Gas station and oil depot maintenance and inspection system based on AR glasses

Abstract

The invention belongs to the technical field of intelligent detection and operation and maintenance, and provides a gas station and oil depot inspection and maintenance and inspection system based on AR glasses, and the system comprises the steps that an AR module detects a storage tank to obtain comprehensive multi-source data; the data visualization module superposes and displays the comprehensive multi-source data at the physical position of the storage tank in real time to obtain real-time picture data of the AR glasses; the path optimization module plans a shortest inspection path and guides an inspector through a ground projection arrow; the collaborative maintenance module receives the real-time picture data of the AR glasses through the collaborative platform, marks the real-time picture data and sends a maintenance instruction; the training drilling module simulates the oil unloading process of the oil tank truck, sends out an operation instruction and records the operation duration; the simulation evaluation module compares actual operation with the operation instruction, sends out virtual leakage early warning information and generates a capability evaluation report; the real-time storage tank state is obtained through the AR technology, and the shortest inspection path is obtained, so that the inspection time and energy consumption are reduced, and meanwhile, the maintenance time is shortened.

Description

Technical Field

[0001] The present invention belongs to the technical field of intelligent detection and operation and maintenance, and specifically relates to a gas station and oil depot inspection, repair and patrol system based on AR glasses. Background Art

[0002] With the rapid development of Industry 4.0 technology, augmented reality (AR) technology has gradually become an important tool for the intelligent transformation of the industrial field; applying it to the inspection, repair and patrol scenarios of modern gas stations and oil depots makes AR technology one of the core driving forces for the digital transformation of the oil and gas industry.

[0003] However, traditional gas stations and oil depots mostly rely on manual regular inspections and independently deployed monitoring devices. Manual inspections have obvious time lags. It is very difficult for inspectors to continuously and real-time monitor the equipment status and environmental parameters. It is difficult to detect early subtle changes with the naked eye and simple tools. By the time the hidden danger develops to an obvious stage and is detected, it may be close to a dangerous state. For the scattered independent monitoring devices, such as temperature sensors, gas detectors, etc., their data is often only displayed on their respective devices, lacking effective data integration and centralized analysis. Staff need to check back and forth between different devices, making it difficult to comprehensively judge the overall safety situation from a macroscopic perspective and unable to discover potential safety hazards in a timely and comprehensive manner, greatly increasing the risk of safety accidents in gas stations and oil depots;

[0004] In the patrol work, it was usually carried out according to experience or fixed routes in the past, lacking scientific route planning. The gas station and oil depot areas are vast and there are many devices. This method may cause the patrolmen to repeatedly pass through some areas or miss some key devices, wasting a lot of time and energy, resulting in low patrol efficiency and making it difficult to ensure the comprehensiveness and accuracy of the patrol. In terms of equipment maintenance, once a failure occurs, on-site maintenance personnel often have limited knowledge and experience and are difficult to quickly determine a solution to complex problems. At the same time, there are obstacles in communication and cooperation with remote experts, and only limited methods such as phone calls and photos can be used to transmit information, unable to let experts intuitively and real-time understand the on-site situation, resulting in a drawn-out maintenance process, a significant extension of the maintenance time, and a significant increase in labor costs;

[0005] Traditional personnel training methods mainly focus on classroom lectures and on-site observations. Classroom lectures lack the sense of immersion in real scenarios, making it difficult for staff to closely link abstract theoretical knowledge with actual operations. Their understanding of the operation process only stays on the surface. Although on-site observations allow staff to see the actual operation process, they cannot participate in it personally and are difficult to truly master operation skills and methods for dealing with emergencies. When encountering emergencies in actual work, due to the lack of sufficient practical experience, staff are often in a hurry and unable to make correct and timely responses, seriously affecting the performance of business capabilities and the effectiveness of emergency handling. Moreover, the assessment of staff in the daily operations of gas stations and oil depots is mostly subjective evaluation after the fact, lacking real-time and objective data support. It is impossible to compare actual operations with standard operation instructions in real time during the operation process, and it is difficult to detect subtle deviations and non-standard behaviors in the operation process.

