Fault inspection and operation and maintenance method, device and medium for tunnel fire detection equipment

Abstract

The present invention relates to the technical field of tunnel equipment operation and maintenance management, in particular to a method, device and medium for fault inspection and operation and maintenance of tunnel fire detection equipment. The historical tunnel fire environment data text associated with the fault inspection clustering label is extracted, and a tunnel fire environment simulation field describing each tunnel fire detection equipment to be inspected is created through the historical tunnel fire environment data text, and a fault inspection excitation signal is allocated to obtain a fault inspection strategy; based on the fault inspection strategy, a hysteresis characteristic state equation of the tunnel fire detection equipment to be inspected is constructed, and a hysteresis compensation parameter for compensating the open-loop phase margin in the hysteresis characteristic state equation is designed according to the ideal tunnel fire detection requirements, and the fault state is analyzed through the hysteresis compensation parameter to obtain a fault inspection result; the present invention can plan a reasonable daily fault inspection strategy for tunnel fire detection equipment and perform daily inspection and operation and maintenance, improve the efficiency and quality of fault inspection, and optimize its detection performance and service life.

Description

Technical Field

[0001] The present invention relates to the technical field of tunnel equipment operation and maintenance management, and in particular to a fault inspection and operation and maintenance method, device and medium for tunnel fire detection equipment. Background Art

[0002] Tunnel fire detection equipment is a key device for early detection of fires in tunnels and timely alarm. Due to the particularity of the tunnel environment, such as closed and poor ventilation conditions, once a fire occurs, it is very easy to cause casualties and property losses. Therefore, fire detection equipment plays a vital role in tunnels. Tunnel fire detection equipment belongs to a type of mechanical and electronic equipment. As the operating time increases, the internal components will age and damage, resulting in failures, making it impossible to efficiently detect tunnel fires. Therefore, regular fault inspections are required, but traditional fault inspection methods usually require the help of manpower and professional inspection equipment to complete accurate fault determination. This method requires multiple people to work together, the steps are cumbersome, time-consuming and labor-intensive, which reduces the efficiency of fault inspections and increases the output of manpower and material costs; at the same time, there are large errors in manual inspections, which makes the fault inspection results inaccurate, which is not conducive to subsequent high-quality daily operation and maintenance, and has low reliability. Therefore, it is necessary to develop an efficient and accurate fault inspection method that can save manual output to solve the above problems. Summary of the invention

[0003] The present invention overcomes the deficiencies of the prior art and provides a method, device and medium for fault inspection and operation and maintenance of tunnel fire detection equipment.

[0004] To achieve the above object, the technical solution adopted by the present invention is:

[0005] A first aspect of the present invention provides a fault inspection and operation and maintenance method for tunnel fire detection equipment, comprising the following steps:

[0006] S102: Obtain and analyze the influence of the migration trend vectors of different tunnel environments in the target tunnel space within a preset time period on the state trend of the tunnel fire benchmark time series change characteristics to determine multiple tunnel fire characteristics, and cluster inspection targets based on the multiple tunnel fire characteristics to obtain the fault inspection clustering label of each tunnel fire detection device to be inspected for the tunnel fire;

[0007] S104: extracting historical tunnel fire environment data text associated with the fault inspection clustering label in the multi-dimensional virtual domain, creating a tunnel fire environment simulation field describing each tunnel fire detection device to be inspected through the historical tunnel fire environment data text and allocating a fault inspection excitation signal to obtain a fault inspection strategy;

[0008] S106: constructing a hysteresis characteristic state equation of the tunnel fire detection equipment to be inspected based on the fault inspection strategy, designing a hysteresis compensation parameter for compensating the open-loop phase margin in the hysteresis characteristic state equation according to the ideal tunnel fire detection requirements, analyzing the fault state through the hysteresis compensation parameter, and obtaining a fault inspection result;

[0009] S108: based on the fault inspection result, the control configuration parameters and control execution nodes of the faulty tunnel fire detection device are obtained, the control execution node execution hysteresis compensation parameters are fuzzy evaluated by the control configuration parameters for the fault membership function of the faulty tunnel fire detection device, and the fault membership function is analyzed to obtain the fault tracing analysis result;

[0010] S110: Determine the first operation and maintenance difficulty coefficient of the active fault and the second operation and maintenance difficulty coefficient of the passive fault of the faulty tunnel fire detection equipment according to the fault tracing analysis result, match the operation and maintenance decision of corresponding difficulty according to the first operation and maintenance difficulty coefficient and the second operation and maintenance difficulty coefficient, and obtain the final operation and maintenance decision.

