Site selection method for tunnel management substations in a quasi-unmanned system for highway tunnels

Abstract

The present invention relates to a method for site selection of tunnel management substations under a quasi-unmanned system for highway tunnels, and belongs to the technical field of highway tunnels. The fifth level of the quasi-unmanned system for highway tunnels is a tunnel management substation, and the sixth level is an intelligent microstation, including an intelligent substation microstation, an intelligent management microstation, an intelligent fire-fighting microstation, a diesel generator box, and a modular Internet of Things water tank; the setting principle is that the total inspection distance of a single tunnel management station or tunnel management substation does not exceed 30km, and the single side does not exceed 15km; the site selection of the tunnel management substation adopts an improved salp group optimization algorithm based on Cat chaos mapping and hierarchical analysis method. The present invention realizes the quasi-unmanned operation of the tunnel substation, solves the many difficulties currently faced by tunnel substations during the construction and operation periods, improves the timeliness and safety of emergency rescue, reduces the inspection pressure of the tunnel management station, and promotes cost reduction and efficiency improvement in the construction, management and maintenance of highway tunnels.

Description

Technical Field

[0001] The invention belongs to the technical field of highway tunnels, and in particular relates to a method for selecting a tunnel management substation site under a quasi-unmanned system of a highway tunnel. Background Art

[0002] The operation and management of highways is crucial to ensuring the safety and efficiency of highway traffic. Tunnels, as one of the key components of highways, pass through mountains or other obstacles, providing vehicles with a direct path that is not restricted by terrain, greatly shortening travel distance and time. The safety of tunnels plays a decisive role in ensuring the overall safety of highways. To achieve tunnel management, the current highway tunnel operation and management system is generally divided into five levels, such as Figure 1 shown.

[0003] The first level is the Highway Transportation Management Department, which includes departments such as the Toll Collection Section and the Maintenance Section. It is responsible for organizing and arranging the daily maintenance of highway tunnels, maintenance project plans, making decisions on the scheduling of maintenance resources, and supervising the implementation of maintenance plans. The second level is the Management Office, a coordinating and dispatching department under the command of the Transportation Management Department. Its organizational structure includes the Toll Collection Section, the Monitoring Sub-Center, and the Maintenance Section. It is responsible for the management, scheduling, supervision, and decision-making of the daily maintenance and fire rescue work of the tunnel management station. The third level is the Management Sub-Office, which specifically divides and implements the work of the Management Office. The fourth level is the Tunnel Management Station, a key grassroots implementation department. Its organizational structure includes fire rescue, inspection and maintenance, a video surveillance center, and a substation. It is responsible for daily, regular, and periodic inspections of tunnel civil engineering, electromechanical equipment, and traffic safety, as well as fire rescue, emergency response, video surveillance, and document compilation and summary. The fifth level is the tunnel substation, whose organizational structure mainly includes the duty room, power generation room, substation room, and monitoring room. Its main function is to provide stable and reliable power supply for tunnel ventilation, lighting, monitoring and other electromechanical equipment, and is responsible for electromechanical equipment inspection, tunnel inspection, timely restart of generators in case of tunnel power outages, dispersal of pedestrians in tunnels, and inspection of high-level water tanks.

[0004] The number of tunnel substations varies with tunnel length. Short tunnels (less than 500m) generally use small box-type substations to power ventilation, lighting, monitoring, and communication equipment in the tunnel; medium, long, and extra-long tunnels use building-type substations for power supply. Among them, medium tunnels (length between 500m and 1000m) have a building-type substation built at the entrance or exit of the tunnel; long tunnels (length between 1000m and 3000m) have a tunnel substation built at the entrance and exit of the tunnel; extra-long tunnels (length greater than 3000m) generally have a building-type tunnel substation built at the entrance and exit of the tunnel, as well as a corresponding number of tunnel substations built in the tunnel according to the tunnel length to power ventilation, lighting, monitoring, and communication equipment in the tunnel.

[0005] In the existing management system, both tunnel management stations and tunnel substations require construction facilities. Tunnel management stations have a large jurisdiction, and to complete daily and regular inspections, they are typically divided into multiple inspection teams. Inspection teams located far from their destinations also need to consider travel time, meaning their working hours include both travel time and actual inspection time. When the inspection destination is far from the tunnel management station, the actual inspection time may be far less than the travel time, placing significant pressure on inspection personnel and incurring significant time and fuel costs. Furthermore, when fires or traffic accidents occur far from the tunnel management station, emergency rescue personnel struggle to reach the scene quickly due to distance and road conditions. This significantly reduces the timeliness of emergency response and can even lead to secondary accidents, resulting in significant loss of life and property.

[0006] Currently, tunnel substations are generally built at the tunnel entrance, requiring an area of ​​approximately 1.5 mu. During construction, they face the following serious problems:

[0007] (1) Difficulty in site selection: The terrain along the expressway is complex, with a high proportion of bridges and tunnels. Some routes cross the "red line area", making site selection, land acquisition, and layout of various facilities at the tunnel entrance difficult.

[0008] (2) Difficulty in construction: The construction area of ​​a tunnel substation is large. At the same time, it is very common for mountain highways to be connected by bridges and tunnels, making construction difficult.

[0009] (3) High risk: Most tunnel entrances along the highway are narrow, with steep slopes and complex geological conditions, which poses a high construction risk.

[0010] (4) Long construction period: The highway has special terrain and geological conditions, complex professional interfaces, and many processes, so the construction period is long.

[0011] (5) High cost: The design involves high costs for substation site preparation and special structure construction. Substations require a large number of on-duty personnel, resulting in high human resource costs for long-term operation. Furthermore, the substation is located in a remote and sparsely populated area, which can create a psychological burden on the substation staff.

[0012] (6) Infeasibility: The land acquisition indicators for water fire protection, substations and related facilities are difficult to meet the requirements of relevant specifications and standards.

[0013] Therefore, how to overcome the shortcomings of the existing technology is an urgent problem to be solved in the field of highway tunnel technology. Summary of the Invention

[0014] The purpose of the present invention is to solve the deficiencies of the prior art and to provide a method for selecting a tunnel management substation site under a quasi-unmanned system for a highway tunnel.

[0015] To achieve the above object, the technical solution adopted by the present invention is as follows:

[0016] The site selection method for tunnel management substations under the quasi-unmanned system of expressway tunnels is to set up tunnel management substations below the tunnel management station; the principle is that the total inspection distance of a single tunnel management station or tunnel management substation shall not exceed 30km, and the inspection distance on one side shall not exceed 15km;

[0017] The location of the tunnel management substation is selected using an improved salp swarm optimization algorithm based on Cat chaos mapping and hierarchical analysis method.

[0018] Furthermore, the location of the tunnel management substation is selected using an improved salp swarm optimization algorithm based on Cat chaos mapping and analytic hierarchy process, which specifically includes the following steps:

[0019] Step 1: Set the population size, number of iterations, and upper and lower bounds of the search space.

[0020] Step 2: Initialize the positions of individual salps and the salp population within the search space.

[0021] Among them, the individuals of the salp group are represented as:

[0022] in, The individual number of the salp is Candidate locations, For the The pile number information of each candidate point, For the The total length of the tunnel from the candidate point to the accident site, For the The number of extra-long tunnels from each candidate point to the accident site, For the The total length of the long downhill slope from the candidate point to the accident site, For the The number of small radius curves from the candidate point to the accident site, For the The total length of the road sections that are prone to rain and fog from the candidate points to the accident site, For the The total length of the frequently icy and snowy road sections from the candidate points to the accident site;

[0023] The characteristic information of n individuals in the salp group is stored in of Matrix, as shown in formula (2):

[0024]

[0025] Where n represents the population size of salps, =1,2,…,n; is the j-th dimension feature of the individual, d is the number of features, j = 1, 2, ..., d; For individual salps, is the position of the characteristic variable of the i-th salp in the j-dimensional space, and the matrix That is the salp population;

[0026] Salp populations mapped by Cat The initial position of is given by formula (4):

[0027]

[0028] in, is the initial position of the i-th salp on the j-th dimension feature variable, is the lower bound of the search space for the j-th dimension feature variable, is the upper bound of the search space of the j-th dimension feature variable, is the final ordinate chaotic sequence value of the i-th individual and the j-th dimension characteristic variable after k iterations;

[0029] Step 3: construct a fitness function and calculate the fitness value of each salp individual at the current position;

