on the charging pile usage information and the vehicle owner's contact information; wherein, the charging recommendation information includes real-time status information of the charging pile and charging pile lock inquiry information; Based on the charging lock response information sent by the vehicle owner, a target charging station is selected and locked, and a navigation route from the parking lot entrance to the target charging station is sent to the vehicle owner. When the target charging pile enters the charging state, a preliminary charging strategy is generated based on the real-time battery level of the charging vehicle and the owner's historical charging habits, and the output power of the target charging pile is controlled based on the preliminary charging strategy. Based on the charging demand prediction results output by the current charging resources and user demand prediction model, a charging adjustment strategy is generated, and the output power of the target charging pile is switched according to the charging adjustment strategy until charging is completed; wherein, the user demand prediction model is trained based on user information in the user database and is used to generate user charging demand prediction results for the parking lot within a preset time period.

[0006] In the above solution, by collecting license plate information within the parking lot entrance area in real time and performing segmented matching queries in a pre-built user database, the contact information of car owners can be obtained accurately and efficiently. Real-time monitoring of charging pile usage information within the parking lot is used to send charging recommendation information to car owners, ensuring they have access to the latest charging pile status. This charging recommendation information includes not only real-time charging pile status information but also charging pile lock-in inquiries, allowing car owners to understand the charging resource availability in the parking lot immediately and choose whether to lock a charging pile, thus rationally planning their charging behavior and improving the overall utilization efficiency of the parking lot's charging resources. Then, based on the charging lock-in response information sent by the car owner, a target charging pile is selected and locked, and a navigation route from the parking lot entrance to the target charging pile is sent to the car owner, providing a one-stop charging service. During charging, a preliminary charging strategy is generated based on the real-time battery level of the vehicle and the car owner's historical charging habits, and the output power of the target charging pile is controlled accordingly, achieving personalized charging services. Then, by combining the current charging resources and the charging demand prediction results output by the user demand prediction model, a charging adjustment strategy is generated. Based on the actual situation, the output power of the charging pile is dynamically adjusted, the power is reasonably allocated, and the overall charging efficiency is improved.

[0007] In one implementation, storing license plate information and corresponding user information in the user database specifically includes: When the charging pile equipment is in the charging state, the charging behavior data of the vehicle is collected in real time based on the charging pile equipment; wherein, the charging behavior data includes parking time, charging start time, charging end time, initial charging amount, final charging amount, and charging power; Based on the charging payment information of the charging pile equipment, vehicle license plate information and user identity information are obtained; wherein, the user identity information includes payment method and contact information; The user information is constructed based on the charging behavior data and user identity information; Using each license plate information as an index, establish a mapping relationship between the license plate information and the corresponding user information; The license plate information is stored in partitions based on a preset byte structure; wherein the preset byte structure is to sequentially split the license plate information into a province code segment, a license issuing authority segment, and a serial number segment.

[0008] In the above solution, real-time collection of charging behavior data allows operators to clearly understand vehicle owners' charging habits and patterns. Obtaining license plate and user identity information from charging payment data enables precise linking of vehicles and owners, facilitating communication and service. Combining these two aspects constructs user information, forming a comprehensive user profile and laying the foundation for personalized services. Establishing a mapping relationship using license plates as an index allows for efficient information retrieval. Partitioning and storing license plates by province code segment, issuing authority segment, and serial number segment improves query efficiency for massive amounts of data, helping parking lots achieve intelligent management, provide high-quality services, and optimize resource allocation.

[0009] In one implementation, the step of inputting the collected license plate information into a pre-built user database for segmented matching and querying to obtain the vehicle owner's contact information corresponding to the license plate information specifically includes: The license plate information is split based on the preset byte structure to obtain several bytes of information; Each byte of information is input into the user database for traversal and querying; When the corresponding byte information exists in the user database, the corresponding vehicle owner's contact information is retrieved based on the pre-established mapping relationship and the license plate information.

[0010] In the above solution, license plate information is split into byte segments using a pre-defined byte structure. This byte segment is then input into the user database for a comprehensive query, enabling precise location of license plate information within massive datasets. This segmented matching query method significantly improves query efficiency, reduces unnecessary search scope, and avoids the time and resource consumption associated with full data comparison. Once the corresponding byte segment information exists in the database, the vehicle owner's contact information can be quickly retrieved based on a pre-established mapping relationship. This provides parking lot operators with an efficient way to communicate important information such as charging recommendations to vehicle owners, improving service response speed and quality, and enhancing the intelligent management level of parking lots.

[0011] In one implementation, the real-time monitoring of charging pile usage information in the parking lot, and sending charging recommendation information to the vehicle owner based on the charging pile usage information and the vehicle owner's contact information, further includes: Based on the real-time monitoring of charging pile usage information in the parking lot by sensor devices, the number of charging piles in use in the parking lot is generated; When the number of charging piles in use exceeds a preset usage threshold, a parking lot charging pile information query request is initiated to a pre-established cloud service platform; wherein, the cloud service platform is used to realize information interaction between various parking lots; The cloud service platform provides charging pile usage recommendation information based on the parking lot charging pile information query request; wherein, the charging pile usage recommendation information is parking lot location information where the charging pile usage information is less than the preset usage quantity threshold and the parking lot location is within a preset distance range from the current parking lot. Based on the charging pile usage recommendation information and the vehicle owner's contact information, a charging recommendation message is sent to the vehicle owner.

[0012] In the above scheme, when the usage information exceeds a preset threshold, a query request is initiated to the cloud service platform. This platform facilitates information exchange between parking lots, overcoming the information limitations of individual parking lots and providing a broader overview of charging pile usage. The charging pile usage recommendations fed back from the cloud focus on parking lots with usage information below the threshold and within a certain distance, guiding drivers to locations with more abundant charging resources and avoiding long waits at the current parking lot. Based on this recommendation information, charging suggestions are sent to drivers, providing them with better charging options and improving their charging experience. This also helps balance charging pile usage across parking lots, improving overall resource utilization efficiency and achieving intelligent and collaborative management of parking lot charging services.

[0013] In one implementation, the step of selecting and locking a target charging station based on the charging lock response information sent by the vehicle owner and sending a navigation route from the parking lot entrance to the target charging station to the vehicle owner specifically includes: The target charging pile is selected based on the charging lock response information according to the preset matching rules; wherein, the preset matching rules include at least two of the following: charging power matching rules, interface type matching rules, distance priority matching rules, and usage frequency equalization matching rules; Send a locking command to the target charging station so that the target charging station switches its usage state to a locked state; When the target charging pile is switched to the locked state, a high-precision 3D map of the parking lot where the target charging pile is located is obtained and the target location of the target charging pile in the high-precision 3D map is marked; wherein, the high-precision 3D map is used to describe the environmental information within the parking lot; A navigation route from the parking lot entrance to the target location is generated based on a graph search algorithm.

[0014] The above solution selects target charging stations based on multiple preset matching rules, comprehensively considering factors such as charging power, interface type, distance, and usage frequency balance. This not only meets the charging needs of car owners but also improves the overall utilization efficiency of charging stations, avoiding excessive resource concentration or idleness. Sending a lock command to the target charging station ensures it is available upon the car owner's arrival, reducing waiting time and enhancing the charging experience. Using a high-precision 3D map to mark the location of the target charging station and combining it with a graph search algorithm to generate navigation routes provides accurate navigation within the parking lot. Even in complex parking environments, car owners can quickly find the target charging station, further optimizing the charging process and enhancing the intelligence and user-friendliness of parking services. In one implementation, when the target charging pile enters the charging state, a preliminary charging strategy is generated based on the real-time battery level of the vehicle and the owner's historical charging habits, and the output power of the target charging pile is controlled based on the preliminary charging strategy, specifically including: The vehicle owner's historical charging habits are generated based on the vehicle owner's user information; wherein, the charging habits include charging duration preferences and charging power preferences; When the target charging pile enters the charging state, the real-time power of the charging vehicle is obtained based on the communication between the target charging pile and the charging vehicle. The initial charging power of the charging vehicle is generated based on the real-time battery level, the charging duration preference, and the charging power preference, and the output power of the target charging pile is controlled based on the initial charging power.

[0015] In the above solution, when the target charging station enters the charging state, the vehicle's battery level is acquired in real time. This data, combined with historical charging habits, generates an initial charging power and controls the charging station's output. This allows the charging process to both meet the owner's preferences and be scientifically adjusted based on the vehicle's real-time battery level. It avoids inconvenience caused by excessively long or short charging times, power mismatches, and optimizes charging efficiency.

[0016] In one implementation, the user demand prediction model is trained based on user information in the user database, and the user demand prediction model is trained using the following method: Based on the aforementioned parking time, time-related features are generated; A charging duration feature is generated based on the charging start time and the charging end time; A charging quantity feature is generated based on the initial charging quantity and the final charging quantity. An average charging power feature is generated based on the charging duration feature and the charging amount feature; A target feature set is constructed based on the time-related features, charging duration features, charging amount features, and average charging power features. The random forest model is trained based on the user information so that it learns the relationship between charging demand and the target feature set; wherein, the random forest model traverses the user information using a grid search method to optimize model parameters. When the preset evaluation index reaches the preset threshold, the random forest model is deemed to have completed training, and the user demand prediction model is obtained.

[0017] The above solution comprehensively and meticulously depicts users' charging behavior patterns by extracting features from multiple aspects such as parking time, charging time, and battery level to construct a target feature set. Utilizing a random forest model to learn the relationship between charging demand and the feature set, combined with grid search for parameter tuning, improves the model's accuracy and generalization ability, ensuring that the model can effectively capture complex patterns in user charging demand. Model training is completed when preset evaluation indicators reach thresholds, guaranteeing the model's reliability and practicality. The resulting user demand prediction model can generate predictions of parking lot user charging demand within a preset time period. This helps parking lot managers to plan and allocate resources in advance, rationally arrange the use of charging piles, improve service quality, avoid resource idleness or shortages, and enhance the operational efficiency and economic benefits of parking lots.