[0011] More specifically, the step S102 includes the following steps:

[0012] Obtain model specification parameter information of all the tunnel fire detection devices to be inspected in the target tunnel space, and obtain the established functions of each tunnel fire detection device to be inspected according to the model specification parameter information;

[0013] Obtain migration trend vectors of different tunnel environments of the target tunnel space within a preset time period, and retrieve from a big data network several tunnel fire benchmark time series variation characteristics when a fire occurs in the target tunnel space under different tunnel environment conditions based on the migration trend vectors of the different tunnel environments;

[0014] Preset multiple local regression points and bandwidth parameters according to the migration trend vectors of different tunnel environments, calculate the kernel function of the benchmark time series change characteristics of several tunnel fires and each local regression point with the bandwidth parameter as the local range, obtain multiple kernel function weights, and construct a local polynomial regression trend model according to the multiple kernel function weights;

[0015] By solving the weighted least squares equation group included in the local polynomial regression trend model, the local polynomial regression coefficient is obtained, and the influence degree of different tunnel environments on the state trend of the tunnel fire benchmark time series change characteristics is determined according to the local polynomial regression coefficient;

[0016] updating each tunnel fire benchmark time series variation feature based on the state trend influence degree to obtain multiple tunnel fire characteristics corresponding to each tunnel fire benchmark time series variation feature after update, and setting access points and access point areas based on the multiple tunnel fire characteristics;

[0017] The OPTICS algorithm and the preset minimum number of access points are introduced, and the domain radius of the current access point is calculated according to the established function to obtain the core distance of the current access point. If the number of access points within the core distance is greater than the minimum number of access points, a distance check is performed on all neighboring access points of the current access point in the access point domain to output the reachable distance of each neighboring access point.

[0018] Update the neighborhood access points based on the reachable distance and the core distance until the preset threshold is reached, add the current access point after the domain access point is updated to the cluster order, and continue to explore the remaining access points in the domain of the access point with distance check to generate a cluster order graph of all current access points;

[0019] The cluster sequence diagrams of all current access points are analyzed to obtain clustering results, and a fault inspection clustering label for the tunnel fire of each tunnel fire detection device to be inspected is determined according to the clustering results.

[0020] More specifically, the step S104 includes the following steps:

[0021] Obtaining an original engineering design drawing of the target tunnel space, constructing a multi-dimensional virtual domain of the target tunnel space according to the original engineering design drawing, and obtaining a layout and installation drawing of different tunnel fire detection equipment to be inspected in the target tunnel space;

[0022] The multi-dimensional virtual domain is divided into N sub-virtual domain grids based on the layout installation diagram, and the inspection log of the target tunnel space is obtained. The historical tunnel fire environment data text associated with the fault inspection clustering label of each tunnel fire detection device to be inspected for the tunnel fire is extracted from each sub-virtual domain grid at different time steps through the inspection log;

[0023] The Cellular Automata simulation method is introduced to subdivide each sub-virtual domain grid into a number of state simulation units, and the fire state of the state simulation unit at different time sequence steps is determined according to the historical tunnel fire environment data text, and the fire time sequence state of each state simulation unit is obtained;

[0024] Calculate the state transition probability matrix of the fire sequence state of the current state simulation unit based on the fire sequence state so that the next state simulation unit will synchronously appear the current fire sequence state, and design and define the fire state transition probability rules of the adjacent state simulation units according to the state transition probability matrix;

[0025] Taking different time sequence steps as the updating premise, the fire time sequence state of each state simulation unit is iteratively updated based on the fire state transition probability rule to generate a new fire time sequence state of each state simulation unit;

[0026] The Cellular Automata simulation method is used to simulate and express the new fire time sequence state of all state simulation units, and the tunnel fire environment simulation field where the tunnel fire detection equipment to be inspected is located in the sub-virtual domain grid is obtained;

[0027] Based on the tunnel fire environment simulation field where the tunnel fire detection equipment to be inspected is located in the sub-virtual domain grid, a fault inspection excitation signal is constructed, and the fault inspection excitation signal is uploaded to the fault inspection terminal corresponding to the tunnel fire detection equipment to be inspected to obtain a fault inspection strategy.