[0030] The fitness function is a function with the minimum emergency rescue time as the goal , as shown in formula (6):

[0031]

[0032]

[0033] in, represents a tunnel management node, which includes a tunnel management station or a tunnel management substation; Indicates the point where the accident occurred. It is from Tunnel management node to the The emergency distance of each accident point, for The weight value of is in the range of (0,1); It is from Tunnel management node to the The total length of the tunnels passed by the accident points is for The weight value of is in the range of (0,1); It is from Tunnel management node to the The number of tunnels that the accident point passes through, for The weight value of is in the range of (0,1); It is from Tunnel management node to the The length of the long downhill section that the accident point passes through, for The weight value of is in the range of (0,1); It is from Tunnel management node to the The number of small radius curves that the accident point passes through, for The weight value of is in the range of (0,1); It is from Tunnel management node to the The length of the rainy and foggy road section where the accident point passes, for The weight value of is in the range of (0,1); It is from Tunnel management node to the The length of the icy and snowy road section where the accident point passes, for The weight value of is in the range of (0,1);

[0034] The distance between any two tunnel management nodes should be 15 km, and the total inspection distance of a single tunnel management station or tunnel management substation must not exceed 30 km;

[0035] The individual fitness value is the inverse of its corresponding objective function value, as shown in formula (11):

[0036]

[0037] Step 4: Sort the salps individuals in descending order of fitness value, select the individuals with the largest fitness value as the food source, set the top 50% of the salps in fitness as leaders, and the rest of the salps as followers;

[0038] Step 5, update the position of the salp group;

[0039] In a salp group, the position update formula of the leader salp is shown in formula (12):

[0040]

[0041] in, is the position of leader s in the j-th dimension characteristic variable, s = 1, 2, …, n; is the location of the food source in the jth dimension characteristic variable, and are all random numbers between [0,1]. is the convergence factor;

[0042] The update formula is shown in formula (13):

[0043]

[0044] in, Indicates the current iteration number, Indicates the maximum number of iterations;

[0045] In a salp group, the position of the follower salp is updated according to Newton's law of motion, as shown in Equation (17):

[0046]

[0047] in, , Indicates the The position of each salp in the j-th dimension of the characteristic variable; Indicates the The position of the previous salp of the salp in the j-th dimension feature variable;

[0048] Step 6, calculating the fitness value of each salp according to the fitness function, and re-determining the food source location;

[0049] Step 7: Determine whether the number of algorithm iterations meets the stopping condition. If the stopping condition is not met, return to step 5, update the position of the salp group, and continue to iterate;

[0050] After the iteration is terminated, the position of the leader salp is output, that is, the optimal value of the candidate location of the tunnel management substation is obtained.

[0051] Further, in step 3, The value of is determined by the analytic hierarchy process.

[0052] Furthermore, the quasi-unmanned system for highway tunnels includes six levels: the first level is the highway transportation management department, the second level is the management office, the third level is the management sub-office station, the fourth level is the tunnel management station, the fifth level is the tunnel management sub-station, and the sixth level is the intelligent micro-station. The intelligent micro-station is set at the lower level of the tunnel management station or the tunnel management sub-station;

[0053] The intelligent micro station includes an intelligent substation, an intelligent management micro station, an intelligent fire protection micro station, a diesel generator box and a modular IoT water tank;

[0054] The intelligent substation is used to ensure the power supply of the tunnel and to perform intelligent management and control of the power supply of the tunnel;

[0055] The intelligent management microstation is used to manage tunnel lighting facilities, ventilation facilities, monitoring facilities, fire protection facilities, as well as intelligent substations, intelligent management microstations, intelligent fire protection microstations, diesel generator boxes, and modular IoT water tanks. The intelligent management microstation is also equipped with necessary emergency supplies.

[0056] The intelligent fire fighting microstation is used to draw water from a modular IoT water tank or high-level water pool to provide stable water pressure for tunnel fire fighting, and to monitor the water pressure, liquid level, and pump operating status of the fire fighting water supply system in real time.

[0057] The diesel generator box is used to monitor the working status, fuel level, faults and operating environment information of the diesel generator, and control the start or stop of the diesel generator according to the mains power on / off status and inspection requirements. The diesel generator serves as a backup power source for the mains power supply system;

[0058] The modular IoT water tank is composed of several independent water storage units, and is equipped with a liquid level detector and a water pressure detector;

[0059] The modular IoT water tank is used as a fire water supply system for tunnels with a length of less than or equal to 5,000 meters. At the same time, the modular IoT water tank can monitor its own liquid level and pressure information in real time.

[0060] Furthermore, tunnels with a length greater than 5,000m use high-level water tanks to provide fire water sources at constant high pressure. Smart fire microstations provide stable water pressure for tunnel fire protection and monitor water pressure and flow information. Tunnels with a length less than or equal to 5,000m use a temporary high-pressure water supply system that combines smart fire microstations and modular IoT water tanks. The modular IoT water tanks serve as fire water sources, and the smart fire microstations provide stable water pressure and monitor water pressure and flow information. The modular IoT water tanks are set at the tunnel entrance.

[0061] Furthermore, tunnel management substations are built together with toll stations or service areas along the highway, or are set up at the entrances of extra-long tunnels.

[0062] Furthermore, the intelligent substation conducts intelligent management and control of the tunnel's power supply, including real-time power monitoring of the voltage, current, and power information of each power distribution circuit of the high-voltage, transformer, and low-voltage cabinets, and remote control of the high-voltage distribution circuit, low-voltage lighting circuit, low-voltage ventilation circuit, fire-fighting equipment, and air conditioning. At the same time, it can analyze the energy consumption of the tunnel's lighting, ventilation, fire-fighting, dynamic environment, and video surveillance circuits.

[0063] In the present invention, normal high pressure and temporary high pressure are standard terms in the fire protection field. Please refer to the relevant descriptions of normal high pressure fire water supply system and temporary high pressure fire water supply system in the fire protection field.

[0064] Compared with the traditional tunnel operation system, the system of the present invention adds a tunnel management substation under the tunnel management station, pre-positions some inspection and emergency rescue functions of the tunnel management station to the tunnel management substation, and replaces the traditional building-based substation with an intelligent microstation.

[0065] Some inspection and emergency services at the tunnel management station are pre-located to the tunnel management substation, and three to four staff members are deployed from the tunnel management station to man the substation. The intelligent microstation is located at the tunnel entrance and serves as a subordinate of the tunnel management station or tunnel management substation. The establishment and location of the tunnel management substation are determined by the tunnel management substation site selection method for the quasi-unmanned highway tunnel system described in this invention.

[0066] In the present invention, the intelligent management microstation is used to intelligently manage the intelligent substation, intelligent fire protection microstation, intelligent diesel generator box, modular Internet of Things water tank, tunnel lighting facilities, ventilation facilities, monitoring facilities, and fire protection facilities; at the same time, it is equipped with necessary emergency supplies to comprehensively ensure the safe operation of the tunnel and a series of intelligent microstations.

[0067] In the present invention, the intelligent substation is used to ensure the safe operation of the tunnel and stable power supply. It also conducts real-time power monitoring of the voltage, current, and power information of the power supply and distribution circuits of the high-voltage, transformer, and low-voltage cabinets, and remote control of the high-voltage distribution circuit, low-voltage lighting circuit, low-voltage ventilation circuit, fire-fighting equipment, and air conditioning. It analyzes the energy consumption of the circuits for lighting, ventilation, fire protection, dynamic environment, and video monitoring in the tunnel, quickly realizes abnormal event identification and fault diagnosis in an unmanned environment, and realizes intelligent management and control of the tunnel power supply.

[0068] The intelligent fire-fighting micro-station is used to draw water from a modular IoT water tank or high-level water pool to provide stable water pressure for tunnel fire protection, and to monitor the water pressure, liquid level, and water pump operating status of the fire water supply system in real time. It is used for all-round, visual remote real-time monitoring of the fire water supply system, which can quickly detect anomalies and faults in the fire protection system, ensuring the safety, reliability and fire-fighting efficiency of the tunnel fire water supply system.