[0006] Therefore, technicians in this field have proposed a gas station and oil depot inspection, repair and patrol system based on AR glasses, aiming to obtain real-time and accurate equipment status and environmental parameters through AR technology, and obtain the shortest patrol path to reduce patrol time and energy consumption, while shortening the repair time and reducing labor costs. Summary of the Invention

[0007] To solve the above technical problems, the present invention provides a gas station and oil depot inspection, repair and patrol system based on AR glasses to solve the problems raised in the background technology.

[0008] A gas station and oil depot inspection, repair and patrol system based on AR glasses includes:

[0009] An AR module, integrated with a FLIR thermal imager, a PID gas sensor and a liquid level sensor, is used to detect the surface temperature of the storage tank, detect VOCs and real-time liquid level, and obtain comprehensive multi-source data;

[0010] A data visualization module is used to connect the AR glasses to the IoT sensors of the oil depot through Bluetooth or Wi-Fi, and superimpose and display the comprehensive multi-source data in real time at the physical location of the storage tank to obtain real-time AR glasses picture data;

[0011] A path optimization module is used to plan the shortest patrol path based on the BIM map of the oil depot using the Dijkstra algorithm and guide the patrol personnel through ground projection arrows;

[0012] A collaborative repair module is used to receive the real-time AR glasses picture data through a collaborative platform, annotate the real-time AR glasses picture data to obtain annotation information, and issue a repair instruction based on the annotation information; the annotation information includes the fault point and the disassembly direction;

[0013] The training and simulation module is used to import the 3D model of the oil depot into the AR glasses, simulate the oil unloading process of the oil tanker, and issue operation instructions and record the operation duration.

[0014] The simulation and evaluation module is used to compare the actual operation with the operation instructions to obtain a comparison result. When the comparison result is inconsistent, a virtual leakage warning message is issued, and an ability evaluation report is generated based on the virtual leakage warning message and the operation duration.

[0015] Preferably, the data visualization module is further used for:

[0016] Obtaining the actual position coordinates (x w , y w , z w ) of the storage tank through the IoT sensors of the oil depot, and after being transformed by the internal parameter matrix K and the external parameter matrix [R|t] of the camera in the AR glasses, it is transformed into the point (x c , y c , z c ) in the camera coordinate system, and then through the projection transformation, it is transformed into the point (x s , y s ) in the screen coordinate system;

[0017] The relationship between the point (x c , y c , z c ) in the camera coordinate system and the actual position coordinates (x w , y w , z w ) of the storage tank is:

[0018]

[0019] The relationship between the point (x s , y s ) in the screen coordinate system and the point (x c , y c , z c ) in the camera coordinate system is:

[0020]

[0021] Wherein, K 11 , K 13 , K 22 , K 23 are the elements of the camera internal parameter matrix K;

[0022] Rendering the comprehensive multi-source data of the storage tank i through a graphics rendering algorithm in the virtual space of the AR glasses at the point (x s , y s) for real-time overlay display to obtain the real-time picture data p of the AR glasses.

[0023] Preferably, the path optimization module is used for:

[0024] Abstract each storage tank in the oil depot BIM map as a node in the graph. When there are n storage tanks in the oil depot, it is represented as a node set V = {v1, v2,..., v n};

[0025] When there is a direct channel connection between two nodes v i and v j , that is, add an edge e ij , and assign a weight ω ij to the edge e ij . The weight ω ij represents the distance and walking time measurement between two nodes, obtaining a weighted graph G = (V, E), where E is the set of edges;

[0026] Use the Dijkstra algorithm to plan the shortest inspection path, set the source node as v s , that is, the starting position of the inspector. Store the nodes with the determined shortest path through the set S. Initially, S = {v s};

[0027] Set a distance label d[i] for each node v i , indicating the current shortest distance estimate from the source node v s to the node v i . Initially, d[s] = 0. For other nodes v i ≠ v s , set d[i] = ∞;