[0018] In one implementation, the step of generating a charging adjustment strategy based on the charging demand prediction results output by the current charging resources and user demand prediction model, and switching the output power of the target charging pile based on the charging adjustment strategy until charging is completed, specifically includes: Obtain the charging resources at the current moment; wherein, the charging resources include the electricity pricing policies for different time periods and the maximum power capacity that the power grid in the area where the parking lot is located can provide to the charging piles in the parking lot at the current moment; The load status of the power grid is determined based on the charging demand forecast results; wherein the load status is determined based on the difference between the charging demand forecast results and the maximum power capacity. The charging needs of the vehicle owner are obtained and classified into urgency levels; wherein, the urgency levels include urgent needs, normal needs and non-urgent needs. Power output rules are pre-set according to the urgency of charging demand under different electricity pricing policies and grid load conditions; wherein, the power output rules are set based on rated power. The power output rules are invoked according to the load status and electricity pricing policy of the power grid, and the output power of the target charging pile is switched based on the power output rules.

[0019] The above-mentioned scheme categorizes vehicle owners' charging needs, accurately identifying charging demands of varying urgency levels. This provides a basis for subsequent charging strategy development, better meeting the diverse needs of vehicle owners. By comprehensively considering charging resources such as electricity pricing policies at different times and the maximum power capacity of the grid, it fully utilizes the grid's load characteristics and electricity price differences, achieving a rational allocation of resources.

[0020] Secondly, this application also provides a parking lot intelligent identification and charging system, including: an information matching module, a charging recommendation module, a route generation module, an initial control module, and a charging adjustment module; The information matching module is used to collect license plate information of vehicles within the parking lot entrance area in real time, and input the collected license plate information into a pre-built user database for segmented matching query to obtain the vehicle owner's contact information corresponding to the license plate information; wherein, the user database stores license plate information and corresponding user information; The charging recommendation module is used to monitor the usage information of charging piles in the parking lot in real time, and send charging recommendation information to the car owner based on the charging pile usage information and the car owner's contact information; wherein, the charging recommendation information includes real-time status information of the charging pile and charging pile lock inquiry information; The route generation module is used to select a target charging pile based on the charging lock response information sent by the vehicle owner, lock it, and send a navigation route from the parking lot entrance to the target charging pile to the vehicle owner. The initial control module is used to generate a preliminary charging strategy based on the real-time battery level of the charging vehicle and the owner's historical charging habits when the target charging pile enters the charging state, and to control the output power of the target charging pile based on the preliminary charging strategy. The charging adjustment module is used to generate a charging adjustment strategy based on the current charging resources and the charging demand prediction results output by the user demand prediction model, and to switch the output power of the target charging pile based on the charging adjustment strategy until charging is completed; wherein, the user demand prediction model is trained based on user information in the user database and is used to generate user charging demand prediction results for the parking lot within a preset time period.

[0021] In the above solution, by collecting license plate information within the parking lot entrance area in real time and performing segmented matching queries in a pre-built user database, the contact information of car owners can be obtained accurately and efficiently. Real-time monitoring of charging pile usage information within the parking lot is used to send charging recommendation information to car owners, ensuring they have access to the latest charging pile status. This charging recommendation information includes not only real-time charging pile status information but also charging pile lock-in inquiries, allowing car owners to understand the charging resource availability in the parking lot immediately and choose whether to lock a charging pile, thus rationally planning their charging behavior and improving the overall utilization efficiency of the parking lot's charging resources. Then, based on the charging lock-in response information sent by the car owner, a target charging pile is selected and locked, and a navigation route from the parking lot entrance to the target charging pile is sent to the car owner, providing a one-stop charging service. During charging, a preliminary charging strategy is generated based on the real-time battery level of the vehicle and the car owner's historical charging habits, and the output power of the target charging pile is controlled accordingly, achieving personalized charging services. Then, by combining the current charging resources and the charging demand prediction results output by the user demand prediction model, a charging adjustment strategy is generated. Based on the actual situation, the output power of the charging pile is dynamically adjusted, the power is reasonably allocated, and the overall charging efficiency is improved.

[0022] In one implementation, storing license plate information and corresponding user information in the user database specifically includes: When the charging pile equipment is in the charging state, the charging behavior data of the vehicle is collected in real time based on the charging pile equipment; wherein, the charging behavior data includes parking time, charging start time, charging end time, initial charging amount, final charging amount, and charging power; Based on the charging payment information of the charging pile equipment, vehicle license plate information and user identity information are obtained; wherein, the user identity information includes payment method and contact information; The user information is constructed based on the charging behavior data and user identity information; Using each license plate information as an index, establish a mapping relationship between the license plate information and the corresponding user information; The license plate information is stored in partitions based on a preset byte structure; wherein the preset byte structure is to sequentially split the license plate information into a province code segment, a license issuing authority segment, and a serial number segment.

[0023] In the above solution, real-time collection of charging behavior data allows operators to clearly understand vehicle owners' charging habits and patterns. Obtaining license plate and user identity information from charging payment data enables precise linking of vehicles and owners, facilitating communication and service. Combining these two aspects constructs user information, forming a comprehensive user profile and laying the foundation for personalized services. Establishing a mapping relationship using license plates as an index allows for efficient information retrieval. Partitioning and storing license plates by province code segment, issuing authority segment, and serial number segment improves query efficiency for massive amounts of data, helping parking lots achieve intelligent management, provide high-quality services, and optimize resource allocation.

[0024] In one implementation, the step of inputting the collected license plate information into a pre-built user database for segmented matching and querying to obtain the vehicle owner's contact information corresponding to the license plate information specifically includes: The license plate information is split based on the preset byte structure to obtain several bytes of information; Each byte of information is input into the user database for traversal and querying; When the corresponding byte information exists in the user database, the corresponding vehicle owner's contact information is retrieved based on the pre-established mapping relationship and the license plate information.

[0025] In the above solution, license plate information is split into byte segments using a pre-defined byte structure. This byte segment is then input into the user database for a comprehensive query, enabling precise location of license plate information within massive datasets. This segmented matching query method significantly improves query efficiency, reduces unnecessary search scope, and avoids the time and resource consumption associated with full data comparison. Once the corresponding byte segment information exists in the database, the vehicle owner's contact information can be quickly retrieved based on a pre-established mapping relationship. This provides parking lot operators with an efficient way to communicate important information such as charging recommendations to vehicle owners, improving service response speed and quality, and enhancing the intelligent management level of parking lots.

[0026] In one implementation, the charging recommendation module is used to monitor the usage information of charging piles in the parking lot in real time, and send charging recommendation information to the vehicle owner based on the charging pile usage information and the vehicle owner's contact information, and further includes: Based on the real-time monitoring of charging pile usage information in the parking lot by sensor devices, the number of charging piles in use in the parking lot is generated; When the number of charging piles in use exceeds a preset usage threshold, a parking lot charging pile information query request is initiated to a pre-established cloud service platform; wherein, the cloud service platform is used to realize information interaction between various parking lots; The cloud service platform provides charging pile usage recommendation information based on the parking lot charging pile information query request; wherein, the charging pile usage recommendation information is parking lot location information where the charging pile usage information is less than the preset usage quantity threshold and the parking lot location is within a preset distance range from the current parking lot. Based on the charging pile usage recommendation information and the vehicle owner's contact information, a charging recommendation message is sent to the vehicle owner.

[0027] In the above scheme, when the number of charging piles in use exceeds a preset threshold, a query request is sent to the cloud service platform. This platform facilitates information exchange between parking lots, overcoming the information limitations of individual parking lots and providing a broader overview of charging pile usage. The cloud-based charging pile usage recommendations focus on parking lots with usage below the threshold and within a certain distance, guiding drivers to locations with more abundant charging resources and avoiding long waits at their current parking lot. Based on this recommendation, charging suggestions are sent to drivers, providing them with better charging options and improving their charging experience. This also helps balance charging pile usage across parking lots, improving overall resource utilization efficiency and achieving intelligent and collaborative management of parking lot charging services.

[0028] In one implementation, the route generation module is used to select a target charging station based on the charging lock response information sent by the vehicle owner, lock the station, and send a navigation route from the parking lot entrance to the target charging station to the vehicle owner, specifically including: The target charging pile is selected based on the charging lock response information according to the preset matching rules; wherein, the preset matching rules include at least two of the following: charging power matching rules, interface type matching rules, distance priority matching rules, and usage frequency equalization matching rules; Send a locking command to the target charging station so that the target charging station switches its usage state to a locked state; When the target charging pile is switched to the locked state, a high-precision 3D map of the parking lot where the target charging pile is located is obtained and the target location of the target charging pile in the high-precision 3D map is marked; wherein, the high-precision 3D map is used to describe the environmental information within the parking lot; A navigation route from the parking lot entrance to the target location is generated based on a graph search algorithm.

[0029] The above solution selects target charging stations based on multiple preset matching rules, comprehensively considering factors such as charging power, interface type, distance, and usage frequency balance. This not only meets the charging needs of car owners but also improves the overall utilization efficiency of charging stations, avoiding excessive resource concentration or idleness. Sending a lock command to the target charging station ensures it is available upon the car owner's arrival, reducing waiting time and enhancing the charging experience. Using a high-precision 3D map to mark the location of the target charging station and combining it with a graph search algorithm to generate navigation routes provides accurate navigation within the parking lot. Even in complex parking environments, car owners can quickly find the target charging station, further optimizing the charging process and enhancing the intelligence and user-friendliness of parking services. In one implementation, the initial control module is used to generate a preliminary charging strategy based on the real-time battery level of the vehicle and the owner's historical charging habits when the target charging pile enters the charging state, and to control the output power of the target charging pile based on the preliminary charging strategy, specifically including: The vehicle owner's historical charging habits are generated based on the vehicle owner's user information; wherein, the charging habits include charging duration preferences and charging power preferences; When the target charging pile enters the charging state, the real-time power of the charging vehicle is obtained based on the communication between the target charging pile and the charging vehicle. The initial charging power of the charging vehicle is generated based on the real-time battery level, the charging duration preference, and the charging power preference, and the output power of the target charging pile is controlled based on the initial charging power.

[0030] In the above solution, when the target charging station enters the charging state, the vehicle's battery level is acquired in real time. This data, combined with historical charging habits, generates an initial charging power and controls the charging station's output. This allows the charging process to both meet the owner's preferences and be scientifically adjusted based on the vehicle's real-time battery level. It avoids inconvenience caused by excessively long or short charging times, power mismatches, and optimizes charging efficiency.

