Parking lot charging pile management system and management method

Abstract

The invention discloses a parking lot charging pile management system and method, and the system comprises a plurality of parking spaces disposed in a parking lot, a plurality of charging piles disposed in the parking spaces, a plurality of dual-spectrum thermal imaging cameras, a system servo processor, a charging pile controller, and a management platform. The double-spectrum thermal imaging camera is used for shooting the state of a parking space and the thermal imaging temperature of a vehicle, the charging pile is electrically connected with the charging pile controller, the double-spectrum thermal imaging camera and the charging pile controller are both electrically connected with the system servo processor, and the system servo processor is in signal connection with the management platform. The charging pile controller comprises an acousto-optic warning indicator and is used for collecting charging data in the charging process of the charging pile. The method is applied to the technical field of new energy charging.

Description

Technical Field

[0001] The present invention is applied to the technical field of new energy charging, and in particular relates to a parking lot charging pile management system and a management method. Background Art

[0002] The electric vehicle industry is in a stage of rapid development of intelligence, and governments of various countries have allocated funds to support corresponding incentive policies for charging piles. At the same time, parking lots are required to be equipped with a certain proportion of charging piles. Due to favorable policies and public attention caused by social events such as electric vehicle fires, fire prevention of charging piles has become a requirement that must be considered by parking lot management. All types of electric vehicle charging are subject to fire risks. 80% of electric vehicle fires occur during the charging process. Due to the characteristics of the concentration of charging piles, once the electric vehicle burns, toxic smoke will quickly spread upwards, and the fire will spread rapidly, causing huge economic losses and personal dangers. If the problem of safe monitoring and management of charging vehicles and parking spaces in parking lots can be solved, safety hazards can be discovered and eliminated as soon as possible, and the safe charging of vehicles and the correct process can be standardized, the risk of fire during electric vehicle charging can be greatly reduced. Summary of the invention

[0003] The technical problem to be solved by the present invention is to overcome the deficiencies of the prior art and provide a parking lot charging pile management system and management method that combines a dual-spectrum thermal imaging camera with charging pile data acquisition to achieve all-round and high-precision detection of the status of charging parking spaces.

[0004] The technical solution adopted by the present invention is: the present invention includes a parking lot charging pile management system and a management method. The parking lot charging pile management system includes a plurality of parking spaces arranged in the parking lot, and a plurality of charging piles arranged in the parking spaces. The charging piles are used to charge electric vehicles in the parking spaces. The management system also includes a plurality of dual-spectrum thermal imaging cameras, a system servo processor, a charging pile controller and a management platform. The dual-spectrum thermal imaging camera is used to capture the status of the parking space and the thermal imaging temperature of the vehicle. The charging pile is electrically connected to the charging pile controller. The dual-spectrum thermal imaging camera and the charging pile controller are both electrically connected to the system servo processor. The system servo processor is signal-connected to the management platform. The charging pile controller includes an audible and visual alarm. The charging pile controller is used to collect charging data during the charging process of the charging pile and set a safe charging data threshold. If the actual charging data exceeds the set threshold, an audible and visual alarm is used to warn, and the information is wirelessly transmitted to the management platform through the system servo processor.

[0005] Furthermore, the dual-spectrum thermal imaging camera includes a high-definition camera and an infrared thermal imaging monitor. The high-definition camera is integrated with an AI image recognition module. The high-definition camera is used to capture the status of a parking space, whether it is vacant or with a vehicle parked. When capturing a parking space with a vehicle parked, the license plate captured at the same time is determined by the AI ​​image recognition module to determine whether the charging space is occupied by a non-electric vehicle. The infrared thermal imaging monitor is used to capture the battery temperature status during the charging process of the electric vehicle.

[0006] Furthermore, the charging pile controller includes a box body and a cover body adapted to the box body, the box body is provided with a control module, a switching power supply and an air switch, the sound and light alarm is a plurality of LED buzzers installed on the cover body, and the LED buzzer, the switching power supply and the air switch are all electrically connected to the control module.

[0007] Furthermore, an electric gate is installed at the entrance of the parking lot, and a display is provided at the electric gate. The electric gate and the display are electrically connected to the system servo processor. Each parking space has a corresponding number and positioning coordinates. The display is used to display the number of vacant parking spaces and the number of vacant electric vehicle charging spaces, as well as to display a parking lot map and mark the locations of vacant parking spaces.

[0008] Furthermore, the parking space includes one or both of a car parking space and a battery vehicle parking space.

