Intelligent gas station management system and method based on AI Internet of Things

Abstract

The invention relates to the technical field of intelligent gas station management, in particular to an intelligent gas station management system and method based on the AI Internet of Things. Firstly, the running state of key equipment is real-time through Internet of Things equipment, potential safety hazards are recognized, timely processing is carried out through an emergency plan, and in the refueling process, the system can automatically recognize a vehicle, obtain user data associated with the vehicle and generate temporary information credentials for the vehicle; according to the method and the system, the user data is ensured not to be out of order in system data transmission with relatively high concurrency, then the oil product is intelligently recommended, the use experience of a customer is greatly improved, and the system analyzes the consumption data of the user and performs personalized oil product recommendation and preferential pushing, so that the user stickiness is improved. By combining AI and Internet of Things technologies, the intelligent gas station management system can realize equipment real-time monitoring, oil fee payment and safety emergency response, so that the operation efficiency of the intelligent gas station is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of smart gas station management, and in particular to a smart gas station management system and method based on AI Internet of Things. Background Art

[0002] With the rapid development of artificial intelligence and Internet of Things technologies, the management model of traditional gas stations faces problems such as inefficiency, excessive energy consumption, safety hazards and insufficient customer experience. As a new generation of gas station management system, smart gas stations integrate advanced technologies such as AI, Internet of Things, and big data analysis to improve the intelligent management level of gas stations, optimize energy scheduling, and improve operational efficiency. At the same time, it realizes fuel payment and real-time safety monitoring, providing users with a more convenient, safe and efficient service experience.

[0003] A Chinese invention patent with publication number CN112330217A discloses a smart gas station integrated management system, including: a central processing unit, a site safety management module, an intelligent patrol management module, an intelligent payment management module and an intelligent delivery management module. The central processing unit is bidirectionally electrically connected to the site safety management module, the intelligent patrol management module, the intelligent payment management module and the intelligent delivery management module respectively. The present invention can effectively identify smoke or flames in a gas station, improve the safety management and control of the station, and avoid accidents caused by vehicle congestion and illegal parking. It centrally controls refueling behavior and personnel behavior, avoids the occurrence of dangers caused by refueling operation errors and staff behavior errors, improves patrol efficiency, effectively saves customers' time in issuing invoices when refueling, and can save customers' time in purchasing goods, avoiding vehicle congestion in gas stations.

[0004] At present, it is difficult to monitor the status of equipment in smart gas stations in real time, fault warnings are delayed, safety hazards cannot be identified in time, the refueling and payment processes are inefficient, customer service lacks personalization, operational data is scattered and difficult to analyze effectively, and during the system's high-concurrency data transmission process, vehicle-related user data may experience data transmission disorder problems, which will directly affect the normal operation of the system and lead to inaccurate data processing during the refueling process. It may happen that the refueling amount, payment information or user fuel preferences cannot be synchronized in time, thereby affecting the accuracy of the system's intelligent refueling and user experience. Summary of the invention

[0005] The purpose of the present invention is to address the problems existing in the background technology and propose a smart gas station management system and method based on AI Internet of Things.

[0006] The technical solution of the present invention is a smart gas station management method based on AI Internet of Things, which includes the following specific implementation steps: S1. During the initialization phase, several IoT sensors are installed at the core of the gas station to collect real-time status data of the gas station, and high-definition cameras are used to conduct real-time video monitoring of the gas station area and entrances and exits; S2. In the intelligent safety management stage of smart gas stations, IoT sensors and high-definition cameras collect real-time status data and environmental monitoring videos of gas stations and transmit them to the data analysis module. The data analysis module then analyzes and processes the data. If an abnormality is found, an alarm is sent to the safety management and response module. The safety management and response module automatically activates the emergency plan after identifying the alarm type. S3, in the smart refueling stage of the smart gas station, when a vehicle enters the gas station, a high-definition camera collects vehicle images and transmits them to the data analysis module, which identifies the license plate, retrieves user data associated with the vehicle, and transmits it to the smart refueling management module, which links the refueling equipment, recommends oil products and refueling quantities based on user preferences and historical records, and prompts the user to pay the fuel fee after refueling is completed; S4. In the personalized marketing stage of smart gas stations, user refueling and consumption data are collected and transmitted to the data analysis module for analysis and processing, static and dynamic tags are generated, and users are grouped using the K-means clustering algorithm. The collaborative filtering algorithm is then used to analyze user preferences. The data analysis results are transmitted to the personalized service module to generate recommended content, which is then pushed through mobile apps, advertising screens, and text messages.