[0031] In one implementation, the user demand prediction model is trained based on user information in the user database, and the user demand prediction model is trained using the following method: Based on the aforementioned parking time, time-related features are generated; A charging duration feature is generated based on the charging start time and the charging end time; A charging quantity feature is generated based on the initial charging quantity and the final charging quantity. An average charging power feature is generated based on the charging duration feature and the charging amount feature; A target feature set is constructed based on the time-related features, charging duration features, charging amount features, and average charging power features. The random forest model is trained based on the user information so that it learns the relationship between charging demand and the target feature set; wherein, the random forest model traverses the user information using a grid search method to optimize model parameters. When the preset evaluation index reaches the preset threshold, the random forest model is deemed to have completed training, and the user demand prediction model is obtained.

[0032] The above solution comprehensively and meticulously depicts users' charging behavior patterns by extracting features from multiple aspects such as parking time, charging time, and battery level to construct a target feature set. Utilizing a random forest model to learn the relationship between charging demand and the feature set, combined with grid search for parameter tuning, improves the model's accuracy and generalization ability, ensuring that the model can effectively capture complex patterns in user charging demand. Model training is completed when preset evaluation indicators reach thresholds, guaranteeing the model's reliability and practicality. The resulting user demand prediction model can generate predictions of parking lot user charging demand within a preset time period. This helps parking lot managers to plan and allocate resources in advance, rationally arrange the use of charging piles, improve service quality, avoid resource idleness or shortages, and enhance the operational efficiency and economic benefits of parking lots.

[0033] In one implementation, the charging adjustment module is used to generate a charging adjustment strategy based on the current charging resources and the charging demand prediction results output by the user demand prediction model, and to switch the output power of the target charging pile based on the charging adjustment strategy until charging is completed, specifically including: Obtain the charging resources at the current moment; wherein, the charging resources include the electricity pricing policies for different time periods and the maximum power capacity that the power grid in the area where the parking lot is located can provide to the charging piles in the parking lot at the current moment; The load status of the power grid is determined based on the charging demand forecast results; wherein the load status is determined based on the difference between the charging demand forecast results and the maximum power capacity. The charging needs of the vehicle owner are obtained and classified into urgency levels; wherein, the urgency levels include urgent needs, normal needs and non-urgent needs. Power output rules are pre-set according to the urgency of charging demand under different electricity pricing policies and grid load conditions; wherein, the power output rules are set based on rated power. The power output rules are invoked according to the load status and electricity pricing policy of the power grid, and the output power of the target charging pile is switched based on the power output rules.

[0034] The above-mentioned scheme categorizes vehicle owners' charging needs, accurately identifying charging demands of varying urgency levels. This provides a basis for subsequent charging strategy development, better meeting the diverse needs of vehicle owners. By comprehensively considering charging resources such as electricity pricing policies at different times and the maximum power capacity of the grid, it fully utilizes the grid's load characteristics and electricity price differences, achieving a rational allocation of resources.

[0035] Thirdly, this application also provides a terminal device, including a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein the processor executes the computer program to implement the parking lot intelligent identification and power replenishment method as described above. Attached Figure Description

[0036] Figure 1 This is a flowchart illustrating a parking lot intelligent identification and charging method provided in one embodiment of the present invention; See Figure 2 , Figure 2 A statistical chart of charging utilization data provided in Huizhou cloud platform according to an embodiment of the present invention; Figure 3 This is a data display diagram of energy consumption report of Huizhou charging cloud platform provided in one embodiment of the present invention; Figure 4 This is a statistical chart of supply and demand data for charging stations on the Huizhou charging cloud platform provided in one embodiment of the present invention; Figure 5 This is a schematic diagram of a parking lot intelligent identification and power replenishment system provided in one embodiment of the present invention. Detailed Implementation

[0037] The specific embodiments of the present invention will be described in further detail below with reference to the accompanying drawings and examples. The following examples are for illustrative purposes only and are not intended to limit the scope of the invention.

[0038] The terms "first" and "second," etc., in the specification, claims, and drawings of this application are used to distinguish different objects, not to describe a specific order. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or apparatus that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to such processes, methods, products, or apparatus.

[0039] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0040] Example 1 See Figure 1 , Figure 1 This is a flowchart illustrating a parking lot intelligent identification and power replenishment method according to an embodiment of the present invention. The embodiment of the present invention provides a parking lot intelligent identification and power replenishment method, including steps 101 to 105, each step being as follows: Step 101: Collect license plate information of vehicles within the parking lot entrance area in real time, and input the collected license plate information into a pre-built user database for segmented matching query to obtain the vehicle owner's contact information corresponding to the license plate information; wherein, the user database stores license plate information and corresponding user information.

[0041] In this embodiment of the invention, to ensure real-time and accurate acquisition of license plate information within the parking lot entrance area, a high-definition license plate recognition camera is installed at the parking lot entrance to recognize the license plates of vehicles entering the parking lot's detection range. Preferably, a high-definition, high-speed license plate recognition camera is selected. These cameras have high resolution and fast capture capabilities, enabling them to clearly capture license plate images as vehicles quickly pass through the entrance. The color of the captured license plate image is then used to determine whether the vehicle at the entrance is a new energy vehicle. Specifically, if the license plate image has a green background and black characters, it is determined to be an image of a new energy vehicle, and the captured vehicle number is input into a pre-built user database for querying. If the license plate image has a blue background and white characters, it is determined to be an image of a gasoline-powered vehicle. Since gasoline-powered vehicles do not require the use of charging stations, there is no need to input their license plate numbers into the user database for querying.

[0042] Real-time collection of license plate information and rapid matching and querying of vehicle owner contact information enables quick identification and processing of vehicles entering the parking lot. This avoids the tedious process of traditional manual registration, reduces vehicle dwell time at the entrance, and improves parking lot throughput. The segmented matching query method utilizes an optimized database structure and efficient algorithms to shorten query time, allowing the parking management system to respond quickly to vehicle entry and improve overall management efficiency.

[0043] In one embodiment, the user database stores license plate information and corresponding user information, specifically including: when the charging pile device is in a charging state, collecting vehicle charging behavior data in real time based on the charging pile device; wherein, the charging behavior data includes parking time, charging start time, charging end time, initial charging level, final charging level, and charging power; obtaining license plate information and user identity information based on the charging payment information of the charging pile device; wherein, the user identity information includes payment method and contact information; constructing the user information based on the charging behavior data and user identity information; establishing a mapping relationship between the license plate information and the corresponding user information using each license plate information as an index; and storing the license plate information in partitions based on a preset byte structure; wherein, the preset byte structure is to sequentially split the license plate information into a provincial code segment, a license plate issuing authority segment, and a serial number segment.

[0044] In this embodiment of the invention, the charging pile equipment is equipped with high-precision sensors, such as current sensors, voltage sensors, and timing devices, to accurately collect data such as parking time, charging start time, charging end time, initial charging level, final charging level, and charging power. These sensors should possess high reliability and stability, and be able to operate accurately under different environmental conditions. To comprehensively obtain user payment information, the charging pile equipment supports multiple payment methods, such as WeChat Pay, Alipay, and bank card payments. Simultaneously, data interfaces are established with major payment platforms to obtain user payment information in real time. After obtaining user identity information, user contact information and identity are verified through methods such as SMS verification codes and facial recognition to ensure the authenticity and validity of the information. Preferably, before storing the collected data, filtering algorithms (such as Kalman filtering) can be used to filter the data and remove the influence of noise interference. Simultaneously, the sensors are calibrated regularly to ensure the accuracy and reliability of the data.

[0045] Then, a user database is constructed using a storage architecture combining a distributed file system (such as Hadoop Distributed File System, HDFS) and a relational database (such as MySQL). HDFS is used to store large-scale raw data, such as charging behavior data and license plate images; MySQL is used to store structured user information and mapping relationships. License plate information and user identity information, after cleaning and standardization, are stored in the MySQL database. License plate information is partitioned according to province code segment, issuing authority segment, and serial number segment, with each partition corresponding to a database table. Simultaneously, a mapping relationship table between license plate information and user information is established for easy querying and management. Preferably, charging behavior data is stored in the distributed file system in raw data form, and can be further partitioned according to time and charging pile number. For example, daily data is stored in a separate folder, and each folder is further divided into subfolders according to charging pile number. Parking lot managers can query the charging records and user information of a specific vehicle through the management system. The system quickly locates the corresponding user information and mapping relationship in the MySQL database based on the license plate information, and simultaneously retrieves the relevant charging behavior data from HDFS.

[0046] In one embodiment, the step of inputting the collected license plate information into a pre-built user database for segmented matching and querying to obtain the vehicle owner's contact information corresponding to the license plate information specifically includes: splitting the license plate information based on the preset byte structure to obtain several bytes of information; inputting each byte of information into the user database for traversal query; when the corresponding byte of information exists in the user database, calling the corresponding vehicle owner's contact information based on the pre-established mapping relationship and the license plate information.

[0047] In this embodiment of the invention, before splitting the license plate information based on a preset byte structure, the collected license plate information is preprocessed. Because problems such as blurred license plate information and character recognition errors may occur during actual collection, preliminary error correction and format standardization are performed on the license plate information. For example, letters in the license plate are converted to uppercase, and redundant spaces and special characters are removed. Preferably, an index can be created for each byte information segment in the user database, such as a provincial code segment index, a license issuing authority segment index, and a sequence number segment index. Indexes can speed up queries and avoid performing a full traversal query each time. For example, a B-tree index or a hash index can be used, selecting the appropriate index type according to different query scenarios. During traversal queries, a progressive query strategy is adopted. The query starts with the provincial code segment to narrow the query range, and then the license issuing authority segment and the sequence number segment are queried sequentially. If no matching result is found in a certain segment, the query can be terminated early to reduce unnecessary query operations. When a unique record is found and the license plate number is a perfect match, the corresponding vehicle owner's contact information field is obtained through the association relationship.

[0048] Step 102: Monitor the usage information of charging piles in the parking lot in real time, and send charging recommendation information to the car owner based on the charging pile usage information and the car owner's contact information; wherein, the charging recommendation information includes real-time status information of the charging pile and charging pile lock inquiry information.