[0009] Furthermore, a management method including the parking lot charging pile management system includes the following steps: S1. Use a high-definition camera to capture the status of parking spaces in the parking lot, and send the number of free parking spaces, the number of charging spaces in the parking lot, and the parking space coordinate information to the management platform; S2. The vehicle enters the parking lot according to the number and coordinates of the displayed free parking spaces and parks in the parking space. If the parking space is a charging space, the high-definition camera captures the license plate number and uses the Yolo real-time target detection algorithm to determine whether the vehicle is an electric vehicle. If the vehicle is not an electric vehicle, the information about the non-electric vehicle occupying the electric parking space is sent to the management platform; S3. If it is determined that the vehicle is a tram, the charging pile of the charging parking space is further determined whether the tram is recharged. If the charging pile has no charging information, the information of the electric parking space occupied by the tram in the non-charging state is sent to the management platform; S4. If the charging pile enters the charging state, the temperature status information of the electric car is detected in real time by the infrared thermal imaging monitor, and the charging pile uploads the charging data to the charging pile controller. If the battery temperature exceeds the set temperature threshold or the actual charging data exceeds the set safe charging data threshold, the sound and light alarm will warn, and the information will be wirelessly transmitted to the management platform through the system servo processor; S5. If the battery temperature and charging data are lower than the set threshold during the charging process of the electric vehicle, the charging state will be maintained. When the battery is fully charged, the management platform will send the full charge information to the owner's mobile phone and start timing at the same time. If the timing after charging exceeds the set time, the information that the electric vehicle has exceeded the time limit for occupying the charging parking space will be sent to the management platform.

[0010] Furthermore, a convolutional block attention module is introduced on the basis of the Yolo real-time target detection algorithm in step S2, the license plate image is acquired through a high-definition camera, and the image is preprocessed through an AI image recognition module to remove noise and enhance image quality; secondly, the license plate is located through edge detection, color segmentation and other technologies to separate it from the background; the overall architecture of Faster R-CNN is used, and the Inception ResNet_v2 network is integrated for feature extraction and final classification regression. Then, the separated license plate is segmented by characters, and each character is recognized to determine the vehicle information.

[0011] Furthermore, the management platform can notify the parking lot management personnel to take corresponding measures through the different information received.

[0012] The beneficial effects of the present invention are: 1. A dual-spectrum thermal imaging camera is installed above the parking space, and the AI ​​algorithm is used to detect whether the parking space is occupied or non-electric vehicles are parked illegally. If the parking space is occupied, the vehicle license plate and other information are recorded; 2. During the charging process of the electric vehicle, the dual-spectrum thermal imaging camera continuously detects temperature, flames, smoke and other dangerous conditions, and measures the current parking space vehicle temperature in real time through thermal imaging temperature detection, and continuously collects the vehicle charging data through the charging pile, thereby ensuring the safe monitoring of the charging process and being able to detect temperature anomalies and charging data anomalies in time, and to warn of possible safety hazards in advance, thereby ensuring the safety of charging facilities and vehicles. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Figure 1 is a management system topology diagram of the present invention; Figure 2 It is a schematic diagram of the structure of the charging pile controller; Figure 3 It is a schematic diagram of Yolo V8 real-time target detection algorithm. DETAILED DESCRIPTION

[0014] like Figures 1 to 3 As shown, in this embodiment, the present invention includes a parking lot charging pile management system and a management method. The parking lot charging pile management system includes a plurality of parking spaces 1 arranged in the parking lot, and a plurality of charging piles 2 arranged in the parking space 1, the charging pile 2 is used to charge the electric car in the parking space 1, the management system also includes a plurality of dual-spectrum thermal imaging cameras 3, a system servo processor 4, a charging pile controller 5 and a management platform 6, the dual-spectrum thermal imaging camera 3 is used to shoot the state of the parking space 1 and the thermal imaging temperature of the vehicle, the charging pile 2 is electrically connected to the charging pile controller 5, the dual-spectrum thermal imaging camera 3 and the charging pile controller 5 are both electrically connected to the system servo processor 4, the system servo processor 4 is signal-connected to the management platform 6, the charging pile controller 5 includes an audible and visual alarm 51, the charging pile controller 5 is used to collect charging data during the charging process of the charging pile 2, and set a safe charging data threshold. If the actual charging data exceeds the set threshold, an alarm is issued through the audible and visual alarm 51, and the information is wirelessly transmitted to the management platform 6 through the system servo processor 4.