[0007] Preferably, the specific implementation steps of S2 are as follows: S21, the IoT sensor collects oil tank level data, temperature data and pressure data, the camera collects environmental monitoring video data in real time, and transmits the above oil tank level data, temperature data, pressure data and environmental monitoring video data to the data analysis module; S22, the data analysis module calls the constructed risk detection integrated model to perform concurrent processing and analysis on the received data. If an abnormality is detected, an alarm instruction is immediately sent to the security management and response module; S23. Once the safety management and response module receives the alarm instruction, the safety management and response module identifies the alarm type and immediately and automatically activates the emergency plan, remotely shuts down the refueling equipment, activates the fire protection system, notifies the emergency management personnel to handle it, and then automatically records the emergency handling process and generates an emergency handling report.

[0008] Preferably, the specific implementation steps of S3 are as follows: S31. When a vehicle enters a gas station, a high-definition camera automatically collects vehicle images and transmits them to a data analysis module; S32, the data analysis module calls the vehicle recognition model, automatically detects the vehicle license plate, and recognizes the vehicle license plate number information through the OCR technology, and then the data analysis module automatically retrieves the user data associated with the vehicle license plate number information, and generates a temporary information credential for the user data associated with the vehicle license plate number information, and then transmits the user data associated with the vehicle and the temporary information credential to the intelligent refueling management module; The user data associated with the vehicle's license plate number information includes refueling preferences, weights of different oil products, and the latest historical refueling records; S33, when the vehicle stops at the refueling area, after the vehicle stops steadily, the intelligent refueling management module reviews the temporary information credentials of the user data associated with the license plate number information of the vehicle, confirms the credibility of the user data associated with the license plate number information of the vehicle, and then generates a recommended oil product and refueling amount, and sends the recommended oil product and refueling amount to the user; S34, the user confirms or modifies the recommended oil product and refueling amount, and the intelligent refueling management module instructs the refueling machine to automatically refuel; S35. During the refueling process, the intelligent refueling management module monitors the operating status data of the refueling equipment in real time through the Internet of Things sensor, and instructs the data analysis module to analyze the operating status of the equipment. If an abnormality is detected, the data analysis module immediately sends an emergency instruction to the intelligent refueling management module, and the intelligent refueling management module immediately stops refueling and sends an alarm instruction to the safety management and response module; S36. After refueling is completed, the intelligent refueling management module automatically calculates the total cost based on the actual refueling amount and the current oil price, and then generates a payment QR code; S37, the user scans the payment QR code to complete the payment. After the intelligent refueling management module detects that the user has completed the payment, the refueling machine is automatically turned off, and the user is prompted through a sound prompt and a display screen that the refueling is completed, and the user can drive away directly; S38. The intelligent refueling management module sends an oil level monitoring instruction to the intelligent monitoring module. The intelligent monitoring module collects the remaining oil level data in the oil tank in real time and transmits the remaining oil level data in the oil tank to the data analysis module. If the data analysis module detects that the remaining oil level is lower than the preset warning value, it automatically sends a replenishment and alarm instruction to the safety management and response module.

[0009] Preferably, the generation process of the temporary information credentials is as follows: S41. Define the finite cyclic group Z q * , and select Z q * A generator g of ; Among them, q is a large prime number, Z q *={1, 2, ..., q-1}; S42, converting the user data associated with the vehicle into a binary string to obtain datav j , and use the random function rand() to generate two integers α and β; Among them, 1≤α≤q-1, and gcd(α,q-1)=1, 1≤β≤q-1, and gcd(β,q-1)=1, α×β≠0 mod(q-1), gcd() is the greatest common divisor function; S43, calculate the first-level information credential Ic1: ; S44, calculate the secondary information credential Ic2 = (H (datav j )-CI G ×Ic1)×β mod (q-1); Where H is the hash function, CI G The temporary information credential generation code generated during the system initialization phase and satisfies 1<CI G <q-1, thereby defining the temporary information credential resolution code CI E : ; S45, calculate the third-level information credential Ic3 = α × β mod (q-1); S46. Generate temporary information credentials IC={Ic1, Ic2, Ic3}.