[0069] The diesel generator box is used to monitor the working status, fuel level, faults and operating environment information of the diesel generator, and control the start or stop of the diesel generator according to the mains power on / off status and inspection requirements. The diesel generator serves as a backup power source for the mains power supply system;

[0070] The modular IoT water tank is formed by combining several 1m³ independent water storage units based on the tunnel fire water consumption. It is used as a tunnel fire water supply system for tunnels less than or equal to 5000m in length. At the same time, the water tank is equipped with a liquid level detector and a water pressure detector, which can monitor the liquid level and pressure information of the modular IoT water tank in real time. The modular IoT water tank is equipped with a pressure-stabilizing pump to stabilize the fire water pressure.

[0071] In the present invention, It represents a tunnel management node, and the tunnel management node includes a tunnel management station or a tunnel management substation, wherein the location of the tunnel management station has been determined during planning and design, and the location of the tunnel management substation is determined by the tunnel management substation location selection method under the quasi-unmanned system of the highway tunnel described in the present invention.

[0072] At present, highway tunnel management faces the problems of wide jurisdiction, heavy inspection pressure, high cost, and untimely emergency rescue. Tunnel substations face many challenges in the construction, management and maintenance process, such as difficult site selection, difficult construction, high risk, long construction period, high cost and impracticability. Therefore, the present invention uses the dual-wheel drive of "system + technology" to investigate the organizational structure and functions of the existing system at all levels, analyze the problems faced by the existing system, and through the research and application of highway tunnel power supply and distribution technology, communication technology, computer technology and embedded technology, use information equipment to optimize the power supply, monitoring and management solutions for tunnel ventilation, lighting, monitoring and communication equipment, realize quasi-unmanned operation of tunnel substations, solve the many difficulties currently faced by tunnel substations during the construction and operation periods, improve the timeliness and safety of emergency rescue, reduce the inspection pressure of tunnel management stations, and promote cost reduction and efficiency improvement in highway tunnel construction, management and maintenance.

[0073] The present invention relates to a method for selecting tunnel management substations within a quasi-unmanned system for highway tunnels. This method optimizes traditional highway tunnel systems and includes the establishment and implementation of a "quasi-unmanned" system for highway tunnels. The method proposes the concept and method for setting up tunnel management substations. Tunnel management substations are located below tunnel management stations in the traditional management system. An objective function is defined based on factors that influence the timeliness of highway emergency rescue, such as emergency distance, number of tunnel clusters, length of extra-long tunnels, and length of long downhill slopes. Using an improved salp swarm optimization algorithm, the optimal location for setting up tunnel management substations is found within the feasible solution space. This maximizes the timeliness of emergency rescue and reduces inspection pressure on tunnel management stations.

[0074] Compared with the prior art, the present invention has the following beneficial effects:

[0075] The application of quasi-unmanned highway systems and related key technologies has significant economic, technical, social and environmental benefits, mainly manifested in the following aspects:

[0076] (1) Social benefit analysis

[0077] The quasi-unmanned highway tunnel system proposes the addition of tunnel management substations to its management architecture. This invention optimizes the location of these substations, significantly improving emergency response capabilities to tunnel incidents, shortening rescue times and protecting public life and property. Furthermore, this system enhances the intelligence of tunnel management and can provide valuable insights for tunnel management in other regions.

[0078] (2) Technical benefit analysis

[0079] This invention, through the deep integration of deep learning algorithms and traffic engineering, has transformed and upgraded the industry's operational management model, transforming the traditional "experience-driven" model into a highly efficient "data-driven" one. This technological innovation not only enhances the intelligence of traffic management but also, through the establishment of a standardized technical framework, provides solid technical support for technological iteration and cross-industry integration in the highway transportation sector.

[0080] (3) Environmental benefit analysis

[0081] By scientifically optimizing the siting of tunnel management substations, this invention helps operators precisely determine the optimal location, significantly shortening the coverage area for inspections and emergency responses. In the event of an emergency on a highway, emergency personnel can quickly arrive at the scene, effectively curb the spread of fire, and quickly clear traffic, reducing vehicle delays. This, in turn, reduces pollutant emissions, improving overall emergency response efficiency and environmental benefits.

[0082] (4) Economic benefit analysis

[0083] By optimizing the tunnel operation and management system and selecting the optimal location for tunnel management substations, this invention significantly improves emergency rescue efficiency for highway accidents and reduces direct and indirect economic losses caused by traffic accidents. Furthermore, by replacing existing building-based tunnel substations with intelligent substations, intelligent management substations, intelligent firefighting substations, intelligent diesel generators, and modular IoT water tanks, this approach addresses the large land requirements and land acquisition challenges of existing substations. It also significantly reduces costs for substation site preparation, substation rooms, and specialized structures, shortening the construction period. Furthermore, the use of a series of intelligent microstations enables unmanned tunnel maintenance, eliminating the need for large numbers of on-duty personnel and significantly reducing long-term operational human resource costs. For example, the Xuanhui Expressway in Yunnan Province, spanning approximately 96.7 km and comprising 30 tunnels, has been estimated to have saved approximately 43.8489 million yuan during the construction period, a 21.67% reduction compared to the traditional approach of building substations and elevated water tanks. The implementation of unmanned maintenance can save approximately 13.716 million yuan in labor costs annually during operation. BRIEF DESCRIPTION OF THE DRAWINGS

[0084] Figure 1 This is a schematic diagram of the structure of the original expressway tunnel management system;

[0085] Figure 2 This is a schematic structural diagram of a quasi-unmanned system for a highway tunnel according to the present invention;

[0086] Figure 3 It is a hierarchical model diagram;

[0087] Figure 4 This is the flow chart of the improved salp swarm optimization algorithm based on Cat chaos mapping and hierarchical analysis method. DETAILED DESCRIPTION

[0088] The present invention is described in further detail below with reference to the embodiments.

[0089] Those skilled in the art will understand that the following examples are intended to illustrate the present invention only and should not be construed as limiting the scope of the present invention. Where specific techniques or conditions are not specified in the examples, the techniques or conditions described in the literature in the art or in the product specifications were used. Materials or equipment used without manufacturer identification are commercially available conventional products.

[0090] Example 1

[0091] The site selection method for tunnel management substations under the quasi-unmanned system of expressway tunnels is to set up tunnel management substations below the tunnel management station; the principle is that the total inspection distance of a single tunnel management station or tunnel management substation shall not exceed 30km, and the inspection distance on one side shall not exceed 15km;

[0092] The location of the tunnel management substation is selected using an improved salp swarm optimization algorithm based on Cat chaos mapping and hierarchical analysis method.

[0093] The location of the tunnel management substation is selected using an improved salp swarm optimization algorithm based on Cat chaos mapping and analytic hierarchy process. The specific steps include:

[0094] Step 1: Set the population size, number of iterations, and upper and lower bounds of the search space.

[0095] Step 2: Initialize the positions of individual salps and the salp population within the search space.

[0096] Among them, the individuals of the salp group are represented as:

[0097] in, The individual number of the salp is Candidate locations, For the The pile number information of each candidate point, For the The total length of the tunnel from the candidate point to the accident site, For the The number of extra-long tunnels from each candidate point to the accident site, For the The total length of the long downhill slope from the candidate point to the accident site, For the The number of small radius curves from the candidate point to the accident site, For the The total length of the road sections that are prone to rain and fog from the candidate points to the accident site, For the The total length of the frequently icy and snowy road sections from the candidate points to the accident site;

[0098] The characteristic information of n individuals in the salp group is stored in of Matrix, as shown in formula (2):

[0099]

[0100] Where n represents the population size of salps, =1,2,…,n; is the j-th dimension feature of the individual, d is the number of features, j = 1, 2, ..., d; For individual salps, is the position of the characteristic variable of the i-th salp in the j-dimensional space, and the matrix That is the salp population;

[0101] Salp populations mapped by Cat The initial position of is given by formula (4):

[0102]

[0103] in, is the initial position of the i-th salp on the j-th dimension feature variable, is the lower bound of the search space for the j-th dimension feature variable, is the upper bound of the search space of the j-th dimension feature variable, is the final ordinate chaotic sequence value of the i-th individual and the j-th dimension characteristic variable after k iterations;

[0104] Step 3: construct a fitness function and calculate the fitness value of each salp individual at the current position;

[0105] The fitness function is a function with the minimum emergency rescue time as the goal , as shown in formula (6):

[0106]

[0107]

