Public place electric bicycle charging management system based on AI image recognition

Abstract

The utility model discloses a public place electric bicycle charging management system based on AI image recognition belongs to electric vehicle charging control technical field. Including power module, AI processing terminal, fire alarm system and a plurality of charging unit, AI processing terminal respectively with power module, fire alarm system and a plurality of charging unit electric connection, AI processing terminal control every charging unit's opening and closing, charging unit includes monitoring camera, metering module and charging socket, fire alarm system includes alarm and fire sprinkler head, fire sprinkler head. The utility model discloses through to electric vehicle and electric vehicle charger's image recognition, carries out monitoring to electric vehicle charging state, combines the fire alarm technology based on image recognition simultaneously, instantaneously cuts off when electric vehicle appears fire, short circuit overload etc. situation, fully guarantees the safety of electric bicycle in the charging process, reduces the personal, property loss caused by fire hazard.

Description

Technical Field

[0001] This utility model relates to the field of electric vehicle charging control technology, specifically to a public place electric bicycle charging management system based on AI image recognition. Background Technology

[0002] In recent years, the proportion of fires caused by electric vehicle charging has been gradually increasing. Data shows that 80% of electric vehicle fires occur while charging, and 90% of injuries and fatalities from electric vehicle fires are due to vehicles being placed in lobbies, hallways, or passageways. Experiments show that the initial flame temperature of an electric vehicle fire reaches 126°C, and the smoke temperature is 43°C. Approximately 7 minutes later, multiple nearby electric vehicles are engulfed in flames, with the flame temperature reaching 1105.9°C and the smoke temperature reaching 500°C.

[0003] Existing electric vehicle charging sheds on the market generally have densely packed vehicles and lack intelligent devices for fire early warning and handling. If an electric vehicle catches fire while charging, the close proximity between adjacent vehicles can cause the flames to spread, potentially resulting in personal injury and economic losses. If a charging management system is adopted to monitor the charging status of electric vehicles and to provide alarms and pre-handling for fire accidents, the personal injury and property losses caused by fire hazards can be reduced. Utility Model Content

[0004] To address the aforementioned issues, this invention provides a public place electric bicycle charging management system based on AI image recognition.

[0005] This utility model is achieved through the following technical solution:

[0006] A public place electric bicycle charging management system based on AI image recognition includes a power module, an AI processing terminal, a fire alarm system and several charging units, wherein the AI ​​processing terminal is electrically connected to the power module, the fire alarm system and several charging units respectively.

[0007] The power module provides voltage to the charging management system;

[0008] The AI ​​processing terminal controls the opening and closing of each charging unit;

[0009] The charging unit includes a monitoring camera, a metering module, and a charging socket. The charging socket has a built-in relay, which receives instructions from the AI ​​processing terminal to control the opening and closing of the socket. The metering module is installed inside the charging socket and measures electrical parameters in real time and uploads them to the AI ​​processing terminal. The monitoring camera captures images of the electric vehicle's charging status and the smoke / fire status and transmits the images to the AI ​​processing terminal.

[0010] The fire alarm system includes an alarm and a fire sprinkler head.

[0011] Preferably, it also includes a mechanical button switch, which transmits signals to the AI ​​processing terminal to control the activation of each charging unit.

[0012] Preferably, the AI ​​processing terminal further includes a wireless communication module, and the AI ​​processing terminal communicates with the mobile terminal through the wireless communication module.

[0013] Preferably, the power module is model XGJ-QT-101.

[0014] Preferably, the AI ​​processing terminal is model FCU3001.

[0015] Preferably, the surveillance camera is model DS-2CD2125FV2.

[0016] Preferably, the wireless communication module adopts one of Bluetooth module, Wi-Fi module, 2G / 3G module and ZigBee module.

[0017] Compared with the prior art, the beneficial effects of this utility model are:

[0018] This invention monitors the charging status of electric vehicles by recognizing images of the electric vehicle and its charger. It also incorporates image recognition-based smoke and fire alarm technology to immediately cut off power when the electric vehicle is exposed to smoke, fire, short circuit, or overload, thus ensuring the safety of electric bicycles during charging and reducing personal injury and property damage caused by fire hazards. Attached Figure Description

[0019] Figure 1 This is a schematic diagram of the structure of the carport in this utility model;

[0020] Figure 2 This is a schematic diagram of the structure of the management system of this utility model.