[0028] More specifically, the step S106 includes the following steps:

[0029] Obtain the actual detection delay rate and actual hysteresis amplitude of the fault inspection excitation signal corresponding to the detection of the current tunnel fire detection device to be inspected, and obtain the detection process and detection response logic of the current tunnel fire detection device to be inspected to execute the fault inspection strategy;

[0030] A discrete difference equation is constructed according to the detection process and the detection response logic, and the actual detection delay rate and the actual hysteresis amplitude are introduced into the discrete difference equation to update the hysteresis state, so as to obtain the hysteresis characteristic state equation of the tunnel fire detection equipment to be inspected;

[0031] Solving the hysteresis characteristic state equation of the tunnel fire detection device to be inspected to obtain an open-loop transfer function of the hysteresis characteristic, and determining the open-loop phase margin of the tunnel fire detection device to be inspected according to the open-loop transfer function of the hysteresis characteristic;

[0032] Obtaining an ideal tunnel fire detection requirement for the tunnel fire detection device to be inspected, extracting an expected detection index threshold when the tunnel fire detection device to be inspected detects a tunnel fire through the ideal tunnel fire detection requirement, and constructing a hysteresis compensation architecture based on the open-loop transfer function;

[0033] Presetting a phase margin threshold according to an expected detection indicator threshold, updating a hysteresis compensation zero point and a hysteresis compensation pole of a hysteresis compensation architecture based on the open-loop phase margin until the open-loop phase margin is greater than the phase margin threshold, and obtaining a hysteresis compensation parameter;

[0034] A detection fault sparse coding set is constructed based on the expected detection index threshold, and the hysteresis compensation parameter is converted into multiple sub-sparse codes. If multiple sub-sparse codes do not exist in the detection fault sparse coding set, the tunnel fire detection device to be inspected is calibrated as a normal tunnel fire detection device;

[0035] If at least one or more sub-sparse codes appear in the detection fault sparse code set, the tunnel fire detection device to be inspected is calibrated as a faulty tunnel fire detection device, and a fault inspection result is obtained.

[0036] More specifically, the step S108 includes the following steps:

[0037] Extract all faulty tunnel fire detection devices based on the fault inspection results, obtain the control network associated with each faulty tunnel fire detection device, apply the hysteresis compensation parameters to the corresponding faulty tunnel fire detection device, and obtain the control configuration parameters of the faulty tunnel fire detection device;

[0038] The bus topology method is introduced to control the topology of the control network associated with each faulty tunnel fire detection device to obtain a control network topology map. The control execution node in each faulty tunnel fire detection device is obtained through the control network topology map, and the active and passive attributes of the control execution node are defined.

[0039] Analyze the feasibility of the control configuration parameters and the execution hysteresis compensation parameters of the active control execution node and the passive control execution node to list several feasible decision schemes, calculate the evaluation fuzzy numbers of each feasible decision scheme on the active attribute and the passive attribute one by one, and construct a decision matrix according to the evaluation fuzzy numbers;

[0040] A hierarchical analysis method is introduced to calculate the attribute weights of the active attributes and the passive attributes to obtain attribute fuzzy weight values, and a fuzzy weighted average value of each feasible decision scheme in the decision matrix is ​​calculated based on the evaluation fuzzy number and the attribute fuzzy weight value, and a fault membership function of the control execution node for generating current lag compensation for the faulty tunnel fire detection equipment is determined according to the fuzzy weighted average value;

[0041] Preset the fault membership function interval of active attributes and the fault membership function interval of passive attributes, and if the fault membership function is within the fault membership function interval of active attributes, mark the control execution node as an active attribute fault node;

[0042] If the fault membership function is within the fault membership function interval of the passive attribute, the control execution node is marked as a passive attribute fault node, and the fault tracing analysis result is obtained.

[0043] More specifically, the step S110 includes the following steps:

[0044] If the fault tracing analysis result of the control execution node shows that it is an active attribute fault node, multiple detection activity trajectories of the control fault tunnel fire detection device when the active attribute fault node fails are obtained, and an actual detection activity trajectory model of the fault tunnel fire detection device is constructed according to the multiple detection activity trajectories;

[0045] Obtaining a preset detection activity control strategy for controlling a faulty tunnel fire detection device when an active attribute fault node fails, and creating an accurate detection activity trajectory model according to the preset detection activity control strategy;

[0046] A hash attention mechanism is introduced to calculate a first hash attention score of the actual detection activity trajectory model compared to the accurate detection activity trajectory model, and an operation and maintenance difficulty coefficient of an active attribute failure of the faulty tunnel fire detection equipment is determined according to the first hash attention score, which is defined as a first operation and maintenance difficulty coefficient;

[0047] If the fault tracing analysis result of the control execution node shows that it is a passive attribute fault node, the disturbance factor that causes the passive attribute fault of the control execution node under the current tunnel environment state is obtained based on the big data network, and the historical migration disturbance area of ​​the disturbance factor is extracted through the operation and maintenance log;