[0108] in, represents a tunnel management node, which includes a tunnel management station or a tunnel management substation; Indicates the point where the accident occurred. It is from Tunnel management node to the The emergency distance of each accident point, for The weight value of is in the range of (0,1); It is from Tunnel management node to the The total length of the tunnels passed by the accident points is for The weight value of is in the range of (0,1); It is from Tunnel management node to the The number of tunnels that the accident point passes through, for The weight value of is in the range of (0,1); It is from Tunnel management node to the The length of the long downhill section that the accident point passes through, for The weight value of is in the range of (0,1); It is from Tunnel management node to the The number of small radius curves that the accident point passes through, for The weight value of is in the range of (0,1); It is from Tunnel management node to the The length of the rainy and foggy road section where the accident point passes, for The weight value of is in the range of (0,1); It is from Tunnel management node to the The length of the icy and snowy road section where the accident point passes, for The weight value of is in the range of (0,1);

[0109] The distance between any two tunnel management nodes should be 15 km, and the total inspection distance of a single tunnel management station or tunnel management substation must not exceed 30 km;

[0110] The individual fitness value is the inverse of its corresponding objective function value, as shown in formula (11):

[0111]

[0112] Step 4: Sort the salps individuals in descending order of fitness value, select the individuals with the largest fitness value as the food source, set the top 50% of the salps in fitness as leaders, and the rest of the salps as followers;

[0113] Step 5, update the position of the salp group;

[0114] In a salp group, the position update formula of the leader salp is shown in formula (12):

[0115]

[0116] in, is the position of leader s in the j-th dimension characteristic variable, s = 1, 2, …, n; is the location of the food source in the jth dimension characteristic variable, and are all random numbers between [0,1]. is the convergence factor;

[0117] The update formula is shown in formula (13):

[0118]

[0119] in, Indicates the current iteration number, Indicates the maximum number of iterations;

[0120] In a salp group, the position of the follower salp is updated according to Newton's law of motion, as shown in Equation (17):

[0121]

[0122] in, , Indicates the The position of each salp in the j-th dimension of the characteristic variable; Indicates the The position of the previous salp of the salp in the j-th dimension feature variable;

[0123] Step 6, calculating the fitness value of each salp according to the fitness function, and re-determining the food source location;

[0124] Step 7: Determine whether the number of algorithm iterations meets the stopping condition. If the stopping condition is not met, return to step 5, update the position of the salp group, and continue to iterate;

[0125] After the iteration is terminated, the position of the leader salp is output, that is, the optimal value of the candidate location of the tunnel management substation is obtained.

[0126] In step 3, The value of is determined by the analytic hierarchy process.

[0127] Example 2

[0128] The site selection method for tunnel management substations under the quasi-unmanned system of expressway tunnels is to set up tunnel management substations below the tunnel management station; the principle is that the total inspection distance of a single tunnel management station or tunnel management substation shall not exceed 30km, and the inspection distance on one side shall not exceed 15km;

[0129] The location of the tunnel management substation is selected using an improved salp swarm optimization algorithm based on Cat chaos mapping and hierarchical analysis method.

[0130] The location of the tunnel management substation is selected using an improved salp swarm optimization algorithm based on Cat chaos mapping and analytic hierarchy process. The specific steps include:

[0131] Step 1: Set the population size, number of iterations, and upper and lower bounds of the search space.

[0132] Step 2: Initialize the positions of individual salps and the salp population within the search space.

[0133] Among them, the individuals of the salp group are represented as:

[0134] in, The individual number of the salp is Candidate locations, For the The pile number information of each candidate point, For the The total length of the tunnel from the candidate point to the accident site, For the The number of extra-long tunnels from each candidate point to the accident site, For the The total length of the long downhill slope from the candidate point to the accident site, For the The number of small radius curves from the candidate point to the accident site, For the The total length of the road sections that are prone to rain and fog from the candidate points to the accident site, For the The total length of the frequently icy and snowy road sections from the candidate points to the accident site;

[0135] The characteristic information of n individuals in the salp group is stored in of Matrix, as shown in formula (2):

[0136]

[0137] Where n represents the population size of salps, =1,2,…,n; is the j-th dimension feature of the individual, d is the number of features, j = 1, 2, ..., d; For individual salps, is the position of the characteristic variable of the i-th salp in the j-dimensional space, and the matrix That is the salp population;

[0138] Salp populations mapped by Cat The initial position of is given by formula (4):

[0139]

[0140] in, is the initial position of the i-th salp on the j-th dimension feature variable, is the lower bound of the search space for the j-th dimension feature variable, is the upper bound of the search space of the j-th dimension feature variable, is the final ordinate chaotic sequence value of the i-th individual and the j-th dimension characteristic variable after k iterations;

[0141] Step 3: construct a fitness function and calculate the fitness value of each salp individual at the current position;

[0142] The fitness function is a function with the minimum emergency rescue time as the goal , as shown in formula (6):

[0143]

[0144]

[0145] in, represents a tunnel management node, which includes a tunnel management station or a tunnel management substation; Indicates the point where the accident occurred. It is from Tunnel management node to the The emergency distance of each accident point, for The weight value of is in the range of (0,1); It is from Tunnel management node to the The total length of the tunnels passed by the accident points is for The weight value of is in the range of (0,1); It is from Tunnel management node to the The number of tunnels that the accident point passes through, for The weight value of is in the range of (0,1); It is from Tunnel management node to the The length of the long downhill section that the accident point passes through, for The weight value of is in the range of (0,1); It is from Tunnel management node to the The number of small radius curves that the accident point passes through, for The weight value of is in the range of (0,1); It is from Tunnel management node to the The length of the rainy and foggy road section where the accident point passes, for The weight value of is in the range of (0,1); It is from Tunnel management node to the The length of the icy and snowy road section where the accident point passes, for The weight value of is in the range of (0,1);

[0146] The distance between any two tunnel management nodes should be 15 km, and the total inspection distance of a single tunnel management station or tunnel management substation must not exceed 30 km;

[0147] The individual fitness value is the inverse of its corresponding objective function value, as shown in formula (11):

[0148]

[0149] Step 4: Sort the salps individuals in descending order of fitness value, select the individuals with the largest fitness value as the food source, set the top 50% of the salps in fitness as leaders, and the rest of the salps as followers;

[0150] Step 5, update the position of the salp group;

[0151] In a salp group, the position update formula of the leader salp is shown in formula (12):

[0152]

[0153] in, is the position of leader s in the j-th dimension characteristic variable, s = 1, 2, …, n; is the location of the food source in the jth dimension characteristic variable, and are all random numbers between [0,1]. is the convergence factor;

[0154] The update formula is shown in formula (13):

[0155]

[0156] in, Indicates the current iteration number, Indicates the maximum number of iterations;

[0157] In a salp group, the position of the follower salp is updated according to Newton's law of motion, as shown in Equation (17):

[0158]

[0159] in, , Indicates the The position of each salp in the j-th dimension of the characteristic variable; Indicates the The position of the previous salp of the salp in the j-th dimension feature variable;

[0160] Step 6, calculating the fitness value of each salp according to the fitness function, and re-determining the food source location;

[0161] Step 7: Determine whether the number of algorithm iterations meets the stopping condition. If the stopping condition is not met, return to step 5, update the position of the salp group, and continue to iterate;

[0162] After the iteration is terminated, the position of the leader salp is output, that is, the optimal value of the candidate location of the tunnel management substation is obtained.

[0163] In step 3, The value of is determined by the analytic hierarchy process.

[0164] The quasi-unmanned system for highway tunnels includes six levels: the first level is the highway transportation management department, the second level is the management office, the third level is the management sub-office station, the fourth level is the tunnel management station, the fifth level is the tunnel management sub-station, and the sixth level is the intelligent micro-station. The intelligent micro-station is set up at the tunnel management station or tunnel management sub-station.

[0165] The intelligent micro station includes an intelligent substation, an intelligent management micro station, an intelligent fire protection micro station, a diesel generator box and a modular IoT water tank;

[0166] The intelligent substation is used to ensure the power supply of the tunnel and to perform intelligent management and control of the power supply of the tunnel;

[0167] The intelligent management microstation is used to manage tunnel lighting facilities, ventilation facilities, monitoring facilities, fire protection facilities, as well as intelligent substations, intelligent management microstations, intelligent fire protection microstations, diesel generator boxes, and modular IoT water tanks. The intelligent management microstation is also equipped with necessary emergency supplies.

[0168] The intelligent fire fighting microstation is used to draw water from a modular IoT water tank or high-level water pool to provide stable water pressure for tunnel fire fighting, and to monitor the water pressure, liquid level, and pump operating status of the fire fighting water supply system in real time.