[0010] Preferably, the process of confirming the credibility of the user data associated with the license plate number information of the vehicle is as follows: S51. Calculate the audit factor generation parameter P f : ; S52, calculate the first-level audit factor Af1 = H (datav j )×P f mod (q-1); Among them, datav j is the binary string form of the user data associated with the vehicle; S53. Calculate the secondary audit factor Af2 = Ic1 × P f mod (q-1); S54. Calculate the audit factor A: ; Among them, CI E It is the temporary information credential parsing code; S55. If A=Ic1, it means that the user data associated with the vehicle is not out of order during the data transmission process, that is, the credibility of the user data associated with the license plate number information of the vehicle is confirmed.

[0011] Preferably, the specific implementation steps of S4 are as follows: S61, collecting the user's refueling behavior data and consumption data in real time, and transmitting the user's refueling behavior data and consumption data to a data analysis module; S62, the data analysis module processes the user's refueling behavior data and consumption data, and generates a set of static and dynamic tags for each user; S63, the data analysis module calls the personalized marketing model and uses the K-means clustering algorithm to classify users into behavior groups: A-type users, B-type users, and C-type users; S64. Based on the collaborative filtering algorithm, the data analysis module analyzes the historical behaviors of other user groups with similar consumption behaviors to the current user, infers the oil product or service preference of the current user, and obtains the user behavior analysis and grouping results; S65. The data analysis module transmits the user's behavior analysis and grouping results to the personalized service module, and the personalized service module generates personalized recommendation content: S66. The personalized service module pushes recommended content to users through mobile apps, advertising screens, and text messages, and records user feedback data. It uses the feedback data as a basis for adaptive optimization of the model and adjusts marketing strategies.

[0012] The technical solution of the present invention is a smart gas station management system based on AI Internet of Things, which is used to execute the above-mentioned smart gas station management method based on AI Internet of Things, including: Intelligent monitoring module, which integrates several IoT sensors and high-definition cameras to collect working status data within the gas station; The data analysis module is run on an AI algorithm box, which is an intelligent edge application server with a built-in domestically produced high-performance AI computing chip. The core is to use deep learning technology to customize the construction of fire detection models, vehicle identification models, customer behavior analysis models, and risk detection integration models; Intelligent refueling management module, responsible for intelligent refueling of vehicles; Personalized service module, providing customers with personalized recommendation services; The safety management and response module automatically activates the emergency plan according to the alarm instructions and notifies the emergency management personnel to intervene.

[0013] Compared with the prior art, the above technical solution of the present invention has the following beneficial technical effects: The present invention designs a smart gas station management system and method based on AI Internet of Things. First, through the deployment of Internet of Things devices, the operating status of key equipment such as oil tanks and gas dispensers is monitored in real time, and automatic alarms are issued under abnormal circumstances to ensure the normal operation of the equipment; secondly, the system monitors vehicles, personnel and the surrounding environment in real time, identifies potential safety hazards, and handles them in a timely manner through emergency plans; during the refueling process, the system can automatically identify the vehicle, obtain user data associated with the vehicle, and generate temporary information credentials for it, ensuring that user data will not be out of order in system data transmission with high concurrency, and then intelligently recommend oil products and realize the oil payment process, which greatly improves the customer's usage experience; at the same time, the system uses the user's consumption data to make personalized oil product recommendations and discount push notifications to enhance customer stickiness. By combining AI and Internet of Things technologies, the present invention can realize real-time equipment monitoring, energy efficiency optimization, oil payment and safety emergency response through the intelligent gas station management system, thereby improving operational efficiency, reducing costs and ensuring safety. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 This is an architecture diagram of a smart gas station management system based on AI Internet of Things proposed by the present invention; Figure 2 A flowchart of the implementation steps for the smart safety management phase of smart gas stations; Figure 3 A flowchart of the implementation steps for the smart refueling phase of a smart gas station; Figure 4 Flowchart of the implementation steps for the personalized marketing stage of smart gas stations. DETAILED DESCRIPTION

[0015] Embodiment 1, as Figure 1 As shown, the present invention proposes a smart gas station management system based on AI Internet of Things, including: an intelligent monitoring module, a data analysis module, an intelligent refueling management module, a personalized service module, and a safety management and response module.