[0028] In each iteration, select the node u with the smallest distance label from the set V - S, denoted as u = argmin v∈V-S d[v];

[0029] Add the node u to the set S. For the nodes v adjacent to the node u, when d[u] + ω uv < d[v], then update the distance label of the node v, d[v] = d[u] + ω uv , where ω uv is the weight of the edge between the node u and the node v;

[0030] Repeat the above steps until all nodes are added to the set S. The obtained distance label d[i] is the shortest distance from the source node v s to the node v i ;

[0031] From the target node v tStart by backtracking the path by recording the predecessor nodes of each node in the shortest path. When the predecessor node of node v i is p[i], then starting from the target node v t , by continuously accessing the predecessor nodes, i.e., v t-1 = p[t], v t-2 = p[t - 1],..., until returning to the source node v s , the shortest inspection path is obtained.

[0032] Preferably, the path optimization module is further configured to:

[0033] After determining the shortest inspection path, convert the path information into ground projection arrows to guide the inspector;

[0034] For the coordinates (x i , y i , z i ) of node v i , and the coordinates (x j , y j , z j ) of node v j , calculate the direction vector of each edge e ij in the shortest inspection path:

[0035]

[0036] According to the vector projection formula, project the direction vector onto the ground plane:

[0037]

[0038] where, is the projection vector, and is the normal vector of the ground plane;

[0039] Determine the direction and position of the arrow according to the projection vector , and guide the inspector to move forward along the shortest path through the arrows projected on the ground.

[0040] Preferably, the collaborative maintenance module is used for:

[0041] The AR glasses communicate with the collaborative platform through a network protocol. The AR glasses act as the sender, and the collaborative platform acts as the receiver. The AR glasses encapsulate the real-time video data p into a data packet P and send it to the collaborative platform; the data packet P includes video data D image and metadata M, expressed as P = (D image , M);

[0042] Annotate the real-time video data D through the collaboration platform image to obtain annotation information, where the annotation information includes the fault point and the disassembly direction;

[0043] Among them, the pixel coordinates of the fault point are (a, b), and the resolution of the real-time video data is W×H, satisfying 0≤a≤W, 0≤b≤H;

[0044] The disassembly direction is based on 0° being horizontally to the right in the real-time video data, and the angle θ is obtained as the disassembly direction, satisfying 0°≤θ<360°;

[0045] The annotation information is represented as a tuple I = ((a, b), θ).

[0046] Preferably, the collaborative maintenance module is further configured to:

[0047] Obtain a maintenance instruction O, which is generated by the collaboration platform according to the annotation information I. The maintenance instruction O includes fault point information, disassembly direction information, and explanatory text T, and is represented as O = (I, T);

[0048] Send the maintenance instruction O back to the AR glasses through the network, and encapsulate the maintenance instruction into a new data packet P using the communication connection O ;

[0049] After receiving the maintenance instruction data packet P O by the AR glasses, parse out the maintenance instruction O, and visually present the fault point and disassembly direction information to the patrol inspector.

[0050] Compared with the prior art, the present invention has the following beneficial effects:

[0051] 1. Through real-time monitoring and early warning, the present invention can timely discover and handle potential safety hazards, reduce the risk of safety accidents in gas stations and oil depots, and ensure the safety of personnel and equipment; and plan the shortest patrol path through the path optimization module, make information acquisition more intuitive through the data visualization module, and realize remote collaboration through the collaborative maintenance module, thereby reducing the time for patrol and maintenance, improving work efficiency, and reducing labor costs.

[0052] 2. The present invention provides an immersive training environment through the training and drill module, which helps the staff better master the operation process and emergency handling methods, and improves their business capabilities and the ability to handle emergencies.

[0053] 3. The present invention realizes real-time monitoring of the operation process and quantitative evaluation of the staff's capabilities by comparing the actual operation with the operation instructions in real time through the simulation and evaluation module, timely triggering warning information, and generating an ability evaluation report, which helps to continuously improve safety management and personnel training work. Brief Description of the Drawings

[0054] Figure 1 The following is a block diagram of the gas station and oil depot inspection, maintenance and patrol system based on AR glasses according to the present invention. Detailed Embodiments

[0055] The following further describes the embodiments of the present invention in detail in conjunction with the drawings and embodiments. The following embodiments are used to illustrate the present invention, but cannot be used to limit the scope of the present invention.