[0049] In one embodiment, the real-time monitoring of charging pile usage information in the parking lot and sending charging recommendation information to the vehicle owner based on the charging pile usage information and the vehicle owner's contact information further includes: real-time monitoring of charging pile usage information in the current parking lot using sensor devices; generating the number of charging piles in use in the parking lot based on the charging pile usage information; when the number of charging piles in use exceeds a preset charging pile usage threshold, initiating a parking lot charging pile information query request to a pre-established cloud service platform; wherein, the cloud service platform is used to realize information interaction between various parking lots; the cloud service platform provides charging pile usage recommendation information based on the parking lot charging pile information query request; wherein, the charging pile usage recommendation information is parking lot location information where the charging pile usage information is less than the preset usage threshold and the distance between the parking lot and the current parking lot is within a preset distance range; and sending charging recommendation information to the vehicle owner based on the charging pile usage recommendation information and the vehicle owner's contact information.

[0050] In this embodiment of the invention, multiple sensor devices, such as current sensors, voltage sensors, and occupancy sensors, are installed on each charging pile in the parking lot. The current and voltage sensors are used to monitor in real time whether the charging pile is charging a vehicle and the charging power, while the occupancy sensor detects whether the charging pile is occupied by a vehicle. Data from each sensor is collected and processed in real time to form charging pile usage information, including the occupancy status of each charging pile, charging duration, and remaining charging time. Based on this charging pile usage information, the number of charging piles currently in use in the parking lot can be calculated. See also... Figure 2 , Figure 2 This is a statistical chart of charging utilization data provided by a Huizhou cloud platform according to an embodiment of the present invention. (Attached) Figure 2It shows the core charging operation data of multiple parking lots in Huizhou area from June 9 to September 9, 2025 (3 months), including total aggregated data such as total charging capacity (246090.747kWh), total number of orders (11646 orders), total charging duration (1363 days 6 hours 26 minutes 52 seconds), and average daily utilization rate (2.47%), and also lists detailed information including charging capacity, number of orders, charging duration and daily utilization rate of multiple stations such as Xinhu Park No.1 Charging Station and City Science Museum Parking Lot Charging Station. These data intuitively reflect the service efficiency of charging piles in each parking lot. A threshold for the number of used charging piles is set in the local server. When the number of real-time monitored used charging piles, that is, the number of charging piles being used, exceeds the threshold, the local server automatically initiates a query request for parking lot charging pile information to the pre-established cloud service platform. After receiving the query request, the cloud service platform uses big data analysis technology to process and analyze the charging pile use information of each parking lot in real time. It screens out parking lots where the number of used charging piles is less than a preset usage quantity threshold, and the distance between the parking lot and the current parking lot is within a preset distance range. The cloud service platform organizes the screened parking lot location information, number of idle charging piles, charging standards and other information into charging pile usage recommendation information, and feeds it back to the local server that initiated the query request. The local server sends charging recommendation information to vehicle owners via short messages, APP push and other methods according to the received charging pile usage recommendation information and the vehicle owner's contact information. The recommendation information not only includes the basic information of the parking lot, but also provides a navigation link and estimated arrival time. When a vehicle owner receives the charging recommendation information and has a charging demand, they can send corresponding information to the local server via short message reply, APP operation and other methods. For example, when a new energy vehicle drives into a parking lot, the high-definition license plate recognition camera at the entrance quickly recognizes the license plate number as "Guangdong A 12345 green plate". The local server quickly finds the owner's mobile phone number 138xxxx5678 corresponding to the license plate in the database through a segmented query method. If the utilization rate of charging piles in the current parking lot is lower than the preset utilization threshold, that is, there are idle charging piles available for charging in the current parking lot, the charging pile usage information of the current parking lot and the charging pile locking inquiry information are directly sent to the owner's mobile phone number. If the utilization rate of charging piles in the current parking lot exceeds the preset threshold (e.g., 80%), the local server initiates a query request to the cloud service platform. After screening, the cloud service platform finds that the utilization rate of charging piles in Parking Lot A, which is 3 kilometers away from the current parking lot, is only 30%, with 10 charging piles idle, and the charging rate is 1.5 yuan per kWh. The cloud service platform feeds back the information of Parking Lot B to the local server of Parking Lot A, and the local server sends the charging recommendation information to the vehicle owner via short message: "Dear vehicle owner, the charging piles in the xx parking lot where you are located are relatively saturated. There are 10 idle charging piles in Parking Lot B 3 kilometers away from you, and the charging rate is 1.5 yuan per kWh. Click [navigation link] to go there."After receiving the text message, the car owner opens the mobile map application by clicking the navigation link and confirms that they are going to parking lot B for charging. After receiving the car owner's confirmation, the local server of the current parking lot communicates with parking lot A to pre-lock a charging station. If the car owner does not need to charge, they can directly enter the parking lot to park after receiving the charging recommendation information."

[0051] Step 103: Based on the charging lock response information sent by the vehicle owner, select the target charging pile, lock it, and send the vehicle owner a navigation route from the parking lot entrance to the target charging pile.

[0052] In this embodiment of the invention, after the car owner sends a response message, the charging station can be quickly located, avoiding the situation where an unavailable charging station cannot be found after arriving at the parking lot, saving time and effort in searching for a charging station. It directly provides the car owner with a navigation route from the parking lot entrance to the target charging station, eliminating the need for the car owner to navigate within the parking lot and reducing the possibility of getting lost in an unfamiliar parking lot. Especially for large multi-story parking lots or complex parking lots, accurate navigation helps car owners reach their destination quickly and accurately, improving the parking and charging experience.

[0053] In one embodiment, the step of selecting and locking a target charging pile based on the charging lock response information sent by the vehicle owner and sending a navigation route from the parking lot entrance to the target charging pile to the vehicle owner specifically includes: selecting a target charging pile based on the charging lock response information according to preset matching rules; wherein, the preset matching rules include at least two of the following: charging power matching rules, interface type matching rules, distance priority matching rules, and usage frequency balancing matching rules; sending a lock command to the target charging pile to cause the target charging pile to switch its usage state to a locked state; when the target charging pile switches to the locked state, acquiring a high-precision 3D map of the parking lot where the target charging pile is located and marking the target location of the target charging pile in the high-precision 3D map; wherein, the high-precision 3D map is used to describe the environmental information within the parking lot; and generating a navigation route from the parking lot entrance to the target location based on a graph search algorithm.

[0054] In this embodiment of the invention, after the vehicle owner sends a charging lock response, the system analyzes the owner's user information, especially charging behavior data, to select a suitable target charging pile. For example, under the charging power matching rule, charging piles matching the owner's required charging power are prioritized. For instance, if the owner needs 60kW fast charging, the system will prioritize selecting charging piles supporting 60kW and above. Interface type matching selects a suitable charging pile based on the vehicle's charging interface type. For example, some new energy vehicles use specific fast charging interfaces, and the system ensures the selected charging pile has the corresponding interface. It should be noted that the charging power matching and interface type data can be obtained from user information in the user database. Each time a target charging pile is selected, the charging power matching rule and interface type matching rule are used as the primary matching methods. Then, the distance priority matching rule and usage frequency balancing rule are considered. Under the premise of satisfying the charging power and interface type requirements, charging piles closer to the parking entrance are selected to reduce the owner's driving distance. The distance from each charging pile to the parking entrance can be calculated using pre-set parking lot map coordinates. Furthermore, the frequency balancing matching rule is used to avoid overuse of some charging piles. It considers the historical usage frequency of each charging pile and prioritizes those with lower usage frequencies to extend their lifespan. Once a target charging pile is selected, the system communicates with it via a specific communication protocol (such as Modbus or CAN) to send a locking command. This command also includes information such as the license plate number. Upon receiving the locking command, the charging pile updates its status from "idle" to "locked" and displays the locked license plate number on its screen, allowing the driver to locate the target charging pile. Then, high-precision 3D map data of the parking lot is retrieved. This map details the parking lot's environment, including floor layout, passageways, parking spaces, elevators, and stairs. The specific location of the target charging pile is marked on the high-precision 3D map, generating precise coordinates. Simultaneously, the coordinates of the parking lot entrance are recorded, providing foundational data for subsequent navigation route generation. In this embodiment of the invention, when generating a navigation route based on a graph search algorithm, key locations within the parking lot, such as intersections, turns, elevator entrances, stairwells, charging station locations, and parking lot entrances, are defined as nodes in the graph based on a high-precision 3D map. Each node has a unique identifier and corresponding 3D coordinates. Passage segments connecting adjacent nodes are defined as edges in the graph. Each edge contains attributes such as the length, direction of travel, and width of the passage. Based on the actual conditions of the passage, each edge is assigned a weight value, which can comprehensively consider factors such as passage length and difficulty of passage. Then, a navigation route is generated based on Dijkstra's algorithm. The distance value of the starting node (parking lot entrance) is set to 0, and the distance values ​​of other nodes are set to infinity. A priority queue is created, and the starting node is added to the queue. The node with the smallest distance value is retrieved from the priority queue and designated as the current node.The algorithm iterates through all adjacent nodes of the current node, calculating the distance from the starting node to an adjacent node via the current node. If the calculated distance is less than the current distance value of the adjacent node, the distance value of the adjacent node and its predecessor node are updated. The updated adjacent node is added to a priority queue. The algorithm terminates when the distance value of the target charging pile node is updated and it is removed from the priority queue. This graph search algorithm starts from the target charging pile node and generates a navigation route from the parking lot entrance to the target charging pile by backtracking its predecessor nodes. Furthermore, when the target charging pile is not in the current parking lot, it also needs to generate a route from the current parking lot entrance to the parking lot in the recommended information. The navigation route from the current parking lot entrance to the parking lot where the target charging pile is located and the navigation route from the entrance of the parking lot where the target charging pile is located to the target charging pile are concatenated to generate the final navigation route. The generated navigation route is then displayed visually on the car owner's mobile application.

[0055] Step 104: When the target charging pile enters the charging state, a preliminary charging strategy is generated based on the real-time battery level of the charging vehicle and the owner's historical charging habits, and the output power of the target charging pile is controlled based on the preliminary charging strategy.