[0015] In this embodiment, the dual-spectrum thermal imaging camera 3 includes a high-definition camera and an infrared thermal imaging monitor. The high-definition camera is integrated with an AI image recognition module. The high-definition camera is used to capture the status of the parking space 1 as being idle or parked with a vehicle. When capturing a parking space with a vehicle parked, the license plate captured at the same time is determined by the AI ​​image recognition module to determine whether the charging parking space is occupied by a non-electric vehicle; the infrared thermal imaging monitor is used to capture the battery temperature status during the charging process of the electric vehicle.

[0016] In this embodiment, the charging pile controller 5 includes a box body 52 and a cover body 53 adapted to the box body 52, and a control module 54, a switching power supply 55 and an air switch 56 are arranged in the box body 52. ​​The sound and light alarm 51 is a plurality of LED buzzers installed on the cover body 53, and the LED buzzer, the switching power supply 55 and the air switch 56 are all electrically connected to the control module 54.

[0017] In this embodiment, an electric gate is installed at the entrance of the parking lot, and a display is provided at the electric gate. The electric gate and the display are electrically connected to the system servo processor 4. Each parking space 1 has a corresponding number and positioning coordinates. The display is used to display the number of vacant parking spaces in parking space 1 and the number of vacant electric vehicle charging spaces, as well as to display a parking space map of the parking lot and mark the locations of vacant parking spaces.

[0018] In this embodiment, the management system also includes a reservation function module and a ground lock located on the charging parking space. The car owner reserves a parking space through the reservation function of the mobile terminal. When entering the parking lot, after the reserved license plate is recognized, the location of the reserved parking space is displayed on the display. When the vehicle enters the reserved parking space, the high-definition camera captures the corresponding reserved license plate and controls the ground lock to descend and unlock.

[0019] In this embodiment, the parking space 1 includes one or both of a car parking space 11 and a battery vehicle parking space 12 .

[0020] In this embodiment, a management method including the parking lot charging pile management system includes the following steps: S1. The status of parking space 1 in the parking lot is captured by a high-definition camera, and the number of free parking spaces 1, the number of charging spaces in parking space 1, and the coordinate information of parking space 1 are sent to the management platform 6; S2. The vehicle enters the parking lot according to the number and coordinate position of the displayed free parking spaces 1, and parks in parking space 1. If the parking space 1 is a charging space, the high-definition camera captures the license plate number, and uses the Yolo V8 real-time target detection algorithm to determine whether the vehicle is an electric vehicle. If the vehicle is not an electric vehicle, the information of the non-electric vehicle occupying the electric parking space is sent to the management platform 6, wherein the YoloV8 algorithm specifically adopts the Car-Yolo algorithm, which is an advanced real-time target detection algorithm that inherits the efficiency and lightness of the YOLO series, and has significant improvements in speed, accuracy and flexibility. It is particularly suitable for applications in complex scenes, and its optimized model architecture can provide better detection effects under complex conditions such as low light or occlusion. Car-Yolo's lightweight design and efficient inference speed are suitable for real-time detection on cameras or embedded devices, and the embedded object tracking technology therein helps the model perform better in real-time detection, and its multi-category high-precision detection capability enables it to perform well in distinguishing different types of vehicles or obstacles; S3. If it is determined that the vehicle is a tram, the charging pile 2 of the charging parking space is further determined whether the tram is recharged. If the charging pile 2 has no charging information, the information of the tram occupying the charging electric parking space in the non-charging state is sent to the management platform 6; S4. If the charging pile 2 enters the charging state, the temperature status information of the electric car is detected in real time by the infrared thermal imaging monitor, and the charging pile 2 uploads the charging data to the charging pile controller 5. If the battery temperature exceeds the set temperature threshold or the actual charging data exceeds the set safe charging data threshold, the sound and light alarm 51 warns, and the information is wirelessly transmitted to the management platform 6 through the system servo processor 4; S5. If the battery temperature and charging data are lower than the set threshold during the charging process of the electric vehicle, the charging state will be maintained. When the battery is fully charged, the full charge information will be sent to the owner's mobile phone through the management platform 6, and the timing will start at the same time. If the timing after charging exceeds the set time, the information that the electric vehicle has exceeded the time limit for occupying the charging parking space will be sent to the management platform 6.