[0056] As shown in the attached Figure 1 figures

[0057] Embodiment 1: The present invention provides a gas station and oil depot inspection, maintenance and patrol system based on AR glasses, including:

[0058] An AR module, integrated with a FLIR thermal imager, a PID gas sensor and a liquid level sensor, for detecting the surface temperature of the storage tank, detecting VOCs and real-time liquid level, and obtaining comprehensive multi-source data;

[0059] The AR engine in the AR glasses retrieves the equipment maintenance manual from the EAM system, accesses the SCADA real-time data through the OPC UA protocol, and deploys an edge server in the oil depot machine room for processing the local image recognition of the AR glasses, reducing the cloud transmission delay. And because the high humidity and oil stains in the oil depot are likely to cause the camera recognition to fail, the AR glasses adopt an oleophobic coating lens. The AR module can collect multi-dimensional data of key equipment and environment in the gas station and oil depot in real time and accurately, providing a basis for subsequent analysis and decision-making, and helping to timely discover potential safety hazards, such as abnormal temperature, gas leakage and abnormal liquid level, etc.

[0060] A data visualization module, for connecting the AR glasses to the IoT sensors of the oil depot through Bluetooth or Wi-Fi, and overlaying and displaying the comprehensive multi-source data in real time at the physical location of the storage tank to obtain the real-time picture data of the AR glasses;

[0061] The actual position coordinates (x w , y w , z w ) of the storage tank are obtained through the IoT sensors of the oil depot, and after being transformed by the internal parameter matrix K and the external parameter matrix [R|t] of the camera in the AR glasses, they are transformed into the points (x c , y c , z c ) in the camera coordinate system, and then transformed into the points (x s , y s ) in the screen coordinate system through projection transformation;

[0062] The points (x c , yc , z c ), and the relationship with the actual position coordinates (x w , y w , z w ) of the storage tank is:

[0063]

[0064] The point (x s , y s ) in the screen coordinate system and the point (x c , y c , z c ) in the camera coordinate system have the following relationship:

[0065]

[0066] Among them, K 11 , K 13 , K 22 , K 23 are the elements of the camera internal parameter matrix K;

[0067] By using the graphics rendering algorithm, the comprehensive multi-source data of storage tank i is drawn on the point (x s , y s ) in the virtual space of the AR glasses for real-time overlay display, obtaining the real-time image data p of the AR glasses.

[0068] The data visualization module overlays and displays the comprehensive multi-source data in real time at the physical location of the storage tank, generating the real-time image of the AR glasses, enabling the inspection personnel to visually see the correspondence between the real-time data of the equipment and the equipment location, without having to look back and forth between different instruments and equipment, improving the inspection efficiency and accuracy, and reducing human errors.

[0069] The path optimization module is used to plan the shortest inspection path based on the oil depot BIM map using the Dijkstra algorithm and guide the inspection personnel through the ground projection arrow;

[0070] Each storage tank in the oil depot BIM map is abstracted as a node in the graph. When there are n storage tanks in the oil depot, it is represented as the node set V = {v1, v2,..., v n};

[0071] When there is a direct channel connection between two nodes v i and v j , an edge e ij is added, and the edge e ij is assigned a weight ω ij , and the weight ω ijA measure representing the distance and travel time between two nodes, obtaining a weighted graph G=(V, E), where E is the set of edges;

[0072] Use Dijkstra's algorithm to plan the shortest inspection path, set the source node as v s , that is, the starting position of the inspector. Store the nodes with the determined shortest path through the set S. Initially, S = {v s};

[0073] Set a distance label d[i] for each node v i , representing the current shortest distance estimate from the source node v s to the node v i . Initially, d[s] = 0. For other nodes v i ≠v s , set d[i] = ∞;