[0056] In this embodiment of the invention, after the target charging pile enters the charging state, real-time battery level of the charging vehicle and the owner's historical charging habits are collected. A preliminary charging strategy is generated based on these two data points, taking into account the vehicle's current battery level, the owner's past charging duration, power preferences, etc. The output power of the target charging pile is then controlled according to this preliminary charging strategy. Combining real-time battery level and historical habits to formulate a strategy allows for the selection of appropriate charging power for different situations, enabling faster charging and saving time for owners seeking quick charging. Simultaneously, because the system monitors the power usage of all charging piles in the parking lot in real time, it dynamically adjusts the output power of each charging pile based on the total capacity of the transformer. In the initial stage of vehicle charging, the initial charging power is rationally allocated to avoid uneven transformer load caused by some charging piles operating at high power for extended periods. For example, when some vehicles are low on battery and urgently need fast charging, while meeting their fast charging needs, the charging power for other non-urgent charging vehicles is appropriately reduced, thus providing charging opportunities for more vehicles within the transformer's capacity.

[0057] In one embodiment, when the target charging pile enters the charging state, a preliminary charging strategy is generated based on the real-time battery level of the charging vehicle and the owner's historical charging habits. The output power of the target charging pile is then controlled based on the preliminary charging strategy. Specifically, this includes: generating the owner's historical charging habits based on the owner's user information; wherein the charging habits include charging duration preferences and charging power preferences; obtaining the real-time battery level of the charging vehicle based on communication between the target charging pile and the charging vehicle when the target charging pile enters the charging state; generating an initial charging power for the charging vehicle based on the real-time battery level, the charging duration preference, and the charging power preference; and controlling the output power of the target charging pile based on the initial charging power.

[0058] In this embodiment of the invention, the charging history of vehicle owners is obtained based on their user information (especially charging behavior data). Charging duration preference analysis and charging power analysis are then performed based on this user data. For example, in the charging market analysis, charging duration is categorized and statistically analyzed according to different time periods (e.g., weekdays, weekends, daytime, nighttime) and seasons (spring, summer, autumn, winter). By calculating the mean, median, and standard deviation, the distribution of charging duration preferences among vehicle owners in various scenarios is accurately determined. For instance, on summer weekends, vehicle owners may prefer longer, slower charging; while on weekday evenings, they may prefer faster charging. Similarly, in the charging power preference analysis, the frequency and probability of vehicle owners choosing different charging powers are analyzed according to different scenario factors. Considering the impact of the vehicle battery's age and health status on charging power selection, a dynamic charging power preference model is constructed. For example, as battery usage time increases, vehicle owners may gradually reduce charging power to protect the battery. Based on the above analysis, the historical charging habits of vehicle owners are generated. Then, when the target charging station enters the charging state (i.e., the charging gun of the charging station connects to the vehicle), the target charging station communicates with the vehicle's battery management system (BMS) through a certain communication protocol (e.g., ISO 15118 protocol) to obtain the vehicle's current battery level. The real-time battery level is divided into several detailed ranges, such as extremely low battery (0-10%), low battery (11%-20%), low-medium battery (21%-40%), medium battery (41%-60%), medium-high battery (61%-80%), high battery (81%-90%), and full battery (91%-100%). For each range, different initial charging power strategies are formulated based on charging time and power preferences. Preferably, when the battery temperature is too high or too low, the initial charging power can be appropriately reduced to protect battery performance. For example, Ms. Zhang drives her electric vehicle to a public charging station, parks her car next to the target charging station, plugs in the charging gun, and the charging station enters the charging state. Ms. Zhang's user information is retrieved from the user database. Analysis revealed that Ms. Zhang typically prefers to charge her car within one hour on weekday evenings, favoring higher charging power. On weekends, however, she prefers longer, slower charging times and is less demanding on charging duration. The charging station communicates with the vehicle's Battery Management System (BMS) via the ISO 15118 protocol, obtaining information such as the vehicle's current battery level of 15%, battery temperature of 22°C, and good battery health. Given that it is currently a weekday evening, Ms. Zhang's real-time battery level is low, and she prefers fast charging, a higher charging power is selected, specifically an initial output power of 60kW. The target charging station then begins charging the vehicle at 60kW. Preferably, during charging, the battery temperature and charging speed can be continuously monitored through communication between the target charging station and the vehicle. When the battery temperature rises to a certain level, the output power can be appropriately reduced to protect the vehicle's battery.

[0059] Step 105: Generate a charging adjustment strategy based on the current charging resources and the charging demand prediction results output by the user demand prediction model, and switch the output power of the target charging pile based on the charging adjustment strategy until charging is completed; wherein, the user demand prediction model is trained based on user information in the user database and is used to generate user charging demand prediction results for the parking lot within a preset time period.

[0060] In this embodiment of the invention, an adjustment strategy is generated based on the charging demand prediction results. This strategy can flexibly switch the output power of charging piles according to actual needs, avoiding resource waste or insufficiency and improving the efficiency of charging resource utilization. With the help of a well-trained user demand prediction model, user charging needs can be accurately estimated, making the charging strategy more aligned with actual conditions and better serving users.

[0061] In one embodiment, the user demand prediction model is trained based on user information in the user database. The user demand prediction model is trained using the following methods: generating time-related features based on parking time; generating charging duration features based on charging start time and charging end time; generating charging amount features based on initial charging amount and final charging amount; generating average charging power features based on charging duration features and charging amount features; constructing a target feature set based on the time-related features, charging duration features, charging amount features, and average charging power features; training a random forest model based on the user information to enable the random forest model to learn the relationship between charging demand and the target feature set; wherein, the random forest model traverses the user information using a grid search method to optimize model parameters; when a preset evaluation index reaches a preset threshold, the random forest model is deemed to have completed training, thus obtaining the user demand prediction model.

[0062] In this embodiment of the invention, information related to user charging behavior is comprehensively collected from the user database. Then, based on parking time-related features, parking time is divided into specific weekday information, time period information, and holiday information. Different weeks often correspond to different user travel and charging patterns; for example, charging demand on weekdays and weekends may differ significantly. The day is divided into multiple time periods, such as morning peak (7-9 am), morning off-peak (9-11 am), lunch break (11 am-1 pm), afternoon off-peak (1 pm-5 pm), evening peak (5 pm-7 pm), and nighttime off-peak (7 pm-7 am the next day). Charging demand and resource usage differ in each time period; this division allows for a more detailed capture of the impact of time on charging demand. It also determines whether the parking time falls on a statutory holiday or special holiday, as people's travel plans and charging needs change significantly during holidays; for example, charging demand increases dramatically before long holidays. Charging duration is an important indicator reflecting user charging habits and device efficiency; different users may have significantly different charging durations due to factors such as travel plans and vehicle battery capacity. Charging time can be obtained from the difference between the start and end times of charging. The charging amount directly reflects the user's charging demand and is one of the key factors in predicting future charging needs. The user's charging amount is obtained by subtracting the initial charging amount from the final charging amount. The average charging power is related to the performance of the charging equipment and the user's charging choice; different charging pile powers and user requirements for charging speed will lead to different average charging power. The average charging power is obtained by dividing the charging amount by the charging time. Based on the above time-related features, charging time features, charging amount features, and average charging power features, a target feature set is constructed. Then, a random forest model is selected as the user demand prediction model. Random forest is an ensemble learning method composed of multiple decision trees, possessing good generalization ability and anti-overfitting ability, and capable of handling nonlinear relationships. When initializing the random forest model, some initial hyperparameters are set, including: the number of decision trees, the maximum tree depth, and the minimum number of sample splits. It should be noted that before inputting user information into the random forest model for training, certain data preprocessing is required, including missing value handling, outlier handling, and data standardization. The data preprocessing steps described above are standard techniques in this field and will not be elaborated upon here. The preprocessed and feature-generated data is divided into training, validation, and test sets according to a certain ratio (e.g., 7:2:1). The training set is used for model training, the validation set is used for hyperparameter tuning and evaluation, and the test set is used for final model performance evaluation. The grid search method is used to tune the parameters of the random forest model. The search range for the model parameters is defined, for example: Number of decision trees (n_estimators): Set to [50, 100, 150, 200].

[0063] The maximum depth of the tree (max_depth) is set to [5,10,15,20].

[0064] Minimum number of sample splits (min_samples_split): set to [2,5,10].

[0065] A grid search is used to traverse all hyperparameter combinations, and the model's performance is evaluated on the validation set. The hyperparameter combination that optimizes the preset evaluation metrics (including mean squared error (MSE), root mean square error (RMSE), and mean absolute error (MAE)) is selected. The random forest model is then trained using the training set. During training, the model learns the relationship between charging demand and various target features, continuously adjusting the structure and parameters of the decision tree to minimize prediction error. The goal of training is to enable the model to learn general patterns and regularities from the training data. The trained model is then evaluated using the validation set, and preset evaluation metrics are calculated. Based on the evaluation results, it is determined whether the model's performance meets the requirements. If the evaluation metrics do not reach the preset threshold, further adjustments to the hyperparameters or data processing are needed, followed by retraining and re-evaluation. For example, a user's parking time is 18:30 on August 10, 2025, charging starts at 18:45, the initial charging level is 20%, and the current charging power of the charging station is 7 kW. This information is converted into corresponding time-related features. Specifically, the day of the week is Sunday (converted to the corresponding numerical code, assuming Sunday is coded as 0); the time of day is evening peak (coded as 4, assuming the evening peak time is coded as 4); and whether it is a holiday is no (coded as 0). User information data with similar time-related characteristics is collected from the user database. Based on this collected user information data, a user demand prediction model is used to predict the charging demand of the entire parking lot during the user's charging time period, and to predict the parking lot's charging volume during this period. The collected user information data is based on the number of charging piles in the parking lot, assuming all charging piles are in operation. User information matching the number of charging piles is collected for prediction. Furthermore, the estimated charging volume for users can be generated based on the target feature set. Specifically, charging duration: since charging is not yet finished, it is not calculated here, but the possible charging duration range can be estimated based on historical data and the current time. Assuming, based on historical data, the average charging time during this period and with this initial charge level is 2 hours; the average charging power is currently 7 kW; and the initial charge level is 20% (converted to a specific charge value, assuming a vehicle battery capacity of 50 kWh, then the initial charge level is 50 × 20% = 10 kWh). These features are input into a trained random forest model, which then outputs a predicted charge level. Let's assume the model's predicted charge level is 30 kWh. Based on this predicted charge level of 30 kWh, a corresponding charging adjustment strategy can be generated in conjunction with current charging resources. See also... Figure 3 , Figure 3This is a data display chart of energy consumption reports from a charging cloud platform in Huizhou, provided in one embodiment of the present invention. The chart presents monthly charging statistics for representative parking lots in Huizhou from January to September 2025. This chart visually verifies the accuracy of the user demand prediction model of the present invention. The model's prediction of monthly charging demand for each parking lot based on historical energy consumption data has a deviation rate of less than 5% from the actual energy consumption report data. It also assists managers in analyzing the matching relationship between transformer load and charging volume, thereby generating corresponding charging adjustment strategies based on the specific circumstances of each parking lot.