[0021] In the picture: 1. Rain shelter; 2. Support frame; 3. Fire sprinkler head; 4. Surveillance camera; 5. Controller; 6. Charging shed; 7. Electric vehicle. Detailed Implementation

[0022] The present invention will be further described below with reference to the accompanying drawings:

[0023] As shown in the attached diagram of the instruction manual. Figure 1-2 As shown, a public place electric bicycle charging management system based on AI image recognition includes a power module, an AI processing terminal, a fire alarm system and several charging units. The AI ​​processing terminal is electrically connected to the power module, the fire alarm system and several charging units respectively.

[0024] The power module provides voltage to the charging management system;

[0025] The AI ​​processing terminal controls the opening and closing of each charging unit;

[0026] The charging unit includes a monitoring camera, a metering module, and a charging socket. The charging socket has a built-in relay that receives instructions from the AI ​​processing terminal to control the opening and closing of the socket. The metering module is located inside the charging socket and measures electrical parameters in real time and uploads them to the AI ​​processing terminal. The monitoring camera captures images of the electric vehicle's charging status and the smoke / fire status and transmits the images to the AI ​​processing terminal.

[0027] A fire alarm system includes alarms and fire sprinkler heads.

[0028] The charging management system is installed in the charging shed for use. The monitoring camera 4 and the fire sprinkler head 3 are installed on the top of the charging shed 6. Each charging unit is equipped with a charging socket, and one charging unit is used to charge one electric vehicle 7.

[0029] It also includes mechanical button switches, which transmit signals to the AI ​​processing terminal to control the activation of each charging unit.

[0030] The AI ​​processing terminal also includes a wireless communication module, which connects the AI ​​processing terminal to the mobile terminal. Users can remotely operate the charging unit via a mobile app to turn it on and off.

[0031] The power module model is XGJ-QT-101, which is used to provide voltage for the charging management system. Other modules with the same function can also be selected.

[0032] The AI ​​processing terminal is model FCU3001. FCU3001 is a high-performance AI edge computing terminal developed based on the NVIDIA Jetson Xavier NX processor.

[0033] The surveillance camera, model DS-2CD2125FV2, is a 2-megapixel infrared network dome camera. It features IP67 dust and water resistance, an IK10 vandal-proof rating, and is suitable for harsh environments.

[0034] The wireless communication module adopts one of the following: Bluetooth module, Wi-Fi module, 2G / 3G module, and ZigBee module.

[0035] The control system works as follows: The AI ​​processing terminal scans each charging unit sequentially. When a charging unit is activated and the electric bicycle is charging, the AI ​​processing terminal activates the monitoring camera and metering module to collect data from that charging unit, including measuring electrical parameters, taking photos, and performing image recognition processing. This determines the charging status of the electric bicycle in that unit and whether there is any smoke or fire, and takes corresponding actions. When the charging algorithm determines that the electric bicycle is fully charged, it disconnects the power supply to the charging unit's socket. When image recognition detects smoke or fire, it disconnects the power supply to all sockets and activates the fire alarm system. The alarm sounds, and fire sprinklers extinguish the fire on the electric bicycle. Based on the image information returned by the monitoring camera, if a fire is detected, the fire alarm system is activated. The AI ​​processing terminal also includes remote communication capabilities, allowing it to be connected to a mobile app for control, billing systems, etc.

[0036] The surveillance camera is only used to capture images of the electric vehicle charging status and smoke / fire activity. It does not require video recording or storage, nor does it need to be constantly on. When the corresponding charging unit is open and an electric vehicle is charging, the camera is activated by an AI-processed terminal command, takes a photo, transmits it to the terminal, and then closes, awaiting the next activation command.