[0074] In each iteration, select the node u with the smallest distance label from the set V - S, denoted as u = argmin v∈V-S d[v];

[0075] Add the node u to the set S. For the nodes v adjacent to the node u, when d[u]+ω uv <d[v], then update the distance label of the node v, d[v] = d[u]+ω uv , where ω uv is the weight of the edge between the node u and the node v;

[0076] Repeat the above steps until all nodes are added to the set S. The obtained distance label d[i] is the shortest distance from the source node v s to the node v i ;

[0077] Starting from the target node v t , backtrack the path by recording the predecessor node of each node in the shortest path. When the predecessor node of the node v i is p[i], then starting from the target node v t , by continuously accessing the predecessor nodes, that is, v t-1 =p[t], v t-2 =p[t - 1],..., until returning to the source node v s , to obtain the shortest inspection path.

[0078] After determining the shortest inspection path, convert the path information into ground projection arrows to guide the inspector; [[ID=6l]]

[0079] For the coordinates (x i , y i , z i of the node vi ), the coordinates (x j of node v j , y j , z j ), calculate the direction vector of each edge e in the shortest inspection path: ij

[0080]

[0081] According to the vector projection formula, project the direction vector onto the ground plane:

[0082]

[0083] where is the projection vector, is the normal vector of the ground plane;

[0084] Determine the direction and position of the arrow according to the projection vector , and guide the inspector to move forward along the shortest path through the arrow projected on the ground.

[0085] Based on the BIM map of the oil depot, the path optimization module uses the Dijkstra algorithm to plan the shortest inspection path, and guides the inspector through the ground projection arrow, providing the inspector with the optimal inspection route, reducing the inspection time and physical consumption, ensuring that the inspection work can be completed comprehensively and efficiently, and avoiding the situation of missing important equipment or areas.

[0086] The collaborative maintenance module is used to receive the real-time video data of the AR glasses through the collaborative platform, annotate the real-time video data of the AR glasses to obtain the annotation information, and issue a maintenance instruction based on the annotation information; the annotation information includes the fault point and the disassembly direction;

[0087] The AR glasses and the collaborative platform are connected through a network protocol. The AR glasses act as the sender, and the collaborative platform acts as the receiver. The AR glasses encapsulate the real-time video data p into a data packet P and send it to the collaborative platform; the data packet P includes the video data D image and the metadata M, expressed as P = (D image , M);

[0088] Annotate the real-time video data D image through the collaborative platform to obtain the annotation information, and the annotation information includes the fault point and the disassembly direction;

[0089] Among them, the pixel coordinates of the fault point are (a, b), and the resolution of the real-time video data is W × H, satisfying 0 ≤ a ≤ W, 0 ≤ b ≤ H;

[0090] ​The disassembly direction is based on the horizontal right in the real-time video data as 0°, and the angle θ is obtained as the disassembly direction, satisfying 0° ≤ θ < 360°;

[0091] The annotation information is represented as a tuple I = ((a, b), θ).

[0092] The collaboration platform generates a maintenance instruction O according to the annotation information I. The maintenance instruction O includes the fault point information, the disassembly direction information, and the explanatory text T, and is represented as O = (I, T);

[0093] The collaboration platform sends the maintenance instruction O back to the AR glasses through the network, and uses the communication connection to encapsulate the maintenance instruction into a new data packet P O ;

[0094] The AR glasses receive the maintenance instruction data packet P O After that, the maintenance instruction O is parsed out, and the fault point and the disassembly direction information are visually presented to the patrol inspector.

[0095] By receiving the real-time video of the AR glasses through the collaboration platform, annotating the video, obtaining the annotation information including the fault point and the disassembly direction, etc., and issuing a maintenance instruction based on this, the efficient collaboration between the remote expert and the on-site patrol inspector is realized. The expert can accurately guide the maintenance work according to the real-time video, improving the efficiency and accuracy of the maintenance, reducing the maintenance time and cost, and reducing the safety risks caused by untimely or inaccurate maintenance.