[0021] In this embodiment, a convolutional block attention module is introduced on the basis of the Yolo V8 real-time target detection algorithm in step S2. The convolutional block attention module is also called CBAM. CBAM is a lightweight and efficient attention mechanism that combines channel attention and spatial attention, so that the model can selectively focus on important micro-channels and spatial positions. Through the introduction of CBAM, the model is enhanced in the extraction of key micro-channels, especially when detecting small vehicles, motorcycles and partially obscured vehicles in parking spaces, it can more accurately capture their features, reduce missed detection, and improve robustness in complex scenes; In this embodiment, more specifically, taking the Macau parking lot as an example, Macau single-plate private cars usually use the letter "M" plus an English letter and a group of 4 digits (M*-**-**). In 2021, the "A" prefix license plate number segment (A*-**-**) will be issued. Macau private car license plates are generally "black background and white characters" style, and tax-free vehicles such as new energy vehicles use black background and yellow characters style. The "Yue Z" license plate is a license plate code specially set up by Guangdong Province for vehicles entering and leaving the mainland from Hong Kong and Macao. It is usually composed of a four-digit number plus the suffix "Hong Kong" or "Macao" Chinese characters (such as: Yue ZXXXX Hong Kong / Yue ZXXXX Macao). Such license plates are generally used together with local private license plates in Hong Kong and Macao. In step S2, the license plate image is obtained through a high-definition camera, and the image is pre-processed by an AI image recognition module to remove noise and enhance image quality. Secondly, the license plate is located and separated from the background through edge detection, color segmentation and other technologies. The overall architecture of Faster R-CNN is used, and the Inception ResNet_v2 network is integrated for feature extraction and final classification regression. Then, the separated license plate is segmented and each character is recognized to determine the vehicle information. In the license plate recognition of electric vehicles, it is mainly achieved through color, three-dimensional numbers, etc. In addition, there are the following technical difficulties in Macau parking lots: the license plates of Macau and Hong Kong are very different, and there are cases where the same car has multiple license plates; data collection is difficult, and data consistency requirements are high; small data samples are a challenge for algorithms. The solutions are: 1. Obtain as much license plate information data as possible through multiple channels in a legal and compliant manner; 2. Select a suitable model for machine learning training for this situation; 3. Set the license plate priority, and give priority to Macau license plates when multiple license plates are identified.

[0022] In this embodiment, the management platform 6 can notify the parking lot management personnel to perform corresponding processing through the received different information.

[0023] In summary, the present invention has the following advantages: 1. Innovative detection method integrating multiple technologies The innovative combination of camera AI image recognition, infrared thermal imaging temperature monitoring and charging pile data collection has achieved all-round and high-precision detection of the status of charging parking spaces, breaking through the limitations of traditional single detection methods.

[0024] 2. Temperature-based safety warning mechanism The introduction of infrared thermal imaging to monitor vehicle battery temperature is not only used for conventional environmental perception, but also innovatively uses temperature data for safety warnings to detect potential safety hazards such as overheating in advance. This is a relatively novel application in charging parking space management.

[0025] 3. Intelligently identify the occupancy of vehicles that are not charged By comprehensively analyzing a variety of data, it is possible to intelligently determine when a vehicle is occupying a parking space but not charging, and handle it in a timely manner, effectively improving the resource utilization rate of the charging parking space. This precise judgment method is innovative among similar solutions.

[0026] 4. Dynamic optimization and self-learning capabilities The system can self-learn and optimize based on long-term accumulated data, such as continuously improving the accuracy of the AI ​​model's judgment on parking space occupancy and adjusting the time threshold for when a car is not charged, etc., to adapt to different usage scenarios and changing needs. Although the embodiments of the present invention are described with practical solutions, they do not constitute limitations on the meaning of the present invention. For those skilled in the art, it is obvious to modify the implementation scheme and combine it with other solutions based on this description.

Claims

1. A parking lot charging pile management system, comprising a plurality of parking spaces (1) arranged in the parking lot, and a plurality of charging piles (2) arranged in the parking spaces (1), wherein the charging piles (2) are used to charge electric vehicles in the parking spaces (1), and characterized in that: The management system further comprises a plurality of dual-spectrum thermal imaging cameras (3), a system servo processor (4), a charging pile controller (5) and a management platform (6). The dual-spectrum thermal imaging camera (3) is used to photograph the status of the parking space (1) and the thermal imaging temperature of the vehicle. The charging pile (2) is electrically connected to the charging pile controller (5). The dual-spectrum thermal imaging camera (3) and the charging pile controller (5) are both electrically connected to the system servo processor (4). The system servo processor (4) is signal-connected to the management platform (6). The charging pile controller (5) comprises an audible and visual alarm (51). The charging pile controller (5) is used to collect charging data of the charging pile (2) during charging and to set a safe charging data threshold. If the actual charging data exceeds the set threshold, an audible and visual alarm (51) is used to issue an alarm, and the information is wirelessly transmitted to the management platform (6) through the system servo processor (4).