[0037] The charging socket has a built-in relay that receives commands from an AI processing terminal to control its on / off state. It also features a built-in metering module that measures electrical parameters such as voltage, current, power, and battery level in real time and uploads this data to the terminal. These measured parameters help determine the charging status of the electric vehicle. The socket also features a protective power cut-off mechanism in case of overvoltage, overcurrent, or overload, with a fast response time. Furthermore, the battery level measurement function can be integrated with billing functionality.

[0038] The image recognition algorithm is based on YOLO11 and combines deep learning and computer vision technologies to achieve fast and accurate object recognition.

[0039] The algorithm involved in the control system of this utility model includes two datasets. One dataset contains the charging status of electric vehicles, including electric vehicles, electric vehicle chargers, indicator lights, digital display tubes, etc., used to determine the charging status of electric vehicles. The other dataset is a smoke and fire detection dataset, used to determine fire events.

[0040] The AI ​​processing terminal performs image recognition twice on each received image to determine the charging status and the fireworks event, respectively.

[0041] Charging Completion Algorithm: After charging begins, the charger indicator lights on the electric bicycle are monitored via image recognition. The charging status is determined based on a combination of different indicator light combinations, real-time current, and power. For example, a red light indicates charging in progress, a green, blue, or other color light indicates charging complete, and the indicator light being off indicates either charging complete or a malfunction. If the charger has two indicator lights, initially only the red light will illuminate, followed by a green light, indicating charging complete; or initially both red and green lights will illuminate, followed by the red light turning off or turning green, indicating charging complete. If multiple indicator lights are present, all indicator lights will illuminate, indicating charging complete.

[0042] When image recognition and current power parameters meet the charging completion criteria, the power to the socket is turned off after a certain delay. After each charging cycle, the image and electrical parameter data are uploaded to the cloud to aid in the training of new models.

[0043] Smoke and fire detection algorithm: The smoke and fire detection algorithm is based on the smoke and fire dataset in the YOLO image recognition algorithm. To prevent false alarms, when the feature correlation is not high, multiple photos will be taken for judgment. Once smoke and fire are detected, the power will be cut off immediately, the smoke and fire image will be uploaded to the cloud for later investigation, and the fire alarm system will be activated.

[0044] In summary, the above description is merely a preferred embodiment of the present utility model and is not intended to limit the scope of the present utility model. All equivalent variations and modifications made in accordance with the shape, structure, features and spirit described in the claims of the present utility model should be included within the scope of the claims of the present utility model.

Claims

1. An AI image recognition-based public place electric bicycle charging management system, characterized in that: It includes a power module, an AI processing terminal, a fire alarm system, and several charging units, wherein the AI ​​processing terminal is electrically connected to the power module, the fire alarm system, and several charging units respectively; The power module provides voltage to the charging management system; The AI ​​processing terminal controls the opening and closing of each charging unit; The charging unit includes a monitoring camera, a metering module, and a charging socket. The charging socket has a built-in relay, which receives instructions from the AI ​​processing terminal to control the opening and closing of the socket. The metering module is installed inside the charging socket and measures electrical parameters in real time and uploads them to the AI ​​processing terminal. The monitoring camera captures images of the electric vehicle's charging status and the smoke / fire status and transmits the images to the AI ​​processing terminal. The fire alarm system includes an alarm and a fire sprinkler head.

2. The public place electric bicycle charging management system based on AI image recognition according to claim 1, characterized in that: It also includes mechanical button switches, which transmit signals to the AI ​​processing terminal to control the activation of each charging unit. 3.The public place electric bicycle charging management system based on AI image recognition according to claim 1, characterized in that: The AI ​​processing terminal also includes a wireless communication module, which communicates with the mobile terminal. 4.The public place electric bicycle charging management system based on AI image recognition of claim 1, characterized in that: The power module is model XGJ-QT-101. 5.The public place electric bicycle charging management system based on AI image recognition according to claim 1, characterized in that: The AI ​​processing terminal is model FCU3001. 6.The public place electric bicycle charging management system based on AI image recognition according to claim 1, characterized in that: The surveillance camera is model DS-2CD2125FV2. 7.The public place electric bicycle charging management system based on AI image recognition of claim 3, characterized in that: The wireless communication module adopts one of the following: Bluetooth module, Wi-Fi module, 2G / 3G module, and ZigBee module.

Images