[0096] The training and rehearsal module is used to import the 3D model of the oil depot into the AR glasses, simulate the oil unloading process of the oil tanker, and issue operation instructions and record the operation duration; it provides a realistic training and rehearsal environment for the patrol inspector and relevant staff, enabling them to be familiar with the oil unloading process and operation specifications in the virtual scenario, improving the operation skills and emergency handling capabilities of the staff, and reducing the mistakes and risks in actual operations.

[0097] The simulation and evaluation module is used to compare the actual operation with the operation instruction to obtain a comparison result. When the comparison result is inconsistent, a virtual leakage warning message is issued, and an ability evaluation report is generated based on the virtual leakage warning message and the operation duration. Through the simulation and evaluation module, the non-standard behaviors in the operation can be found in time, and the staff can be reminded of the safety risks through the virtual leakage warning message. At the same time, the ability evaluation report can help the management understand the business capabilities of the staff, provide a basis for subsequent training and assessment, and contribute to improving the overall safety management level.

[0098] Embodiment 2: This embodiment is basically the same as the previous embodiment, except that the simulation and evaluation module is further used for:

[0099] The operation instructions are a series of standard operation steps and parameters preset in the training and rehearsal module. The operation instructions contain n steps, and the set of standard parameters corresponding to each step is S i ={s i1 , s i2 ,..., s im}, where i = 1, 2,..., n, n represents the step number, and m represents the number of parameters for each step;

[0100] During the actual operation process, the AR glasses collect the operation data of the patrolman in real time. The set of parameters corresponding to each step of the actual operation is A i ={a i1 , a i2 ,..., a im};

[0101] For each step i, the difference between the actual operation parameters and the standard operation parameters is calculated to determine whether they are consistent. For each parameter j, the error is calculated as e ij =|a ij -s ij |;

[0102] The consistency of each step is comprehensively evaluated by defining a step error metric E i , taking the weighted sum of the parameter errors where ω j is the weight of parameter j, and

[0103] Then, a threshold τ i is defined. When E i ≤τ i , it is considered that the operation of step i is consistent, denoted as C i =1; otherwise, it is denoted as C i =0;

[0104] By performing a logical sum operation on the consistency results of all steps, the overall comparison result C is obtained, that is:

[0105]

[0106] When C = 1, it means that the actual operation is consistent with the operation instructions; when C = 0, it means that there is an inconsistent situation.

[0107] When the comparison result C = 0, the simulation evaluation module issues a virtual warning message. The virtual warning message contains the faulty step number and the error parameters; the set of faulty step numbers is F = {i|C i = 0}, and the warning message W is expressed as W = (F, {E i |i ∈ F}).

[0108] Record the total actual operation duration T a , and the standard duration set in the operation instruction is T s , calculate the deviation rate of the operation duration as

[0109] Adopt the weighted average method. Let the weight of the comparison result be α, and the weight of the deviation rate of the operation duration be β, and α + β = 1; for the comparison result, when C = 1, the score S C = 100; when C = 0, S C Adjust according to the number of fault steps and the degree of error, expressed as S C = 100 - ∑ i∈F kE i , where k is the adjustment coefficient;

[0110] The operation duration score S T Calculate according to the deviation rate, expressed as S T = 100 - 100R T ;

[0111] The final ability evaluation score S = αS C + βS T .

[0112] As can be seen from the above, the AR module integrates a variety of sensors, realizes the fusion acquisition of multi-source data, and can more comprehensively reflect the actual situation of gas stations and oil depots. Compared with the traditional single-sensor monitoring method, it has higher accuracy and reliability; the data visualization module combines AR technology with the IoT sensors of the oil depot, and provides a new and intuitive way of information acquisition for the patrol inspectors by overlaying data on the physical location of the equipment in real time, breaking the limitations of scattered information and inconvenient access in traditional patrols; the path optimization module uses the BIM map of the oil depot and the Dijkstra algorithm to plan the shortest patrol path, making full use of the detailed information of the BIM model and the optimization ability of the algorithm. Compared with the traditional empirical patrol path planning, it is more scientific and reasonable, and can effectively improve the patrol efficiency; the collaborative maintenance module and the training and rehearsal module respectively realize remote collaborative maintenance and immersive training and rehearsal. Through AR glasses and the collaborative platform, they break the limitations of time and space, provide a new mode for expert-guided maintenance and personnel training, and improve the maintenance efficiency and training effect, which are relatively rare in the traditional management of gas stations and oil depots; the simulation evaluation module can compare the actual operation with the operation instructions in real time, trigger warning information in time, and generate an ability evaluation report, realizing the real-time monitoring of the operation process and the quantitative evaluation of the staff's ability, which helps to continuously improve safety management and personnel training work, and this is also a function lacking in the existing technology.