2. A parking lot charging pile management system according to claim 1, characterized in that: The dual-spectrum thermal imaging camera (3) comprises a high-definition camera and an infrared thermal imaging monitor. The high-definition camera is integrated with an AI image recognition module. The high-definition camera is used to photograph the parking space (1) to determine whether it is vacant or has a vehicle parked therein. When photographing a parking space with a vehicle parked therein, the license plate photographed at the same time is used to determine whether the parking space is occupied by a non-electric vehicle through the AI ​​image recognition module. The infrared thermal imaging monitor is used to photograph the battery temperature status during the charging process of the electric vehicle.

3. A parking lot charging pile management system according to claim 3, characterized in that: The charging pile controller (5) comprises a box (52) and a cover (53) adapted to the box (52); a control module (54), a switch power supply (55) and an air switch (56) are arranged in the box (52); the sound and light warning device (51) is a plurality of LED buzzers installed on the cover (53); the LED buzzer, the switch power supply (55) and the air switch (56) are all electrically connected to the control module (54).

4. A parking lot charging pile management system according to claim 1, characterized in that: An electric gate is installed at the entrance of the parking lot, and a display is arranged at the electric gate. The electric gate and the display are electrically connected to the system servo processor (4). Each parking space (1) has a corresponding number and positioning coordinates. The display is used to display the number of vacant parking spaces (1) and the number of vacant electric vehicle charging spaces, as well as to display a parking space map of the parking lot and mark the locations of vacant parking spaces.

5. A parking lot charging pile management system according to claim 1, characterized in that: The parking space (1) comprises one or both of a car parking space (11) and a battery vehicle parking space (12).

6. A management method for a parking lot charging pile management system comprising any one of claims 1 to 3, characterized in that: It includes the following steps: S1. Using a high-definition camera to capture the status of parking spaces (1) in the parking lot, and sending the number of free parking spaces (1), the number of charging spaces in the parking spaces (1), and the coordinate information of the parking spaces (1) to the management platform (6); S2. The vehicle enters the parking lot according to the number and coordinate position of the displayed free parking spaces (1) and parks in the parking space (1). If the parking space (1) is a charging space, the high-definition camera captures the license plate number and uses the Yolo real-time target detection algorithm to determine whether the vehicle is an electric vehicle. If the vehicle is not an electric vehicle, the information about the non-electric vehicle occupying the electric parking space is sent to the management platform (6); S3. If it is determined that the vehicle is an electric car, then whether the electric car is recharging is further determined by whether the charging pile (2) of the charging parking space is working. If the charging pile (2) has no charging information, the information that the electric car is not charging and occupies the charging electric parking space is sent to the management platform (6); S4. If the charging pile (2) enters the charging state, the temperature status information of the electric vehicle is detected in real time by the infrared thermal imaging monitor, and the charging pile (2) uploads the charging data to the charging pile controller (5). If the battery temperature exceeds the set temperature threshold or the actual charging data exceeds the set safe charging data threshold, the sound and light alarm (51) issues an alarm, and the information is wirelessly transmitted to the management platform (6) through the system servo processor (4); S5. If the battery temperature and charging data are both lower than the set threshold during the charging process of the electric vehicle, the charging state is maintained. When the battery is fully charged, the management platform (6) sends a full charge message to the owner's mobile phone and starts counting at the same time. If the time after charging exceeds the set time, the information that the electric vehicle has exceeded the time limit for occupying the charging parking space is sent to the management platform (6).

7. The management method according to claim 6, characterized in that: Based on the Yolo real-time target detection algorithm in step S2, a convolutional block attention module is introduced to obtain the license plate image through a high-definition camera, and the image is preprocessed through an AI image recognition module to remove noise and enhance image quality; secondly, the license plate is located through edge detection, color segmentation and other technologies to separate it from the background; the overall architecture of Faster R-CNN is used, and the InceptionResNet_v2 network is integrated for feature extraction and final classification regression. Then, the separated license plate is segmented into characters, and each character is recognized to determine the vehicle information.

8. The management method according to claim 6, characterized in that: The management platform (6) can notify the parking lot management personnel to carry out corresponding processing based on the different information received.

Images