[0016] The intelligent monitoring module integrates several IoT sensors and high-definition cameras to monitor various facilities in the gas station in real time and collect working status data in the gas station; Among them, the working status data in the gas station includes but is not limited to the amount of fuel, the duration of fueling, the tank level data, the temperature data, the pressure data, and the environmental monitoring video data; The data analysis module runs on an AI algorithm box, which is an intelligent edge application server with a built-in domestically produced high-performance AI computing chip and high-performance decoding and reasoning capabilities. The core is to use deep learning technology to customize and build a series of AI security models. The types of AI security models and their functions are shown in Table 1: Table 1 Security model types and their functions Model category Function Oil level monitoring model Safe Behavior Compare the tank level with the preset warning value; Vehicle Identification Model Intelligent refueling management Automatically identify vehicle information through license plate recognition technology and associate the owner's user data; Customer behavior analysis model Personalized service Intelligently detect user preferences and formulate personalized marketing plans Risk Detection Integrated Model Safe Behavior It integrates equipment status monitoring model, fire detection model, and personnel behavior detection model to analyze real-time monitoring video to determine whether there are safety hazards or violations (including but not limited to fire risks in gas stations, oil leaks, smoking, refueling without turning off the engine, personnel misoperation, and illegal intrusion) The intelligent refueling management module is responsible for intelligent refueling of vehicles, realizing functions including but not limited to automatic refueling billing and fuel payment, improving refueling efficiency and customer satisfaction; The personalized service module provides customers with personalized recommendation services, including but not limited to oil product recommendations and promotional information push, to improve user stickiness; The safety management and response module monitors and analyzes the safety risks of gas stations in real time, automatically identifies safety hazards, and automatically activates emergency plans based on risk levels, notifying emergency management personnel to intervene and ensure the safety of gas station operations.

[0017] Embodiment 2, a smart gas station management method based on AI Internet of Things proposed by the present invention is applied to a smart gas station management system based on AI Internet of Things in Embodiment 1, and its specific implementation steps are as follows: Initialization phase: S101. Install several IoT sensors at the core locations of the gas station (including but not limited to oil tanks, gas pumps, and gas station entrances) to collect real-time status data of the gas station; Among them, the real-time status data of the gas station includes but is not limited to the real-time status data of the gas station including the amount of fuel, the duration of fuel, the tank level data, the temperature data, and the pressure data; S102. Use high-definition cameras to conduct real-time video surveillance of the refueling area and entrances and exits to ensure that the behavior of vehicles and personnel can be captured; S103. Use low-power wide area network technologies including but not limited to 5G, LoRa, and NB-IoT to establish communication links between various IoT sensors, high-definition cameras, and data analysis modules.

[0018] Smart gas station intelligent safety management stage: In the intelligent safety management stage of smart gas stations, IoT sensors and high-definition cameras collect oil tank data and environmental monitoring videos, and transmit them to the data analysis module. The AI ​​security model is then used to concurrently process the data. If an abnormality is found, an alarm is sent to the safety management and response module. After the safety management and response module identifies the alarm type, it automatically starts the emergency plan, remotely shuts down the equipment and notifies the management personnel, and finally records and generates a report, such as Figure 2 As shown in the figure, the specific implementation steps of the intelligent safety management stage of smart gas stations are as follows: S201, a number of IoT sensors collect oil tank level data, temperature data and pressure data, and a high-definition camera collects environmental monitoring video data in real time, and transmits the above oil tank level data, temperature data, pressure data and environmental monitoring video data to a data analysis module; S202. The data analysis module calls the customized AI security model to perform concurrent processing and analysis on the received data: If the risk detection integrated model detects abnormal equipment status data (including but not limited to fuel dispenser failure and oil tank leakage), the data analysis module immediately sends an alarm instruction to the safety management and response module; If the risk detection integrated model detects an abnormal environment or abnormal behavior of personnel (including but not limited to fire, smoking, long-term stay or illegal intrusion), the data analysis module immediately sends an alarm notification to the security management and response module, and sends relevant information and alarm instructions to the security management and response module; If the oil volume monitoring model detects that the oil tank level is lower than the preset warning value, the data analysis module automatically sends a replenishment or alarm instruction to the safety management and response module; S203. Once the safety management and response module receives the alarm instruction, the safety management and response module identifies the alarm type and immediately and automatically starts the emergency plan, remotely shuts down the refueling equipment, starts the fire protection system, notifies the emergency management personnel to handle it, and then automatically records the emergency handling process and generates an emergency handling report; Among them, the alarm types include but are not limited to fire, oil leakage, equipment failure, oil shortage or overflow.