[0113] Importantly, it should be noted that the construction and arrangement of the present application shown in multiple different exemplary embodiments are merely illustrative. Although only a few embodiments are described in detail in this disclosure, those who refer to this disclosure should easily understand that many modifications are possible without substantially departing from the novel teachings and advantages of the subject matter described in this application. Other substitutions, modifications, changes, and omissions may be made in the design, operating conditions, and arrangement of the exemplary embodiments without departing from the scope of the present invention. Therefore, the present invention is not limited to a particular embodiment, but extends to various modifications that still fall within the scope of the appended claims.

[0114] In addition, in order to provide a concise description of the exemplary embodiments, all features of the actual embodiments may not be described (i.e., those features that are not relevant to the best mode currently considered for carrying out the present invention or those features that are not relevant to implementing the present invention).

[0115] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical solutions of the present invention, and they should all be covered by the scope of the claims of the present invention.

Claims

1. A gas station and oil depot inspection, maintenance and patrol system based on AR glasses, characterized in that, Including: An AR module integrated with a FLIR thermal imager, a PID gas sensor, and a liquid level sensor, used to detect the surface temperature of the storage tank, detect VOCs, and real-time liquid level, obtaining comprehensive multi-source data; A data visualization module, used to connect the AR glasses to the IoT sensors of the oil depot via Bluetooth or Wi-Fi, and overlay and display the comprehensive multi-source data in real-time at the physical location of the storage tank, obtaining the real-time picture data of the AR glasses; A path optimization module, used to plan the shortest inspection path based on the BIM map of the oil depot using the Dijkstra algorithm, and guide the inspector through ground projection arrows; A collaborative maintenance module, used to receive the real-time picture data of the AR glasses through a collaborative platform, annotate the real-time picture data of the AR glasses to obtain annotation information, and issue a maintenance instruction based on the annotation information; the annotation information includes the fault point and the disassembly direction; A training and simulation module, used to import the 3D model of the oil depot into the AR glasses, simulate the oil tanker unloading process, and issue operation instructions and record the operation duration; A simulation and evaluation module, used to compare the actual operation with the operation instructions to obtain a comparison result. When the comparison result is inconsistent, issue a virtual leakage warning message, and generate a capacity evaluation report based on the virtual leakage warning message and the operation duration.

2. The inspection and maintenance and patrol system for gas stations and oil depots based on AR glasses according to claim 1, characterized in that The data visualization module is further used for: The actual position coordinates (x w , y w , z w ) of the storage tank are obtained through the IoT sensors in the oil depot. After being transformed by the internal parameter matrix K and the external parameter matrix [R|t] of the camera in the AR glasses, they are converted to the points (x c , y c , z c ) in the camera coordinate system, and then are converted to the points (x s , y s ) in the screen coordinate system through projection transformation; Among them, the relationship between the point (x c , y c , z c ) in the camera coordinate system and the actual position coordinates (x w , y w , z w ) of the storage tank is as follows: The point (x s , y s ) in the screen coordinate system and the point (x c , y c , z c ) in the camera coordinate system have the following relationship: Among them, K 11 , K 13 , K 22 , K 23 are the elements of the camera intrinsic matrix K; Render the comprehensive multi-source data of storage tank i in the virtual space of the AR glasses through a graphics rendering algorithm and draw it at the point (x s , y s ) in the screen coordinate system for real-time overlay display to obtain the real-time picture data p of the AR glasses.