[0169] The diesel generator box is used to monitor the working status, fuel level, faults and operating environment information of the diesel generator, and control the start or stop of the diesel generator according to the mains power on / off status and inspection requirements. The diesel generator serves as a backup power source for the mains power supply system;

[0170] The modular IoT water tank is composed of several independent water storage units, and is equipped with a liquid level detector and a water pressure detector;

[0171] The modular IoT water tank is used as a fire water supply system for tunnels with a length of less than or equal to 5,000 meters. At the same time, the modular IoT water tank can monitor its own liquid level and pressure information in real time.

[0172] Tunnels with a length greater than 5,000m use high-level water tanks to provide fire water sources at constant high pressure. Smart fire microstations provide stable water pressure for tunnel fire protection and monitor water pressure and flow information. Tunnels with a length less than or equal to 5,000m use a temporary high-pressure water supply system that combines smart fire microstations and modular IoT water tanks. The modular IoT water tanks serve as fire water sources, and the smart fire microstations provide stable water pressure and monitor water pressure and flow information. The modular IoT water tanks are installed at the tunnel entrance.

[0173] Tunnel management substations are built together with toll stations or service areas along the highway, or are set up at the entrances of extra-long tunnels.

[0174] The intelligent substation conducts intelligent management and control of the tunnel's power supply, including real-time power monitoring of the voltage, current, and power information of the power supply and distribution circuits of the high-voltage, transformer, and low-voltage cabinets, and remote control of the high-voltage distribution circuit, low-voltage lighting circuit, low-voltage ventilation circuit, fire-fighting equipment, and air conditioning. At the same time, it can analyze the energy consumption of the tunnel's lighting, ventilation, fire-fighting, dynamic environment, and video surveillance circuits.

[0175] Example 3

[0176] 1. Optimized design of tunnel substation

[0177] For tunnel substations, following the principles of "standardized design, intelligent configuration, factory-based production, prefabricated construction, and information-based management," an integrated "three-station, two-tank" solution was developed based on electrical equipment pre-installation technology, prefabricated cabin main structure load analysis technology, intelligent sensing technology, Internet of Things technology, communications technology, intelligent control technology, big data analysis technology, and safety protection technology to replace and upgrade the relevant functions of traditional building-based substations. These include a prefabricated cabin-type intelligent substation microstation, an intelligent management microstation, an intelligent fire protection microstation, a diesel generator box, and a modular Internet of Things water tank. The functions of each microstation are as follows:

[0178] (1) Intelligent substation: While ensuring the safe operation and stable power supply of the tunnel, the system can realize intelligent management and control of the tunnel power supply by analyzing the energy consumption of the lighting, ventilation, fire protection, dynamic environment and video surveillance circuits of the tunnel through real-time power monitoring of the voltage, current and power information of the power distribution circuits of the high-voltage, transformer and low-voltage cabinets, remote control of the high-voltage distribution circuit, low-voltage lighting circuit, low-voltage ventilation circuit, fire-fighting equipment and air conditioning.

[0179] (2) Intelligent Management Micro-station: This system realizes intelligent management of tunnel lighting, ventilation, monitoring, fire protection, and security facilities. Through real-time monitoring, safety warning, equipment management, and environmental monitoring, the operation and management of the “three stations and two containers” are digitalized and intelligent. It is equipped with necessary emergency supplies to comprehensively guarantee the safe operation of the tunnel and the “three stations and two containers”.

[0180] (3) Intelligent fire fighting microstation: It is used to draw water from the modular IoT water tank or high-level water pool to provide stable water pressure for tunnel fire fighting, and to monitor the water pressure, liquid level, and working status of the fire fighting water supply system in real time to ensure the safety, reliability and fire extinguishing efficiency of the tunnel fire fighting water supply system.

[0181] (4) Diesel generator box: When the power grid is cut off, the diesel generator is automatically started to generate electricity and the power supply is automatically switched to the diesel generator. In addition, the diesel generator box is equipped with intelligent monitoring equipment, which can monitor the working status of the diesel generator and fuel level information in real time, and can remotely control the start and stop of the diesel generator.

[0182] (5) Modular IoT water tank: In the traditional design, a high-level fire water tank is set up at the entrance (exit) of the tunnel. Considering the unstable geological conditions at the entrance (exit) of the tunnel, the steep terrain in the cave entrance area makes it difficult to lay out the water tank and the construction is difficult. At the same time, in order to avoid the high-level fire water tank from encroaching on ecological forest land, ecological red line, water source protection area and basic farmland, the optimized design scheme combines the length of each tunnel, the tunnel fire safety level, actual engineering experience and on-site terrain conditions. Under the premise of meeting the driving safety during the operation period, the tunnel fire water supply system is optimized: according to the tunnel fire water consumption, a modular water tank is formed by combining several 1m³ independent water storage units. The fire hydrant water supply mode for tunnels less than or equal to 5000m is adjusted from normal high-pressure water supply to temporary high-pressure water supply. Tunnels with a length of more than 5000m have high maintenance and operation safety requirements. In order to ensure the reliability and stability of the tunnel water supply, the fire hydrant water supply mode maintains the original design scheme, adopting a normal high-pressure water supply system, and extracting water through the intelligent fire micro station to provide a stable fire water flow for tunnel fire fighting. For tunnels 5,000 meters or less in length, after eliminating the elevated water tank, a modular IoT water tank, along with a smart firefighting microstation, is installed at the tunnel entrance to provide a temporary high-pressure water supply for tunnel firefighting. The capacity of the modular IoT water tank is calculated based on industry firefighting standards and tunnel length.

[0183] By replacing traditional building-based tunnel substations with an intelligent "three stations, two boxes" solution, the construction of buildings and personnel is eliminated. This reduces costs, shortens construction time, and reduces complexity during the construction phase; and reduces expenses and improves efficiency during the operational phase. The "three stations, two boxes" solution provides key technical support for the tunnel's quasi-unmanned operation system, enabling real-time monitoring, comprehensive management, and intelligent operation and maintenance.

[0184] 2. Optimization plan for the construction, management and maintenance system of expressway tunnels

[0185] Regarding the highway tunnel management system, building on the existing system, an innovative concept has been proposed: a quasi-unmanned system and tunnel management substations. Tunnel management substations are established below tunnel management stations. These are co-located with toll booths and service areas along the expressway, or located at the entrances of extra-long tunnels. Three to four staff members are stationed at the substations, responsible for some of the emergency response, management, and inspection duties of the original tunnel management stations. Under the original system, tunnel management stations generally had a large jurisdiction and long inspection distances, some exceeding 50 km. For highways in the Yunnan-Guizhou Plateau, the high bridge-tunnel ratio, the preponderance of extra-long tunnels, and the long inspection distances create significant inspection and emergency response pressure for staff, significantly prolonging emergency response times and posing safety risks to drivers and passengers. The quasi-unmanned management system aims to limit the total inspection distance of a single tunnel management station or tunnel management substation to no more than 30 km, with a single-sided inspection distance of no more than 15 km. This aims to maximize the timeliness of highway emergency response, shorten inspection distances, and reduce inspection pressure. If there are fewer tunnels and better road conditions within the jurisdiction, that is, fewer bends, long downhill slopes, rain, fog, and ice and snow sections, the inspection distance of the station can be appropriately increased; otherwise, the inspection distance can be appropriately shortened. In addition, the original building-type tunnel substation is optimized to "three stations and two boxes". Figure 2 shown.

[0186] Under the quasi-unmanned system of highway tunnels, setting up tunnel management substations at appropriate points along the highway is conducive to reducing the inspection pressure of tunnel management stations and improving the timeliness and safety of emergency rescue. However, the selection of the setting points of tunnel management substations is affected by many factors, such as emergency distance, tunnel group location, extra-long tunnel location, long downhill location, multi-bend road section location, small radius bend location, rainy and foggy road section location, icy and snowy road section location and tunnel management station location. Therefore, when setting up tunnel management substations, it is necessary to fully consider the above factors to avoid the subjectivity and randomness of site selection to ensure that the tunnel management substation is optimally located and the substation plays its maximum role. Therefore, the present invention adopts an improved Salp Sea Squid swarm optimization algorithm for modeling and solving, which provides a reference for the optimal location selection of tunnel management substations.