[0048] Obtaining a control criterion for a passive attribute fault node, obtaining a preset detection execution range of the passive attribute fault node according to the control criterion, and calculating an interference area value between a historical migration disturbance region of a disturbance factor and the preset detection execution range;

[0049] A threshold of allowable interference area is preset, and a second hash attention score of the interference area value compared with the threshold of allowable interference area is calculated through a hash attention mechanism. The operation and maintenance difficulty coefficient of the passive attribute failure of the faulty tunnel fire detection equipment is determined according to the second hash attention score, which is defined as the second operation and maintenance difficulty coefficient;

[0050] Acquire several operation and maintenance decisions associated with the faulty tunnel fire detection equipment and the operation and maintenance difficulty level solved by each operation and maintenance decision, construct a hash index table of several operation and maintenance decisions based on the operation and maintenance difficulty level, import the first operation and maintenance difficulty coefficient and the second operation and maintenance difficulty coefficient into the hash index table for index query, obtain the final operation and maintenance decision and upload it to the inspection and operation and maintenance robot terminal.

[0051] A second aspect of the present invention provides a fault inspection and operation and maintenance device for tunnel fire detection equipment, the fault inspection and operation and maintenance device comprising a memory and a processor, the memory storing a fault inspection and operation and maintenance method program for tunnel fire detection equipment, and when the fault inspection and operation and maintenance method program for tunnel fire detection equipment is executed by the processor, any one of the steps of the fault inspection and operation and maintenance method for tunnel fire detection equipment is implemented.

[0052] A third aspect of the present invention provides a computer-readable storage medium, which stores a computer detection program. When the computer detection program is executed by at least one processor, it implements any one of the fault inspection and operation and maintenance methods for tunnel fire detection equipment.

[0053] The present invention solves the technical defects existing in the background technology, and the beneficial technical effects of the present invention are:

[0054] The historical tunnel fire environment data text associated with the fault inspection clustering label in the multi-dimensional virtual domain is extracted, and a tunnel fire environment simulation field describing each tunnel fire detection device to be inspected is created through the historical tunnel fire environment data text and a fault inspection excitation signal is allocated to obtain a fault inspection strategy; the hysteresis characteristic state equation of the tunnel fire detection device to be inspected is constructed based on the fault inspection strategy, and a hysteresis compensation parameter for compensating the open-loop phase margin in the hysteresis characteristic state equation is designed according to the ideal tunnel fire detection requirement, and the fault state is analyzed through the hysteresis compensation parameter to obtain a fault inspection result; the control configuration parameters and control execution nodes of the faulty tunnel fire detection device are obtained based on the fault inspection result, and the control execution node execution hysteresis compensation parameter is fuzzy evaluated by the control configuration parameter for the fault membership function of the faulty tunnel fire detection device, and the fault membership function is analyzed to obtain a fault tracing analysis result; the first operation and maintenance difficulty coefficient of the active attribute fault of the faulty tunnel fire detection device and the second operation and maintenance difficulty coefficient of the passive attribute fault are determined according to the fault tracing analysis result, and the operation and maintenance decision of the corresponding difficulty is matched according to the first operation and maintenance difficulty coefficient and the second operation and maintenance difficulty coefficient, so as to obtain a final operation and maintenance decision. The present invention can plan a reasonable daily fault inspection strategy for tunnel fire detection equipment, improve the efficiency and quality of fault inspection, and at the same time perform daily inspection and maintenance on tunnel fire detection equipment to optimize the detection performance and service life of tunnel fire detection equipment. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, drawings of other embodiments can be obtained based on these drawings without paying creative work.

[0056] Figure 1 A first method flow chart of a fault inspection and operation and maintenance method for tunnel fire detection equipment is shown;

[0057] Figure 2 A second method flow chart of the fault inspection and operation and maintenance method of tunnel fire detection equipment is shown;

[0058] Figure 3 The overall structure diagram of the fault inspection and operation and maintenance device of the tunnel fire detection equipment is shown. DETAILED DESCRIPTION

[0059] In order to more clearly understand the above-mentioned purpose, features and advantages of the present invention, the present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be noted that the embodiments of the present application and the features in the embodiments can be combined with each other without conflict.

[0060] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Therefore, the protection scope of the present invention is not limited to the specific embodiments disclosed below.