[0066] In one embodiment, the step of generating a charging adjustment strategy based on the charging demand prediction results output by the current charging resources and user demand prediction model, and switching the output power of the target charging pile based on the charging adjustment strategy until charging is completed, specifically includes: acquiring the charging resources at the current moment; wherein, the charging resources include electricity price policies for different time periods and the maximum power capacity that the power grid in the area where the parking lot is located can provide to the parking lot charging pile at the current moment; determining the load status of the power grid based on the charging demand prediction results; wherein, the load status is determined based on the difference between the charging demand prediction results and the maximum power capacity; acquiring the charging needs of the vehicle owner and classifying the charging needs according to their urgency; wherein, the urgency types include emergency needs, normal needs, and non-emergency needs; pre-setting power output rules under different electricity price policies and power grid load statuses according to the urgency type of the charging needs; wherein, the power output rules are set based on the rated power; calling the corresponding power output rules according to the power grid load status and electricity price policy, and switching the output power of the target charging pile based on the power output rules.

[0067] In this embodiment of the invention, a data interface is established between the parking lot management system and the local power supplier to obtain electricity pricing policies for different time periods in real time. Preferably, in addition to simple low-price, high-price, and flat-period policies, detailed records are also needed on the specific time range, price fluctuation range, and any special pricing rules that may exist (such as holiday pricing, dynamic adjustments to peak-valley pricing, etc.). Through collaboration with the power grid management department, smart meters or power monitoring equipment are used to obtain the maximum power capacity that the power grid in the parking lot area can provide to the parking lot's charging piles at any given time. Simultaneously, historical data on this capacity is obtained, and its variation patterns in different time periods and seasons are analyzed. The load status is determined based on the difference between the charging demand forecast and the maximum power capacity, and the concept of load factor can also be introduced. Load factor = Charging demand forecast / Maximum power capacity. Based on different load factor ranges, the power grid load status is divided into low load, high load, and flat load states. Preferably, it can be further subdivided into multiple levels such as extremely light load, light load, flat load, heavy load, and extremely heavy load according to the manager's needs; this is only an example and not intended to limit the scope. For different combinations of grid load conditions and electricity pricing policies, detailed power output rules are set for various urgency levels. For example, when the grid load is light and the electricity price is low, the output power of the target charging station is increased to the rated power for vehicles with urgent needs; the output power is increased to 80% of the rated power for vehicles with normal needs; and the output power is increased to 60% of the rated power for vehicles with non-urgent needs.

[0068] For example, by analyzing a car owner's historical charging records, their charging habits and patterns can be understood, such as the times they frequently charge and the amount of electricity required per charge, serving as a reference for classifying charging needs. Furthermore, when using a target charging station, car owners will be required to input basic information such as their desired charging capacity and expected departure time. Based on the urgency level being divided into urgent needs, normal needs, and non-urgent needs, the corresponding classification criteria are as follows: Emergency needs: The car owner clearly states that a certain amount of electricity must be charged within a short period of time (such as within 1 hour) to meet the needs of emergency travel; or the vehicle's remaining battery power is extremely low and cannot support reaching the next charging point.

[0069] Typical needs: Car owners do not have a specific emergency time limit, but hope to complete charging within a reasonable time, generally expecting parking time to be between 2 and 4 hours.

[0070] Non-urgent needs: Car owners have a longer parking time (e.g., more than 4 hours), do not have strict requirements for charging time, and are more concerned about charging costs; the vehicle has relatively sufficient remaining battery power, which can meet the recent travel needs even without immediate charging.

[0071] After categorizing car owners' charging needs by urgency, the system continuously collects key information about the power grid in the parking lot's area through a real-time data interface with the power grid company. This includes real-time load, frequency, voltage, available capacity, and electricity price details for different time periods. Simultaneously, smart meters and related monitoring equipment are deployed within the parking lot to monitor local total power consumption, electricity usage, power factor, and other data in real time, accurately grasping the interaction between the parking lot and the power grid. When the power grid is in a low-load period and electricity prices are low, the output power of the target charging pile is increased to its rated power. Preferably, if the parking lot is equipped with energy storage devices, the low-priced electricity is used to charge the energy storage devices, increasing their energy reserves for later use, while still meeting the car owners' charging needs. When the power grid is in a high-load period and electricity prices are high, for example, for vehicles with urgent needs, the output power is reduced to 60%-70% of the rated power; for vehicles with normal and non-urgent needs, it can be reduced to 30%-50%. For vehicles with non-urgent needs, certain preferential policies (such as parking fee reductions, discounts on the next charge, etc.) can be offered to encourage them to suspend charging or extend charging time to off-peak hours. Preferably, information on electricity prices and power adjustments can be sent to car owners via an app or SMS to gain their understanding and cooperation. For example, informing car owners that current electricity prices are high and that charging power will be appropriately reduced to lower costs, with an expected extension of charging time. When the grid is under load and electricity prices are flat, priority should be given to ensuring charging speeds for vehicles with urgent needs, increasing output power to meet their emergency charging requirements and ensuring charging is completed in the shortest possible time. For vehicles with regular charging needs, output power is dynamically adjusted based on the vehicle's real-time battery level and expected parking time, minimizing charging costs while ensuring charging is completed within the parking time. For vehicles with non-urgent needs, output power is appropriately reduced to lower charging costs. Simultaneously, negotiations can be held with car owners regarding their willingness to accelerate charging speeds and offer more favorable charging prices during off-peak hours.

[0072] See Figure 4 , Figure 4 This is a statistical chart of supply and demand data for charging stations on a Huizhou charging cloud platform, provided in one embodiment of the present invention. (Attached) Figure 4This data displays information on September 8, 2025 (single-day) for multiple charging stations in Huizhou City, including the Science and Technology Museum parking lot and the Convention and Exhibition Center parking lot. This includes the number of charging piles, the number of charging guns, the maximum number of charging guns that can be used, the maximum load capacity, the highest charging peak, and the highest peak demand. For example, the Science and Technology Museum parking lot has 5 charging piles and 10 charging guns, a maximum load capacity of 560 kWh, a highest charging peak of 345 kWh, and a highest peak demand of 630 kWh, representing 113% of the total demand. This data clearly presents the supply and demand situation for charging piles in each parking lot. The comparison between peak demand and maximum load capacity helps determine transformer load pressure and provides a basis for developing strategies to meet the charging needs of more vehicles without increasing the number of transformers. For example, for parking lots where peak demand exceeds the maximum load capacity, the load can be balanced by dynamically adjusting the charging power. Another example is a large commercial parking lot with 100 charging piles equipped with an intelligent charging management system. Between 2:00 AM and 5:00 AM, the grid load is low and electricity prices are low (RMB 0.3 / kWh). The maximum power capacity available to the parking lot is 500kW. There are 20 vehicles charging in the parking lot: 10 for regular use, 5 for emergency use, and 5 for non-emergency use. The intelligent charging management system, based on the vehicles' battery characteristics and charging needs, increases the output power of charging stations supporting high-power fast charging to their rated power. For example, for an electric vehicle supporting 100kW fast charging, the charging station output power is increased to 100kW. Preferably, the low-priced electricity can also be used to charge the parking lot's energy storage equipment, which has a charging power of 50kW. Between 12:00 PM and 2:00 PM, the grid load is high and electricity prices are high (RMB 1.2 / kWh). The maximum power capacity available to the parking lot is 300kW. There are 30 vehicles charging in the parking lot, including 8 vehicles with emergency needs, 12 vehicles with normal needs, and 10 vehicles with non-emergency needs. The system reduces the charging pile output power of vehicles with emergency needs to 60% of their rated power, vehicles with normal needs to 40%, and vehicles with non-emergency needs to 30%. For example, a vehicle with a rated power of 80kW will have its output power reduced to 32kW. Simultaneously, the system sends electricity price and power adjustment information to car owners via app or SMS, and offers a 5 yuan parking fee reduction to owners of vehicles with non-emergency needs to encourage them to suspend charging. Between 4:00 PM and 6:00 PM, when the grid load is stable and the electricity price is at its flat rate (0.6 yuan / kWh), the grid load is relatively stable, and the maximum power capacity available to the parking lot is 400kW. At this time, 25 vehicles are charging in the parking lot, including 5 vehicles with emergency needs, 10 vehicles with normal needs, and 10 vehicles with non-emergency needs.For vehicles with urgent charging needs, the output power of their charging stations is increased to meet their emergency charging requirements. For example, if a vehicle needs to charge from 20% to 80% within one hour, its output power is increased to 60kW. For vehicles with normal charging needs, the output power is dynamically adjusted based on their expected parking time and real-time battery level. For example, if a vehicle with normal charging needs is expected to park for two hours and currently has 40% battery, its output power is adjusted to 40kW. For vehicles with non-urgent charging needs, the output power is appropriately reduced to 20kW, and the owner is informed that they can enjoy a 20% discount when charging during off-peak hours in the early morning. Preferably, during the charging process, real-time charging information, including current output power, estimated charging time, and charging cost, can be provided to the owner through a mobile app or charging station display screen to improve user satisfaction and engagement.

[0073] In this embodiment of the invention, a parking lot intelligent identification and power replenishment device is also provided, including a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, it implements the above-described parking lot intelligent identification and power replenishment method.