[0187] The salp swarm algorithm is inspired by the collective foraging behavior of marine organisms called salps. These creatures do not gather in "swarms" but rather form "chains," with each salp connected end to end, following the movement of the leader. A leader at the front of the salp chain explores the environment and searches for food, while followers follow behind. The leader leads only its immediate followers and does not influence the rest of the population. Therefore, the leader's influence on the following followers gradually decreases, maintaining a good diversity among the following individuals.

[0188] In the salp swarm optimization algorithm, individuals represent possible solutions, while the swarm represents a subset of the solution space. Individuals achieve co-evolution by simulating interactions between salps. The algorithm adjusts the positions of individuals to find the optimal solution to the problem while preserving the diversity of the swarm to avoid falling into local optima.

[0189] Algorithm steps:

[0190] 1. Individual representation:

[0191] Each individual represents a possible tunnel management substation location. The candidate location is any point along the highway. The feature set of each individual includes the pile number information, as well as the total length of tunnels from the point to the accident site, the number of extra-long tunnels, the total length of long downhill sections, the number of small-radius curves, the total length of sections frequently experiencing rain and fog, and the total length of sections frequently experiencing ice and snow. This is expressed by formula (1):

[0192] in, The individual number of the salp is Candidate locations, For the The pile number information of each candidate point, For the The total length of the tunnel from the candidate point to the accident site, For the The number of extra-long tunnels from each candidate point to the accident site, For the The total length of the long downhill slope from the candidate point to the accident site, For the The number of small radius curves from the candidate point to the accident site, For the The total length of the road sections that are prone to rain and fog from the candidate points to the accident site, For the The total length of frequently icy and snowy road sections from the candidate points to the accident site.

[0193] 2. Initialize the salp population:

[0194] The initialization individuals are randomly generated in the search space. If the search space to be solved is d-dimensional, the characteristic information of n individuals of the salp group is stored in of Matrix, as shown in formula (2):

[0195]

[0196] Where n represents the population size of salps, =1,2,…,n; is the j-th dimension feature of the individual, d is the number of features, j = 1, 2, ..., d; For individual salps, is the position of the characteristic variable of the i-th salp in the j-dimensional space, and the matrix For roads with fixed tunnel management station locations, a corresponding number of salps are set aside in the initial population to represent the location of the tunnel management station. During subsequent population updates, the salps representing the tunnel management station are not updated.

[0197] In order to improve the efficiency and accuracy of the search, the uniformity of the first-generation population distribution has an important impact on the convergence speed. Conventional algorithms use the method of directly generating random populations, which may cause some individuals to move away from the optimal position or fall into local optimal problems, prolonging the convergence speed. Therefore, it is extremely necessary to adopt a more efficient initialization method to make the initial population as evenly distributed as possible in the search space.

[0198] Chaotic sequences have the characteristics of randomness, ergodicity and regularity, and can produce uniformly distributed populations, and the salp populations generated by them have good diversity. At present, the commonly used chaotic sequences are Logistic mapping, Iterative mapping, Cat mapping and Tent mapping, among which the Logistic mapping value is not uniform enough; Iterative mapping is extremely sensitive to initial values, and small gaps may be magnified during the iteration process, leading to different results; Tent mapping is prone to falling into a cycle within a small range, and may also produce unstable periodic points; Cat mapping has good ergodicity, which can make the state points evenly distributed in space, which is especially beneficial in problems that require global search. Therefore, the present invention adopts the Cat mapping method to increase the diversity and distribution uniformity of the initialized population, thereby improving the efficiency of the algorithm. The specific steps are as follows:

[0199] Cat mapping is a reversible two-dimensional chaotic mapping, and its mathematical expression is shown in formula (3):

[0200]

[0201] Where i is the individual number, For individual dimensional features, k is the number of iterations, is the horizontal coordinate value of the chaotic sequence of the i-th individual and the j-th dimension variable at the k-th iteration, is the ordinate value of the chaotic sequence of the i-th individual and the j-th dimension variable at the k-th iteration. The present invention selects Sequence to generate the first generation population, The corresponding population can be obtained in the solution space by performing inverse mapping. The initial position of the salp population after Cat mapping As shown in formula (4):

[0202]

[0203] in, is the initial position of the i-th individual on the j-th dimension characteristic variable, is the lower bound of the search space for the j-th dimension feature variable, is the upper bound of the search space of the j-th dimension feature variable, is the final ordinate chaotic sequence value of the i-th individual and the j-th dimension characteristic variable after k iterations, is the search space, i.e. the entire highway.

[0204] 3. Fitness function:

[0205] A decisive factor in the site selection of tunnel management substations is the timeliness of emergency rescue. This factor is crucial for ensuring safe passage on expressways, enabling timely handling of traffic accidents, preventing them from escalating, and minimizing loss of life and property. The timeliness of emergency rescue on expressways is influenced by multiple factors, including the distance between the accident site and the tunnel management station or substation, the number of tunnel clusters involved in the emergency, the number of long or extra-long tunnels, the length of long and steep downhill slopes, the number of tight-radius curves, and the length of rainy, foggy, and icy sections. Therefore, when selecting a tunnel management station, the optimal location should be determined by comprehensively considering these factors across all candidate locations along the expressway.

[0206] The inspection pressure on highways mainly comes from the inspection distance, the number of inspection tunnel groups, and the length of long or extra-long tunnels. The factors affecting inspection pressure are included in the factors affecting the timeliness of emergency rescue. Therefore, the location of the tunnel management substation selected according to the key influencing factors of the timeliness of emergency rescue can also meet the minimum overall inspection pressure on the highway.

[0207] When a traffic accident, fire, landslide, or mudslide occurs on a highway, the timeliness of emergency rescue is affected by the emergency distance, the length of the tunnels passed, the number of tunnels, the length of the long downhill slope, the number of curves, the length of the rainy and foggy sections, and the length of the icy and snowy sections, as shown in formula (5):

[0208]

[0209] in, represents a tunnel management node, including a tunnel management station or a tunnel management substation, wherein the location of the tunnel management station is determined during planning and design, and the location of the tunnel management substation is determined by the tunnel management substation location selection method under the quasi-unmanned system for highway tunnels of the present invention. Indicates the point where the accident occurred. It is from Tunnel management node to the Emergency distance to each accident point; It is from Tunnel management node to the The total length of tunnels passed by each accident point; It is from Tunnel management node to the The number of tunnels that each accident site passes through; It is from Tunnel management node to the The length of the long downhill section that the accident point passes through; It is from Tunnel management node to the The number of small-radius curves that each accident point passes through; It is from Tunnel management node to the The length of the rainy and foggy road section where each accident point passes; It is from Tunnel management node to the The length of the icy and snowy road section passed by the accident site.

[0210] In order to shorten the time from the tunnel management station or tunnel management substation to the accident site as much as possible and improve the timeliness of emergency rescue, based on the above factors, a function with the minimum emergency rescue time as the goal is constructed. , as shown in formula (6):

[0211]

[0212]

[0213] in, represents a tunnel management node, including a tunnel management station or a tunnel management substation, wherein the location of the tunnel management station is determined during planning and design, and the location of the tunnel management substation is determined by the tunnel management substation location selection method under the quasi-unmanned system for highway tunnels of the present invention. Indicates the point where the accident occurred. It is from Tunnel management node to the The emergency distance of each accident point, for The weight value of is in the range of (0,1); It is from Tunnel management node to the The total length of the tunnels passed by the accident points is for The weight value of is in the range of (0,1); It is from Tunnel management node to the The number of tunnels that the accident point passes through, for The weight value of is in the range of (0,1); It is from Tunnel management node to the The length of the long downhill section that the accident point passes through, for The weight value of is in the range of (0,1); It is from Tunnel management node to the The number of small radius curves that the accident point passes through, for The weight value of is in the range of (0,1); It is from Tunnel management node to the The length of the rainy and foggy road section where the accident point passes, for The weight value of is in the range of (0,1); It is from Tunnel management node to the The length of the icy and snowy road section where the accident point passes, for The weight value ranges from (0 to 1). Any two tunnel management nodes should be at least 15 km apart, and the total inspection distance of a single tunnel management station or tunnel management substation must not exceed 30 km. The ultimate goal is to find the location of a tunnel management substation that minimizes the above objective function.