[0019] Smart gas station intelligent refueling stage: In the smart refueling stage of smart gas stations, when a vehicle enters a gas station, a high-definition camera collects vehicle images and transmits them to the data analysis module, recognizes the license plate, retrieves user data, and transmits them to the smart refueling management module. The smart refueling management module links the refueling equipment, recommends oil products and refueling quantities based on user preferences and historical records, monitors the equipment status in real time and handles abnormalities. After refueling is completed, the user is prompted to pay the fuel fee. Figure 3 As shown in the figure, the specific implementation steps of the smart gas station intelligent refueling stage are as follows: S301, when a vehicle enters a gas station, a high-definition camera automatically collects vehicle images and transmits them to a data analysis module; S302, the data analysis module calls the vehicle recognition model, adopts the YOLOv7 (You Only Look Once version 7) model, automatically detects the vehicle license plate, and recognizes the vehicle license plate number information through OCR (Optical Character Recognition) technology, and then the data analysis module automatically retrieves the user data associated with the vehicle license plate number information, including the refueling preferences {p1, p2, ..., p i , ..., p n}, weights of different oil products {w1, w2, ..., wi , ..., w n}、Latest historical refueling record C history and transmit the user data associated with the vehicle to the intelligent refueling management module; Among them, p i Indicates the oil type of any oil preference of the vehicle, w i Indicates p i The corresponding weight, i is the sequence number, i=1, 2, ..., n; If the data analysis module fails to retrieve the user data associated with the vehicle's license plate number information, the user is a new user, and the data analysis module prompts the user to complete the binding of the license plate and the user account through the APP or the self-service terminal in the station; In order to avoid data disorder in the user data associated with the vehicle during the system data transmission process with high concurrency and ensure the accuracy of intelligent refueling of the intelligent refueling management module, the data analysis module generates a temporary information credential for the vehicle information and the user data associated with the vehicle. The generation process is as follows: S3021. Define the finite cyclic group Z q * , and select Z q * A generator g of ; Among them, q is a large prime number, Z q * ={1, 2, ..., q-1}; S3022, convert the user data associated with the vehicle into a binary string to obtain datav j , and use the random function rand() to generate two integers α and β; Among them, 1≤α≤q-1, and gcd(α,q-1)=1, 1≤β≤q-1, and gcd(β,q-1)=1, α×β≠0 mod(q-1), gcd() is the greatest common divisor function; S3023. Calculate the first-level information credential Ic1: ; S3024, calculate the secondary information credential Ic2 = (H (datav j )-CI G ×Ic1)×β mod (q-1); Where H is the hash function, CI G The temporary information credential generation code generated during the system initialization phase and satisfies 1<CI G <q-1, thereby defining the temporary information credential resolution code CI E : ; S3025, calculate the third-level information credential Ic3 = α × β mod (q-1); S3026. Generate temporary information credential IC={Ic1, Ic2, Ic3}; S303: When the vehicle stops at the refueling area, the intelligent refueling management module is linked with the refueling equipment through the Internet of Things network. After the vehicle stops steadily, the distance sensor on the refueling machine detects the vehicle position, automatically activates the refueling machine, and automatically recommends oil products and refueling quantities. The specific implementation process is as follows: S3031. The intelligent refueling management module receives the temporary information credential IC={Ic1, Ic2, Ic3} of the user data associated with the vehicle, and implements the following verification process: (1) Calculate the audit factor generation parameter P f : ; (2) Calculate the first-level audit factor Af1 = H(datav j )×P f mod (q-1); (3) Calculate the secondary audit factor Af2 = Ic1 × P f mod (q-1); (4) Calculate the audit factor A: ; (5) If A = Ic1, it means that the user data associated with the vehicle is not out of order during the data transmission process, which means that the user data associated with the vehicle is credible, and the refueling preference {p1, p2, ..., p i , ..., p n}, weights of different oil products {w1, w2, ..., w i , ..., w n}, Latest historical refueling record C history}; S3032, the intelligent refueling management module is based on the vehicle's oil product preference and the weights of different oil products {w1, w2, ..., w i , ..., w n}, automatically recommend oil models ; S3033, based on the latest historical refueling record C history , and read the maximum capacity C of the vehicle's fuel tank max 、The current remaining oil volume in the tank C current , automatically recommends fueling amount Q recommend =min(|C max -C current |, C history ), and then the user confirms it through the mobile APP or the self-service terminal on the site; S304, the user confirms the oil model P through the mobile phone APP or the self-service terminal in the station commend and refueling amount Q recommend , or change the oil model P through the mobile APP or the self-service terminal on the station commend and refueling amount Q recommend After the user confirms the oil selection, the intelligent refueling management module instructs the refueling machine to automatically start working. During this period, the refueling machine will dynamically display the refueling amount and real-time cost for the user's reference; S305. During the refueling process, the intelligent refueling management module monitors the operating status data of the refueling equipment in real time through the Internet of Things sensor, and instructs the data analysis module to analyze the operating status of the equipment. If an abnormality is detected (including but not limited to oil pipe leakage, refueling machine failure), the data analysis module immediately sends an emergency instruction to the intelligent refueling management module, and the intelligent refueling management module immediately stops refueling and sends an alarm instruction to the safety management and response module; S306. After refueling is completed, the intelligent refueling management module automatically calculates the total cost based on the actual refueling amount and the current oil price, displays it on the screen, and generates a payment QR code; S307, the user scans the payment QR code to complete the payment. After the intelligent refueling management module detects that the user has completed the payment, it prompts the user through a sound prompt and a display screen that the refueling is completed. At this time, the refueling machine is automatically closed, and the user is prompted to drive away directly; S308. The intelligent refueling management module sends an oil level monitoring instruction to the intelligent monitoring module. The intelligent monitoring module collects the remaining oil level data in the oil tank in real time and transmits the remaining oil level data in the oil tank to the data analysis module. If the data analysis module detects that the remaining oil level is lower than the preset warning value, it automatically sends a replenishment and alarm instruction to the safety management and response module.