[0074] In this embodiment of the invention, a computer-readable storage medium is also provided, which includes a stored computer program, wherein the computer program controls the device where the computer-readable storage medium is located to execute the above-described intelligent parking lot identification and power replenishment method when it is running.

[0075] For example, a computer program can be divided into one or more modules, one or more of which are stored in memory and executed by a processor to perform the present invention. One or more modules can be a series of computer program instruction segments capable of performing a specific function, which describe the execution process of the computer program in a parking lot intelligent identification and power replenishment device.

[0076] The parking lot intelligent identification power supply device can be a computing device such as a desktop computer, laptop, handheld computer, or cloud server. The parking lot intelligent identification power supply device may include, but is not limited to, a processor, memory, and display. Those skilled in the art will understand that the above components are merely examples of the parking lot intelligent identification power supply device and do not constitute a limitation on the device. It may include more or fewer components, or a combination of certain components, or different components. For example, the parking lot intelligent identification power supply device may also include input / output devices, network access devices, buses, etc.

[0077] The processor can be a Central Processing Unit (CPU), or other general-purpose processors, Digital Signal Processors (DSPs), Application Specific Integrated Circuits (ASICs), Field-Programmable Gate Arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor can be a microprocessor or any conventional processor. The processor is the control center of the fuel cell performance recovery equipment, connecting various parts of the intelligent identification and power replenishment equipment in the parking lot through various interfaces and lines.

[0078] The memory can be used to store computer programs and / or modules. The processor implements various functions of the fuel cell performance recovery device by running or executing the computer programs and / or modules stored in the memory, and by calling the data stored in the memory. The memory can mainly include a program storage area and a data storage area. The program storage area can store the operating system, at least one application program required for a function (such as sound playback function, text conversion function, etc.), etc.; the data storage area can store data created according to the use of the mobile phone (such as audio data, text message data, etc.). In addition, the memory can include high-speed random access memory, and can also include non-volatile memory, such as hard disk, RAM, plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, at least one disk storage device, flash memory device, or other volatile solid-state storage device.

[0079] The module for intelligent parking lot identification and power replenishment, if implemented as a software functional unit and sold or used as an independent product, can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the above embodiments of the present invention can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include: any entity or device capable of carrying computer program code, recording media, USB flash drives, portable hard drives, magnetic disks, optical disks, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signals, telecommunication signals, and software distribution media, etc. Those skilled in the art can understand and implement this without any inventive effort.

[0080] This invention provides an intelligent identification and charging method for parking lots. By collecting license plate information within the parking lot entrance area in real time and performing segmented matching queries in a pre-built user database, it can accurately and efficiently obtain the contact information of car owners. It monitors the usage information of charging piles within the parking lot in real time and sends charging recommendation information to car owners accordingly, ensuring they have access to the latest charging pile status. The charging recommendation information includes not only real-time charging pile status information but also charging pile lock inquiry information, allowing car owners to understand the charging resource availability in the parking lot immediately and choose whether to lock a charging pile, rationally planning their charging behavior and improving the overall utilization efficiency of charging resources in the parking lot. Then, based on the charging lock response information sent by the car owner, a target charging pile is selected and locked, and a navigation route from the parking lot entrance to the target charging pile is sent to the car owner, providing a one-stop charging service. During charging, a preliminary charging strategy is generated based on the real-time battery level of the vehicle and the car owner's historical charging habits, and the output power of the target charging pile is controlled accordingly, achieving personalized charging service. Then, by combining the current charging resources and the charging demand prediction results output by the user demand prediction model, a charging adjustment strategy is generated. Based on the actual situation, the output power of the charging pile is dynamically adjusted, the power is reasonably allocated, and the overall charging efficiency is improved.

[0081] Example 2 See Figure 5 , Figure 5This is a schematic diagram of a parking lot intelligent identification and charging system provided in one embodiment of the present invention. The present invention provides a parking lot intelligent identification and charging system, including: an information matching module 201, a charging recommendation module 202, a route generation module 203, an initial control module 204, and a charging adjustment module 205; The information matching module 201 is used to collect license plate information of vehicles within the parking lot entrance area in real time, and input the collected license plate information into a pre-built user database for segmented matching query to obtain the vehicle owner's contact information corresponding to the license plate information; wherein, the user database stores license plate information and corresponding user information; The charging recommendation module 202 is used to monitor the usage information of charging piles in the parking lot in real time, and send charging recommendation information to the car owner based on the charging pile usage information and the car owner's contact information; wherein, the charging recommendation information includes real-time status information of the charging pile and charging pile lock inquiry information; The route generation module 203 is used to select a target charging pile based on the charging lock response information sent by the vehicle owner, lock it, and send a navigation route from the parking lot entrance to the target charging pile to the vehicle owner. The initial control module 204 is used to generate a preliminary charging strategy based on the real-time battery level of the charging vehicle and the owner's historical charging habits when the target charging pile enters the charging state, and to control the output power of the target charging pile based on the preliminary charging strategy. The charging adjustment module 205 is used to generate a charging adjustment strategy based on the current charging resources and the charging demand prediction results output by the user demand prediction model, and to switch the output power of the target charging pile based on the charging adjustment strategy until the charging ends; wherein, the user demand prediction model is trained based on user information in the user database, and is used to generate user charging demand prediction results for the parking lot within a preset time period.

[0082] In one embodiment, the user database stores license plate information and corresponding user information, specifically including: when the charging pile device is in a charging state, collecting vehicle charging behavior data in real time based on the charging pile device; wherein, the charging behavior data includes parking time, charging start time, charging end time, initial charging level, final charging level, and charging power; obtaining license plate information and user identity information based on the charging payment information of the charging pile device; wherein, the user identity information includes payment method and contact information; constructing the user information based on the charging behavior data and user identity information; establishing a mapping relationship between the license plate information and the corresponding user information using each license plate information as an index; and storing the license plate information in partitions based on a preset byte structure; wherein, the preset byte structure is to sequentially split the license plate information into a provincial code segment, a license issuing authority segment, and a serial number segment.

[0083] In one embodiment, the step of inputting the collected license plate information into a pre-built user database for segmented matching and querying to obtain the vehicle owner's contact information corresponding to the license plate information specifically includes: splitting the license plate information based on the preset byte structure to obtain several bytes of information; inputting each byte of information into the user database for traversal query; when the corresponding byte of information exists in the user database, calling the corresponding vehicle owner's contact information based on the pre-established mapping relationship and the license plate information.

[0084] In one embodiment, the charging recommendation module is used to monitor the usage information of charging piles in the parking lot in real time, and send charging recommendation information to the car owner based on the charging pile usage information and the car owner's contact information. The module further includes: monitoring the usage information of charging piles in the current parking lot in real time using sensor devices, generating the number of charging piles in use in the parking lot based on the charging pile usage information; when the number of charging piles in use exceeds a preset charging pile usage threshold, initiating a parking lot charging pile information query request to a pre-established cloud service platform; wherein, the cloud service platform is used to realize information interaction between various parking lots; the cloud service platform provides charging pile usage recommendation information based on the parking lot charging pile information query request; wherein, the charging pile usage recommendation information is parking lot location information where the charging pile usage information is less than the preset usage threshold and the distance between the parking lot and the current parking lot is within a preset distance range; and sending a charging recommendation message to the car owner based on the charging pile usage recommendation information and the car owner's contact information.

[0085] In one embodiment, the route generation module is used to select a target charging pile based on the charging lock response information sent by the vehicle owner, lock it, and send a navigation route from the parking lot entrance to the target charging pile to the vehicle owner. Specifically, this includes: selecting a target charging pile based on a preset matching rule for the charging lock response information; wherein the preset matching rule includes at least two of the following: charging power matching rule, interface type matching rule, distance priority matching rule, and usage frequency balancing matching rule; sending a lock command to the target charging pile to switch its usage state to a locked state; when the target charging pile switches to the locked state, acquiring a high-precision 3D map of the parking lot where the target charging pile is located and marking the target location of the target charging pile in the high-precision 3D map; wherein the high-precision 3D map is used to describe the environmental information within the parking lot; and generating a navigation route from the parking lot entrance to the target location based on a graph search algorithm.

[0086] In one embodiment, the initial control module is used to generate a preliminary charging strategy based on the real-time battery level of the charging vehicle and the owner's historical charging habits when the target charging pile enters the charging state, and to control the output power of the target charging pile based on the preliminary charging strategy. Specifically, this includes: generating the owner's historical charging habits based on the owner's user information; wherein the charging habits include charging duration preferences and charging power preferences; obtaining the real-time battery level of the charging vehicle based on communication between the target charging pile and the charging vehicle when the target charging pile enters the charging state; generating an initial charging power for the charging vehicle based on the real-time battery level, the charging duration preference, and the charging power preference, and controlling the output power of the target charging pile based on the initial charging power.

[0087] In one embodiment, the user demand prediction model is trained based on user information in the user database. The user demand prediction model is trained through the following methods: generating time-related features based on parking time; generating charging duration features based on charging start time and charging end time; generating charging amount features based on charging start amount and charging end amount; generating average charging power features based on charging duration features and charging amount features; constructing a target feature set based on the time-related features, charging duration features, charging amount features, and average charging power features; training a random forest model based on the user information so that the random forest model learns the relationship between charging demand and the target feature set; wherein, the random forest model traverses the user information using a grid search method to achieve model parameter tuning; when a preset evaluation index reaches a preset threshold, the random forest model is determined to be trained successfully, and the user demand prediction model is obtained.

[0088] In one embodiment, the charging adjustment module is used to generate a charging adjustment strategy based on the current charging resources and the charging demand prediction results output by the user demand prediction model, and to switch the output power of the target charging pile based on the charging adjustment strategy until charging is completed. Specifically, this includes: acquiring the charging resources at the current time; wherein the charging resources include electricity price policies for different time periods and the maximum power capacity that the power grid in the parking lot area can provide to the parking lot charging pile at the current time; determining the load status of the power grid based on the charging demand prediction results; wherein the load status is determined based on the difference between the charging demand prediction results and the maximum power capacity; acquiring the vehicle owner's charging needs and classifying the charging needs by urgency; wherein the urgency types include emergency needs, normal needs, and non-emergency needs; pre-setting power output rules under different electricity price policies and power grid load statuses according to the urgency type of the charging needs; wherein the power output rules are set based on the rated power; and calling the corresponding power output rules according to the power grid load status and electricity price policy, and switching the output power of the target charging pile based on the power output rules.