[0214] Since the objective function is affected by There are seven influencing factors in total, and the magnitude and dimension of each influencing factor are different. Therefore, it is not possible to simply solve the problem by weighting and summing the factors. In order to accurately reflect the degree of influence of each factor on the target, each factor is normalized according to formula (7):

[0215]

[0216] Among them, h is the original data value of the impact factor, is the minimum observed value of the factor among all possible values, is the maximum observed value of the factor among all possible values, The normalized value of this factor is the dimensionless value after mapping the original value to the interval [0,1].

[0217] in The value of is determined by the Analytical Hierarchy Process (AHP), with the following steps:

[0218] (1) Establishing a hierarchical model

[0219] The model is divided into target class and criterion class that affects the target. By constructing a hierarchical structure of the interrelationships between factors, each factor is weighted, and then the relative weights are calculated to obtain the order of merit. Figure 3 shown.

[0220] (2) Constructing a judgment matrix

[0221] The 1-9 scale method proposed by Professor Satty is used to describe the importance of two factors, and multiple indicators are compared pairwise. express and Towards the target The impact ratio of , Form a judgment matrix. The comparison rules are shown in Table 1:

[0222] Table 1 1-9 scale definition table

[0223]

[0224] Experts score according to the above table to form a judgment matrix as shown in formula (8):

[0225]

[0226] The score indicating the emergency distance and the importance of the emergency distance, A score indicating the importance of the emergency distance compared to the length of the tunnel passed. A score indicating the importance of emergency distance compared to the number of tunnels, A score indicating the importance of the emergency distance compared to the length of the long downhill slope, A score indicating the importance of emergency distance compared to the number of curves, The score indicating the importance of emergency distance compared to the length of rainy and foggy road sections, The score represents the importance of the emergency distance compared to the length of the icy and snowy road section. Similarly, the importance of each of the seven important factors is compared in pairs.

[0227] (3) Hierarchical single sorting and consistency test

[0228] Calculate the square root of the product of all elements in each row of the judgment matrix, as shown in formula (9):

[0229]

[0230] in, For the matrix The geometric mean of the row elements, After normalization according to formula (10), we can get The value of .

[0231]

[0232] for The weight value of for The weight value of for The weight value of for The weight value of for The weight value of for The weight value of for The weight value of .

[0233] When combining the salp swarm optimization algorithm to solve the minimization problem, the minimum value of the objective function corresponds to the maximum value of the fitness function, so the individual fitness value is the inverse of its corresponding objective function value, as shown in formula (11):

[0234]

[0235] For the The fitness values ​​of n salps are sorted from large to small according to their fitness values, and the individual with the largest fitness value is selected as the food source, which is the current optimal tunnel management substation location. The current optimal value is recorded in the optimal solution set of the tunnel management substation candidate points to provide a decision reference for the final location selection.

[0236] 4. Collaborative behavior simulation:

[0237] Individual salps are divided into two roles: leaders and followers. These individuals can share information, such as the distribution of tunnels and the location of specific tunnels. The leader, located at the front of the salp chain, guides the movement of the entire group; the followers, located at the back of the chain, follow the leader.

[0238] 5. Update the location of the salp group:

[0239] The position update formula of the leader salp is shown in formula (12):

[0240] In a salp group, the position update formula of the leader salp is shown in formula (12):

[0241]

[0242] in, is the position of leader s in the j-th dimension characteristic variable, s = 1, 2, …, n; is the location of the food source in the jth dimension characteristic variable, and Both are random numbers between [0,1]. The former represents the step factor, and the latter mainly determines the direction of leader movement. is the convergence factor, which is the key parameter for the algorithm to balance global search and local search. When it is greater than or equal to 1, the algorithm conducts global exploration, and when it is less than 1, the algorithm conducts local development. As the number of iterations increases, It gradually decreases from 2 to 0, and its update formula is shown in formula (13):

[0243]

[0244] in, Indicates the current iteration number, Indicates the maximum number of iterations. hour, , enhance global exploration, as the current iteration number increases, Gradually decreases when hour, .

[0245] The position of the follower salp is updated according to Newton's law of motion, as shown in Equation (14):

[0246]

[0247] in , Indicates the The position of the follower salp at the j-th dimension feature variable. Indicates the The position of the previous salp of the salp in the j-th dimension feature variable, For time, is the initial velocity, acceleration As shown in formula (15):

[0248]

[0249] in It is expressed by formula (16):

[0250]

[0251] Due to time is the difference between two iteration times, which can be taken as unit time, that is, , and the initial velocity , so formula (14) can be expressed as:

[0252]

[0253] When the number of iterations does not reach the maximum number of iterations, the behavior mechanism of the salp chain is simulated by using Equations (12) and (17) to update the positions of the leader, the following salps, and the food source.

[0254] 6. Re-evaluate fitness:

[0255] After the salp population is updated, the fitness function is used to calculate the fitness value of each salp, and the location of the food source is re-determined.

[0256] 7. Iteration stop judgment:

[0257] Determine whether the number of algorithm iterations meets the stopping condition. If the stopping condition is not met, continue to execute the iteration.

[0258] After the iteration ends, the optimal solution set for the candidate points stores the optimal values ​​of the population after each update. The later the value stored in the solution set, the higher the fitness value, and the stronger its reference for selecting the location of the tunnel management substation. Based on road characteristics and actual needs, one or more points with high fitness from the optimal solution set can be selected to establish a tunnel management substation.

[0259] The algorithm flow is as follows Figure 4 shown.

[0260] Application Examples

[0261] The proposed quasi-unmanned highway tunnel system and related key technologies have been implemented in Menglu (Mengxing to Jiangcheng to Lvchun Expressway) in Yunnan Province. The system will be gradually implemented on the Xuanhui Expressway (Xuanwei to Huize), Huiqiao Expressway (Huize to Qiaojia), Anchu Expressway (Anning to Chuxiong), and Shiqiu Expressway (Shizong to Qiubei). Using the Dali section of the Dayong Expressway as an example, the key steps for implementing the quasi-unmanned system management solution and key technologies are as follows:

[0262] The Dali section of the Dayong Expressway starts in Haidong Town, Dali City, and ends in Pianjiao Town, Yongsheng County, Lijiang City, connecting with the Dali Expressway. It is 53 km long and has a total of 8 tunnels. It is under the unified management of the Qingshan Tunnel Management Station. The facilities along the line are shown in Table 2:

[0263] Table 2 Facilities along the Dali section of Dayong Expressway

[0264]

[0265] In the original design, the Qingshan Tunnel Management Station had a wide jurisdiction, with tunnels mainly concentrated in the first 16 km, including one extra-long tunnel, two long tunnels, three medium tunnels, and one short tunnel. The high density of tunnels made it difficult to ensure timely emergency rescue and put a lot of pressure on inspections. Based on the proposed tunnel management substation site selection method for the quasi-unmanned system of highway tunnels, the following is derived by improving the salp group optimization algorithm:

[0266] 1. From Table 2, we can see that the stake number of the candidate point The search space is the Dali section of Dayong Expressway. , According to the setting principle, the distance between a tunnel management station and a tunnel management substation should not be less than 15km. Therefore, only Waxi Toll Station, Binchuan Toll Station and Lijiao Toll Station are candidate locations for tunnel management substations on this section of road.

[0267] 2. Assuming an accident occurs at the exit of Pulianpeng Tunnel, the initial characteristic parameters of each candidate station are obtained based on the actual characteristics of the road section as shown in Table 3:

[0268] Table 3 Initial feature parameters of candidate points

[0269]

[0270] According to formula (2), the initial matrix of the salp population can be obtained as:

[0271]

[0272] 3. Calculate the influence weight of each characteristic factor through the hierarchical analysis method. Compare all factors pairwise based on the 1-9 scale to obtain the judgment matrix of the factors at this level:

[0273]

[0274] According to formula (9) and (10), we can get: , , , , =0.10, , .

[0275] 4. Calculate the objective function of the initial candidate point according to formula (6). The objective function value of Waxi toll station is , the objective function value of Binchuan toll station is , the objective function value of the force angle toll station is .

[0276] 5. Calculate the initial fitness according to formula (11). The fitness of Waxi Toll Station is , the fitness of Binchuan toll station is , the fitness of Lijiao toll station is According to the fitness value, Waxi Toll Station is used as the food source to record the optimal solution set of candidate locations of the tunnel management substation.

[0277] 6. For highways where both the tunnel management station and the candidate points are uncertain, the positions of the leader and follower of the salp need to be updated according to Equations (12) and (14), then return to step 5, calculate the fitness value, and then record the point with the largest fitness into the optimal solution set until the maximum number of iterations is reached.