[0020] Smart gas station personalized marketing stage: In the personalized marketing stage of smart gas stations, user refueling and consumption data are collected in real time and transmitted to the data analysis module for processing, static and dynamic tags are generated, and users are grouped using the K-means clustering algorithm. Then, user preferences are analyzed using the collaborative filtering algorithm. The data analysis results are transmitted to the personalized service module to generate recommended content and push it through mobile phone apps, advertising screens, and SMS messages, such as Figure 4 As shown in the figure, the implementation steps of the personalized marketing stage of smart gas stations are as follows: S401, collecting the user's refueling behavior data and consumption data in real time, and transmitting the user's refueling behavior data and consumption data to a data analysis module; S402: After processing the user's refueling behavior data and consumption data, the data analysis module generates a set of static and dynamic tags for each user: Among them, static tags include but are not limited to license plate number and vehicle model, and dynamic tags include but are not limited to user's fuel selection, refueling frequency, refueling time period, and consumption amount; S403, the data analysis module calls the personalized marketing model and uses the K-means clustering algorithm to classify users into behavior groups: A-type users, B-type users, and C-type users; Among them, Category A users prefer high-grade gasoline, Category B users pay more attention to promotional activities, and Category C users have a higher frequency but a lower consumption amount; S404, based on the collaborative filtering algorithm, the data analysis module infers the oil product or service preference of the current user by analyzing the historical behaviors of other user groups with similar consumption behaviors to the current user, and obtains the user's behavior analysis and grouping results; S405: The data analysis module transmits the user's behavior analysis and grouping results to the personalized service module, and the personalized service module generates personalized recommendation content: S406. The personalized service module pushes the recommended content to users through mobile phone apps, advertising screens, and text messages, and records the user's feedback data (including whether to participate in the promotion and whether to accept the recommended oil products). The feedback data is used as the basis for adaptive optimization of the model to adjust the marketing strategy.