[0089] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working process of the device described above can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0090] This invention provides an intelligent parking lot identification and charging system. By collecting license plate information from within the parking lot entrance area in real time and performing segmented matching queries in a pre-built user database, it can accurately and efficiently obtain the contact information of car owners. It monitors the usage information of charging piles within the parking lot in real time and sends charging recommendation information to car owners accordingly, ensuring they have access to the latest charging pile status. The charging recommendation information includes not only real-time charging pile status information but also charging pile lock inquiry information, allowing car owners to understand the charging resource availability in the parking lot immediately and choose whether to lock a charging pile, rationally planning their charging behavior and improving the overall utilization efficiency of parking lot charging resources. Then, based on the charging lock response information sent by the car owner, a target charging pile is selected and locked, and a navigation route from the parking lot entrance to the target charging pile is sent to the car owner, providing a one-stop charging service. During charging, a preliminary charging strategy is generated based on the real-time battery level of the vehicle and the car owner's historical charging habits, and the output power of the target charging pile is controlled accordingly, achieving personalized charging services. Then, by combining the current charging resources and the charging demand prediction results output by the user demand prediction model, a charging adjustment strategy is generated. Based on the actual situation, the output power of the charging pile is dynamically adjusted, the power is reasonably allocated, and the overall charging efficiency is improved.

[0091] The above are merely preferred embodiments of the present invention. It should be noted that those skilled in the art can make several improvements and substitutions without departing from the technical principles of the present invention, and these improvements and substitutions should also be considered within the scope of protection of the present invention.

Claims

1. A method for intelligent identification and power replenishment in parking lots, characterized in that, include: The system collects license plate information of vehicles within the parking lot entrance area in real time, and inputs the collected license plate information into a pre-built user database for segmented matching and querying to obtain the contact information of the vehicle owners corresponding to the license plate information; wherein, the user database stores license plate information and corresponding user information; The system monitors the usage information of charging piles in the parking lot in real time, and sends charging recommendation information to the vehicle owner based on the charging pile usage information and the vehicle owner's contact information; wherein, the charging recommendation information includes real-time status information of the charging pile and charging pile lock inquiry information; Based on the charging lock response information sent by the vehicle owner, a target charging station is selected and locked, and a navigation route from the parking lot entrance to the target charging station is sent to the vehicle owner. When the target charging pile enters the charging state, a preliminary charging strategy is generated based on the real-time battery level of the charging vehicle and the owner's historical charging habits, and the output power of the target charging pile is controlled based on the preliminary charging strategy. A charging adjustment strategy is generated based on the charging demand prediction results output by the current charging resources and user demand prediction model, and the output power of the target charging pile is switched according to the charging adjustment strategy until charging is completed; wherein, the user demand prediction model is trained based on user information in the user database and is used to generate user charging demand prediction results for the parking lot within a preset time period.

2. The intelligent identification and charging method for parking lots as described in claim 1, characterized in that, Storing license plate information and corresponding user information in the user database specifically includes: When the charging pile equipment is in the charging state, the charging behavior data of the vehicle is collected in real time based on the charging pile equipment; wherein, the charging behavior data includes parking time, charging start time, charging end time, initial charging amount, final charging amount, and charging power; Based on the charging payment information of the charging pile equipment, vehicle license plate information and user identity information are obtained; wherein, the user identity information includes payment method and contact information; The user information is constructed based on the charging behavior data and user identity information; Using each license plate information as an index, establish a mapping relationship between the license plate information and the corresponding user information; The license plate information is stored in partitions based on a preset byte structure; wherein the preset byte structure is to sequentially split the license plate information into a province code segment, a license issuing authority segment, and a serial number segment.

3. The intelligent identification and charging method for parking lots as described in claim 2, characterized in that, The step of inputting the collected license plate information into a pre-built user database for segmented matching and querying to obtain the vehicle owner's contact information corresponding to the license plate information specifically includes: The license plate information is split based on the preset byte structure to obtain several bytes of information; Each byte of information is input into the user database for traversal and querying; When the corresponding byte information exists in the user database, the corresponding vehicle owner's contact information is retrieved based on the pre-established mapping relationship and the license plate information.

4. The intelligent identification and charging method for parking lots as described in claim 1, characterized in that, The method of real-time monitoring of charging pile usage information in the parking lot, and sending charging recommendation information to the vehicle owner based on the charging pile usage information and the vehicle owner's contact information, also includes: Based on the real-time monitoring of charging pile usage information in the parking lot by sensor devices, the number of charging piles in use in the parking lot is generated; When the number of charging piles in use exceeds a preset usage threshold, a parking lot charging pile information query request is initiated to a pre-established cloud service platform; wherein, the cloud service platform is used to realize information interaction between various parking lots; The cloud service platform provides charging pile usage recommendation information based on the parking lot charging pile information query request; wherein, the charging pile usage recommendation information is parking lot location information where the charging pile usage information is less than the preset usage quantity threshold and the parking lot location is within a preset distance range from the current parking lot. Based on the charging pile usage recommendation information and the vehicle owner's contact information, a charging recommendation message is sent to the vehicle owner.

5. The intelligent identification and charging method for parking lots as described in claim 1, characterized in that, The process of selecting and locking a target charging station based on the charging lock response information sent by the vehicle owner, and sending a navigation route from the parking lot entrance to the target charging station to the vehicle owner, specifically includes: The target charging pile is selected based on the charging lock response information according to the preset matching rules; wherein, the preset matching rules include at least two of the following: charging power matching rules, interface type matching rules, distance priority matching rules, and usage frequency equalization matching rules; Send a locking command to the target charging station so that the target charging station switches its usage state to a locked state; When the target charging pile is switched to the locked state, a high-precision 3D map of the parking lot where the target charging pile is located is obtained and the target location of the target charging pile in the high-precision 3D map is marked; wherein, the high-precision 3D map is used to describe the environmental information within the parking lot; A navigation route from the parking lot entrance to the target location is generated based on a graph search algorithm.

6. The intelligent identification and charging method for parking lots as described in claim 1, characterized in that, When the target charging pile enters the charging state, a preliminary charging strategy is generated based on the real-time battery level of the vehicle and the owner's historical charging habits. The output power of the target charging pile is then controlled based on this preliminary charging strategy. Specifically, this includes: The vehicle owner's historical charging habits are generated based on the vehicle owner's user information; wherein, the charging habits include charging duration preferences and charging power preferences; When the target charging pile enters the charging state, the real-time power of the charging vehicle is obtained based on the communication between the target charging pile and the charging vehicle. The initial charging power of the charging vehicle is generated based on the real-time battery level, the charging duration preference, and the charging power preference, and the output power of the target charging pile is controlled based on the initial charging power.

7. The intelligent identification and charging method for parking lots as described in claim 2, characterized in that, The user demand prediction model is trained based on user information in the user database, and is trained using the following method: Based on the aforementioned parking time, time-related features are generated; A charging duration feature is generated based on the charging start time and the charging end time; A charging quantity feature is generated based on the initial charging quantity and the final charging quantity. An average charging power feature is generated based on the charging duration feature and the charging amount feature; A target feature set is constructed based on the time-related features, charging duration features, charging amount features, and average charging power features. The random forest model is trained based on the user information so that it learns the relationship between charging demand and the target feature set; wherein, the random forest model traverses the user information using a grid search method to optimize model parameters. When the preset evaluation index reaches the preset threshold, the random forest model is deemed to have completed training, and the user demand prediction model is obtained.

8. The intelligent identification and charging method for parking lots as described in claim 1, characterized in that, The process of generating a charging adjustment strategy based on the charging demand prediction results output by the current charging resources and user demand prediction model, and switching the output power of the target charging pile based on the charging adjustment strategy until charging is completed, specifically includes: Obtain the charging resources at the current moment; wherein, the charging resources include the electricity pricing policies for different time periods and the maximum power capacity that the power grid in the area where the parking lot is located can provide to the charging piles in the parking lot at the current moment; The load status of the power grid is determined based on the charging demand forecast results; wherein the load status is determined based on the difference between the charging demand forecast results and the maximum power capacity. The charging needs of the vehicle owner are obtained and classified into urgency levels; wherein, the urgency levels include urgent needs, normal needs and non-urgent needs. Power output rules are pre-set according to the urgency of charging demand under different electricity pricing policies and grid load conditions; wherein, the power output rules are set based on rated power. The power output rules are invoked according to the load status and electricity pricing policy of the power grid, and the output power of the target charging pile is switched based on the power output rules.

9. A parking lot intelligent identification and power replenishment system, characterized in that, include: Information matching module, charging recommendation module, route generation module, initial control module, and charging adjustment module; The information matching module is used to collect license plate information of vehicles within the parking lot entrance area in real time, and input the collected license plate information into a pre-built user database for segmented matching query to obtain the vehicle owner's contact information corresponding to the license plate information; wherein, the user database stores license plate information and corresponding user information; The charging recommendation module is used to monitor the usage information of charging piles in the parking lot in real time, and send charging recommendation information to the car owner based on the charging pile usage information and the car owner's contact information; wherein, the charging recommendation information includes real-time status information of the charging pile and charging pile lock inquiry information; The route generation module is used to select a target charging pile based on the charging lock response information sent by the vehicle owner, lock it, and send a navigation route from the parking lot entrance to the target charging pile to the vehicle owner. The initial control module is used to generate a preliminary charging strategy based on the real-time battery level of the charging vehicle and the owner's historical charging habits when the target charging pile enters the charging state, and to control the output power of the target charging pile based on the preliminary charging strategy. The charging adjustment module is used to generate a charging adjustment strategy based on the current charging resources and the charging demand prediction results output by the user demand prediction model, and to switch the output power of the target charging pile based on the charging adjustment strategy until charging is completed; wherein, the user demand prediction model is trained based on user information in the user database and is used to generate user charging demand prediction results for the parking lot within a preset time period.

10. A terminal device, characterized in that, It includes a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein the processor executes the computer program to implement the parking lot intelligent identification and charging method as described in any one of claims 1 to 8.

Images