[0278] By comparing the adaptability of various candidate locations, the Dali section of Dayong Expressway finally chose to set up a tunnel management substation at the Waxi toll station. This can not only significantly improve the timeliness of emergency rescue on this section of road, but also greatly share the inspection pressure of the Qingshan Tunnel Management Station.

[0279] The basic principles, main features, and advantages of the present invention are shown and described above. Those skilled in the art should understand that the present invention is not limited to the foregoing embodiments. The foregoing embodiments and descriptions are merely illustrative of the principles of the present invention. Various changes and modifications may be made to the present invention without departing from the spirit and scope of the present invention. Such changes and modifications are intended to fall within the scope of the present invention. The scope of protection claimed in the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for selecting a tunnel management substation site under a quasi-unmanned system for a highway tunnel, characterized in that: Set up a tunnel management substation under the tunnel management station; The principle is that the total inspection distance of a single tunnel management station or tunnel management substation shall not exceed 30km, and the inspection distance on one side shall not exceed 15km; The location of the tunnel management substation is selected using an improved salp swarm optimization algorithm based on Cat chaos mapping and analytic hierarchy process. The location of the tunnel management substation is selected using an improved salp swarm optimization algorithm based on Cat chaos mapping and analytic hierarchy process. The specific steps include: Step 1: Set the population size, number of iterations, and upper and lower bounds of the search space. Step 2: Initialize the positions of individual salps and the salp population within the search space. Among them, the individuals of the salp group are represented as: , in, The individual number of the salp is Candidate locations, For the The pile number information of each candidate point, For the The total length of the tunnel from the candidate point to the accident site, For the The number of extra-long tunnels from each candidate point to the accident site, For the The total length of the long downhill slope from the candidate point to the accident site, For the The number of small radius curves from the candidate point to the accident site, For the The total length of the road sections that are prone to rain and fog from the candidate points to the accident site, For the The total length of the frequently icy and snowy road sections from the candidate points to the accident site; The characteristic information of n individuals in the salp group is stored in of Matrix, as shown in formula (2): ， Where n represents the population size of salps, =1,2,…,n; is the j-th dimension feature of the individual, d is the number of features, j = 1, 2, ..., d; For individual salps, is the position of the characteristic variable of the i-th salp in the j-dimensional space, and the matrix That is the salp population; Salp populations mapped by Cat The initial position of is given by formula (4): ， in, is the initial position of the i-th salp on the j-th dimension feature variable, is the lower bound of the search space of the j-th dimension feature variable, is the upper bound of the search space of the j-th dimension feature variable, is the final ordinate chaotic sequence value of the i-th individual and the j-th dimension characteristic variable after k iterations; Step 3: construct a fitness function and calculate the fitness value of each salp individual at the current position; The fitness function is a function with the minimum emergency rescue time as the goal , as shown in formula (6): ， ， in, represents a tunnel management node, which includes a tunnel management station or a tunnel management substation; Indicates the point where the accident occurred. It is from Tunnel management node to the The emergency distance of each accident point, for The weight value of is in the range of (0,1); It is from Tunnel management node to the The total length of the tunnels passed by the accident points is for The weight value of is in the range of (0,1); It is from Tunnel management node to the The number of tunnels that the accident point passes through, for The weight value of is in the range of (0,1); It is from Tunnel management node to the The length of the long downhill section that the accident point passes through, for The weight value of is in the range of (0,1); It is from Tunnel management node to the The number of small radius curves that the accident point passes through, for The weight value of is in the range of (0,1); It is from Tunnel management node to the The length of the rainy and foggy road section where the accident point passes, for The weight value of is in the range of (0,1); It is from Tunnel management node to the The length of the icy and snowy road section where the accident point passes, for The weight value of is in the range of (0,1); The distance between any two tunnel management nodes should be 15 km, and the total inspection distance of a single tunnel management station or tunnel management substation must not exceed 30 km; The individual fitness value is the inverse of its corresponding objective function value, as shown in formula (11): ， Step 4: Sort the salps individuals in descending order of fitness value, select the individuals with the largest fitness value as the food source, set the top 50% of the salps in fitness as leaders, and the rest of the salps as followers; Step 5, update the position of the salp group; In a salp group, the position update formula of the leader salp is shown in formula (12): ， in, is the position of leader s in the j-th dimension characteristic variable, s = 1, 2, …, n; is the location of the food source in the jth dimension characteristic variable, and are all random numbers between [0,1]. is the convergence factor; The update formula is shown in formula (13): ， in, Indicates the current iteration number, Indicates the maximum number of iterations; In a salp group, the position of the follower salp is updated according to Newton's law of motion, as shown in Equation (17): ， in, , Indicates the The position of the previous salp of the salp in the j-th dimension feature variable; Step 6, calculating the fitness value of each salp according to the fitness function, and re-determining the food source location; Step 7: Determine whether the number of algorithm iterations meets the stopping condition. If the stopping condition is not met, return to step 5, update the position of the salp group, and continue to iterate; After the iteration is terminated, the position of the leader salp is output, which means the optimal value of the candidate location of the tunnel management substation is obtained; In step 3, The value of is determined by the analytic hierarchy process.

2. The method for selecting a tunnel management substation in a quasi-unmanned system for a highway tunnel according to claim 1 is characterized in that: The quasi-unmanned system for highway tunnels includes six levels: the first level is the highway transportation management department, the second level is the management office, the third level is the management sub-office station, the fourth level is the tunnel management station, the fifth level is the tunnel management sub-station, and the sixth level is the intelligent micro-station. The intelligent micro-station is set up at the tunnel management station or tunnel management sub-station. The intelligent micro station includes an intelligent substation, an intelligent management micro station, an intelligent fire protection micro station, a diesel generator box and a modular IoT water tank; The intelligent substation is used to ensure the power supply of the tunnel and to perform intelligent management and control of the power supply of the tunnel; The intelligent management microstation is used to manage tunnel lighting facilities, ventilation facilities, monitoring facilities, fire protection facilities, as well as intelligent substations, intelligent management microstations, intelligent fire protection microstations, diesel generator boxes, and modular IoT water tanks. The intelligent management microstation is also equipped with necessary emergency supplies. The intelligent fire fighting microstation is used to draw water from a modular IoT water tank or high-level water pool to provide stable water pressure for tunnel fire fighting, and to monitor the water pressure, liquid level, and pump operating status of the fire fighting water supply system in real time. The diesel generator box is used to monitor the working status, fuel level, faults and operating environment information of the diesel generator, and control the start or stop of the diesel generator according to the mains power on / off status and inspection requirements. The diesel generator serves as a backup power source for the mains power supply system; The modular IoT water tank is composed of several independent water storage units, and is equipped with a liquid level detector and a water pressure detector; The modular IoT water tank is used as a fire water supply system for tunnels with a length of less than or equal to 5,000 meters. At the same time, the modular IoT water tank can monitor its own liquid level and pressure information in real time.

3. The method for selecting a tunnel management substation site under a quasi-unmanned system for a highway tunnel according to claim 2 is characterized in that: Tunnels with a length greater than 5,000m use high-level water tanks to provide fire water sources at constant high pressure. Smart fire microstations provide stable water pressure for tunnel fire protection and monitor water pressure and flow information. Tunnels with a length less than or equal to 5,000m use a temporary high-pressure water supply system that combines smart fire microstations and modular IoT water tanks. The modular IoT water tanks serve as fire water sources, and the smart fire microstations provide stable water pressure and monitor water pressure and flow information. The modular IoT water tanks are installed at the tunnel entrance.

4. The method for selecting a tunnel management substation site under a quasi-unmanned system for a highway tunnel according to claim 2 is characterized in that: Tunnel management substations are built together with toll stations or service areas along the highway, or are set up at the entrances of extra-long tunnels.

5. The method for selecting a tunnel management substation site under a quasi-unmanned system for a highway tunnel according to claim 2 is characterized in that: The intelligent substation conducts intelligent management and control of the tunnel's power supply, including real-time power monitoring of the voltage, current, and power information of the power supply and distribution circuits of the high-voltage, transformer, and low-voltage cabinets, and remote control of the high-voltage distribution circuit, low-voltage lighting circuit, low-voltage ventilation circuit, fire-fighting equipment, and air conditioning. At the same time, it can analyze the energy consumption of the tunnel's lighting, ventilation, fire-fighting, dynamic environment, and video surveillance circuits.

Images