3. The inspection and maintenance and patrol system for gas stations and oil depots based on AR glasses according to claim 1, characterized in that The path optimization module is further used for: Abstract each storage tank in the oil depot BIM map as a node in the graph. When there are n storage tanks in the oil depot, it is represented as a node set V = {v1, v2,..., v n}; When there is a direct channel connection between two nodes v i and v j , that is, add an edge e ij , and assign a weight ω ij to the said edge e ij , where the weight ω ij represents the distance between the two nodes and a measure of the travel time, obtaining a weighted graph G=(V, E), where E is the set of edges; Use the Dijkstra algorithm to plan the shortest inspection path, and set the source node as v s , which is the starting position of the inspector. Store the nodes with the determined shortest paths in the set S. Initially, S = {v s}; For each node v i Set the distance label d[i], which represents the current shortest distance estimate from the source node v s to the node v i Initially, d[s] = 0. For other nodes v i ≠ v s , set d[i] = ∞; In each iteration, select the node u with the smallest distance label from the set V - S, denoted as u = argmin v∈V-S d[v]; Add node u to set S. For node v adjacent to node u, when d[u] + ω uv < d[v], then update the distance label of node v as d[v] = d[u] + ω uv , where ω uv is the weight of the edge between node u and node v; Repeat the above steps until all nodes are added to the set S, and the resulting distance label d[i] is the shortest distance from the source node v s node v i ; Starting from the target node v t Begin, backtrack the path by recording the predecessor node of each node in the shortest path. When the predecessor node of node v i is p[i], then starting from the target node v t Begin, by continuously accessing the predecessor nodes, that is, v t-1 = p[t], v t-2 = p[t - 1],..., until returning to the source node v s , the shortest inspection path is obtained.

4. The inspection and maintenance and patrol system for gas stations and oil depots based on AR glasses according to claim 3, characterized in that, The path optimization module is further used for: After determining the shortest inspection path, convert the path information into a ground projection arrow to guide the inspector; For node v i with coordinates (x i , y i , z i ), and node v j with coordinates (x j , y j , z j ), calculate the direction vector of each edge e ij in the shortest inspection path: According to the vector projection formula, project the direction vector onto the ground plane: Among them, is the projection vector, is the normal vector of the ground plane; According to the projection vector determine the direction and position of the arrow, and guide the inspector to move forward along the shortest path through the arrow projected on the ground.

5. The inspection and maintenance and patrol system for gas stations and oil depots based on AR glasses according to claim 1, characterized in that The collaborative maintenance module is further used for: The AR glasses communicate with the collaborative platform through a network protocol. The AR glasses act as the sender, and the collaborative platform acts as the receiver. The AR glasses encapsulate the real-time picture data p into a data packet P and send it to the collaborative platform; The data packet P includes video data D image and metadata M, expressed as P = (D image , M); Annotate the real-time video data D through the collaboration platform image to obtain annotation information, where the annotation information includes the fault point and the disassembly direction; Among them, the pixel coordinates of the fault point are (a, b), and the resolution of the real-time picture data is W×H, satisfying 0≤a≤W, 0≤b≤H; The disassembly direction is based on the horizontal right in the real-time picture data as 0°, and the obtained angle θ is used as the disassembly direction, satisfying 0°≤θ<360°; The annotation information is represented as a tuple I = ((a, b), θ).

6. The inspection and maintenance and patrol system for gas stations and oil depots based on AR glasses according to claim 5, wherein: The collaborative maintenance module is further used for: Obtain a maintenance instruction O, which is generated by the collaborative platform according to the annotation information I. The maintenance instruction O includes fault point information, disassembly direction information, and explanatory text T, and is represented as O = (I, T); Send the maintenance instruction O back to the AR glasses via the network, and encapsulate the maintenance instruction into a new data packet P using the communication connection O ; After the AR glasses receive the maintenance instruction data packet P O it parses out the maintenance instruction O and visually presents the fault point and disassembly direction information to the inspection operator.

Images