[0021] The embodiments of the present invention are described in detail above with reference to the accompanying drawings, but the present invention is not limited thereto, and various changes can be made within the knowledge scope of technicians in the relevant technical field without departing from the purpose of the present invention.

Claims

1. A smart gas station management method based on AI Internet of Things, characterized in that: The specific implementation steps include the following: S1. During the initialization phase, several IoT sensors are installed at the core of the gas station to collect real-time status data of the gas station, and high-definition cameras are used to conduct real-time video monitoring of the gas station area and entrances and exits; S2. In the intelligent safety management stage of smart gas stations, IoT sensors and high-definition cameras collect real-time status data and environmental monitoring videos of gas stations and transmit them to the data analysis module. The data analysis module then analyzes and processes the data. If an abnormality is found, an alarm is sent to the safety management and response module. The safety management and response module automatically activates the emergency plan after identifying the alarm type. S3, in the smart refueling stage of the smart gas station, when a vehicle enters the gas station, a high-definition camera collects vehicle images and transmits them to the data analysis module, which identifies the license plate, retrieves user data associated with the vehicle, and transmits it to the smart refueling management module, which links the refueling equipment, recommends oil products and refueling quantities based on user preferences and historical records, and prompts the user to pay the fuel fee after refueling is completed; S4. In the personalized marketing stage of smart gas stations, user refueling and consumption data are collected and transmitted to the data analysis module for analysis and processing, static and dynamic tags are generated, and users are grouped using the K-means clustering algorithm. The collaborative filtering algorithm is then used to analyze user preferences. The data analysis results are transmitted to the personalized service module to generate recommended content, which is then pushed through mobile apps, advertising screens, and text messages.

2. According to claim 1, a smart gas station management method based on AI Internet of Things is characterized in that: The specific implementation steps of S2 are as follows: S21, the IoT sensor collects oil tank level data, temperature data and pressure data, the camera collects environmental monitoring video data in real time, and transmits the above oil tank level data, temperature data, pressure data and environmental monitoring video data to the data analysis module; S22, the data analysis module calls the constructed risk detection integrated model to perform concurrent processing and analysis on the received data. If an abnormality is detected, an alarm instruction is immediately sent to the security management and response module; S23. Once the safety management and response module receives the alarm instruction, the safety management and response module identifies the alarm type and immediately and automatically activates the emergency plan, remotely shuts down the refueling equipment, activates the fire protection system, notifies the emergency management personnel to handle it, and then automatically records the emergency handling process and generates an emergency handling report.

3. According to claim 1, a smart gas station management method based on AI Internet of Things is characterized in that: The specific implementation steps of S3 are as follows: S31. When a vehicle enters a gas station, a high-definition camera automatically collects vehicle images and transmits them to a data analysis module; S32, the data analysis module calls the vehicle recognition model, automatically detects the vehicle license plate, and recognizes the vehicle license plate number information through the OCR technology, and then the data analysis module automatically retrieves the user data associated with the vehicle license plate number information, and generates a temporary information credential for the user data associated with the vehicle license plate number information, and then transmits the user data associated with the vehicle and the temporary information credential to the intelligent refueling management module; The user data associated with the vehicle's license plate number information includes refueling preferences, weights of different oil products, and the latest historical refueling records; S33, when the vehicle stops at the refueling area, after the vehicle stops steadily, the intelligent refueling management module reviews the temporary information credentials of the user data associated with the license plate number information of the vehicle, confirms the credibility of the user data associated with the license plate number information of the vehicle, and then generates a recommended oil product and refueling amount, and sends the recommended oil product and refueling amount to the user; S34, the user confirms or modifies the recommended oil product and refueling amount, and the intelligent refueling management module instructs the refueling machine to automatically refuel; S35. During the refueling process, the intelligent refueling management module monitors the operating status data of the refueling equipment in real time through the Internet of Things sensor, and instructs the data analysis module to analyze the operating status of the equipment. If an abnormality is detected, the data analysis module immediately sends an emergency instruction to the intelligent refueling management module, and the intelligent refueling management module immediately stops refueling and sends an alarm instruction to the safety management and response module; S36. After refueling is completed, the intelligent refueling management module automatically calculates the total cost based on the actual refueling amount and the current oil price, and then generates a payment QR code; S37, the user scans the payment QR code to complete the payment. After the intelligent refueling management module detects that the user has completed the payment, the refueling machine is automatically turned off, and the user is prompted through a sound prompt and a display screen that the refueling is completed, and the user can drive away directly; S38. The intelligent refueling management module sends an oil level monitoring instruction to the intelligent monitoring module. The intelligent monitoring module collects the remaining oil level data in the oil tank in real time and transmits the remaining oil level data in the oil tank to the data analysis module. If the data analysis module detects that the remaining oil level is lower than the preset warning value, it automatically sends a replenishment and alarm instruction to the safety management and response module.

4. According to claim 3, a smart gas station management method based on AI Internet of Things is characterized in that: The process of generating temporary information credentials is as follows: S41. Define the finite cyclic group Z q * , and select Z q * A generator g of ; Among them, q is a large prime number, Z q * ={1, 2, ..., q-1}; S42, converting the user data associated with the vehicle into a binary string to obtain datav j , and use the random function rand() to generate two integers α and β; Among them, 1≤α≤q-1, and gcd(α,q-1)=1, 1≤β≤q-1, and gcd(β,q-1)=1, α×β≠0 mod (q-1), gcd() is the greatest common divisor function; S43, calculate the first-level information credential Ic1: ; S44, calculate the secondary information credential Ic2 = (H (datav j )-CI G ×Ic1)×β mod (q-1); Where H is the hash function, CI G The temporary information credential generation code generated during the system initialization phase and satisfies 1<CI G <q-1, thereby defining the temporary information credential resolution code CI E : ; S45, calculate the third-level information credential Ic3 = α × β mod (q-1); S46. Generate temporary information credentials IC={Ic1, Ic2, Ic3}.

5. The method for managing a smart gas station based on AI Internet of Things according to claim 3 is characterized in that: The process of confirming the authenticity of the user data associated with the vehicle's license plate number information is as follows: S51. Calculate the audit factor generation parameter P f : ; S52, calculate the first-level audit factor Af1 = H (datav j )×P f mod (q-1); Among them, datav j A binary string representation of user data associated with the vehicle; S53. Calculate the secondary audit factor Af2 = Ic1 × P f mod (q-1); S54. Calculate the audit factor A: ; Among them, CI E It is the temporary information credential parsing code; S55. If A=Ic1, it means that the user data associated with the vehicle is not out of order during the data transmission process, that is, the credibility of the user data associated with the license plate number information of the vehicle is confirmed.

6. The method for managing a smart gas station based on AI Internet of Things according to claim 1 is characterized in that: The specific implementation steps of S4 are as follows: S61, collecting the user's refueling behavior data and consumption data in real time, and transmitting the user's refueling behavior data and consumption data to a data analysis module; S62, the data analysis module processes the user's refueling behavior data and consumption data, and generates a set of static and dynamic tags for each user; S63, the data analysis module calls the personalized marketing model and uses the K-means clustering algorithm to classify users into behavior groups: A-type users, B-type users, and C-type users; S64. Based on the collaborative filtering algorithm, the data analysis module analyzes the historical behaviors of other user groups with similar consumption behaviors to the current user, infers the oil product or service preference of the current user, and obtains the user behavior analysis and grouping results; S65. The data analysis module transmits the user's behavior analysis and grouping results to the personalized service module, and the personalized service module generates personalized recommendation content: S66. The personalized service module pushes recommended content to users through mobile apps, advertising screens, and text messages, and records user feedback data. It uses the feedback data as a basis for adaptive optimization of the model and adjusts marketing strategies.

7. A smart gas station management system based on AI Internet of Things, which is used to execute a smart gas station management method based on AI Internet of Things as described in any one of claims 1 to 6, characterized in that: include: Intelligent monitoring module, which integrates several IoT sensors and high-definition cameras to collect working status data within the gas station; The data analysis module is operated by an AI algorithm box, which is an intelligent edge application server with a built-in domestic high-performance AI computing chip. The core is to use deep learning technology to customize the construction of fire detection models, vehicle identification models, customer behavior analysis models and risk detection integration models; Intelligent refueling management module, responsible for intelligent refueling of vehicles; Personalized service module, providing customers with personalized recommendation services; The safety management and response module automatically activates the emergency plan according to the alarm instructions and notifies the emergency management personnel to intervene.

Images