[0061] The first aspect of the present invention provides a fault inspection and operation and maintenance method for a tunnel fire detection device, such as Figure 1 As shown, the following steps are included:

[0062] S102: Obtain and analyze the influence of the migration trend vectors of different tunnel environments in the target tunnel space within a preset time period on the state trend of the tunnel fire benchmark time series change characteristics to determine multiple tunnel fire characteristics, and cluster inspection targets based on the multiple tunnel fire characteristics to obtain the fault inspection clustering label of each tunnel fire detection device to be inspected for the tunnel fire;

[0063] S104: extracting historical tunnel fire environment data text associated with the fault inspection clustering label in the multi-dimensional virtual domain, creating a tunnel fire environment simulation field describing each tunnel fire detection device to be inspected through the historical tunnel fire environment data text and allocating a fault inspection excitation signal to obtain a fault inspection strategy;

[0064] S106: constructing a hysteresis characteristic state equation of the tunnel fire detection equipment to be inspected based on the fault inspection strategy, designing a hysteresis compensation parameter for compensating the open-loop phase margin in the hysteresis characteristic state equation according to the ideal tunnel fire detection requirements, analyzing the fault state through the hysteresis compensation parameter, and obtaining a fault inspection result;

[0065] S108: based on the fault inspection result, the control configuration parameters and control execution nodes of the faulty tunnel fire detection device are obtained, the control execution node execution hysteresis compensation parameters are fuzzy evaluated by the control configuration parameters for the fault membership function of the faulty tunnel fire detection device, and the fault membership function is analyzed to obtain the fault tracing analysis result;

[0066] S110: Determine the first operation and maintenance difficulty coefficient of the active fault and the second operation and maintenance difficulty coefficient of the passive fault of the faulty tunnel fire detection equipment according to the fault tracing analysis result, match the operation and maintenance decision of corresponding difficulty according to the first operation and maintenance difficulty coefficient and the second operation and maintenance difficulty coefficient, and obtain the final operation and maintenance decision.

[0067] More specifically, the step S102 includes the following steps:

[0068] Obtain model specification parameter information of all the tunnel fire detection devices to be inspected in the target tunnel space, and obtain the established functions of each tunnel fire detection device to be inspected according to the model specification parameter information;

[0069] Obtain migration trend vectors of different tunnel environments of the target tunnel space within a preset time period, and retrieve from a big data network several tunnel fire benchmark time series variation characteristics when a fire occurs in the target tunnel space under different tunnel environment conditions based on the migration trend vectors of the different tunnel environments;

[0070] Preset multiple local regression points and bandwidth parameters according to the migration trend vectors of different tunnel environments, calculate the kernel function of the benchmark time series change characteristics of several tunnel fires and each local regression point with the bandwidth parameter as the local range, obtain multiple kernel function weights, and construct a local polynomial regression trend model according to the multiple kernel function weights;

[0071] By solving the weighted least squares equation group included in the local polynomial regression trend model, the local polynomial regression coefficient is obtained, and the influence degree of different tunnel environments on the state trend of the tunnel fire benchmark time series change characteristics is determined according to the local polynomial regression coefficient;

[0072] updating each tunnel fire benchmark time series variation feature based on the state trend influence degree to obtain multiple tunnel fire characteristics corresponding to each tunnel fire benchmark time series variation feature after update, and setting access points and access point areas based on the multiple tunnel fire characteristics;

[0073] The OPTICS algorithm and the preset minimum number of access points are introduced, and the domain radius of the current access point is calculated according to the established function to obtain the core distance of the current access point. If the number of access points within the core distance is greater than the minimum number of access points, a distance check is performed on all neighboring access points of the current access point in the access point domain to output the reachable distance of each neighboring access point.

[0074] Update the neighborhood access points based on the reachable distance and the core distance until the preset threshold is reached, add the current access point after the domain access point is updated to the cluster order, and continue to explore the remaining access points in the domain of the access point with distance check to generate a cluster order graph of all current access points;

[0075] The cluster sequence diagrams of all current access points are analyzed to obtain clustering results, and a fault inspection clustering label for the tunnel fire of each tunnel fire detection device to be inspected is determined according to the clustering results.

[0076] It should be noted that the station tunnel is claustrophobic, and there are many electrical equipment or cables arranged inside the tunnel. Due to the special setting of the tunnel ventilation system, the internal environment of the tunnel is relatively complex and changeable, and the fire characteristics of the tunnel are also affected and changed due to changes in the environment. For example, the wind speed in the tunnel environment may be high, especially when there is a mechanical ventilation system or hot air flow caused by fire, which causes the flame to spread rapidly, resulting in a large amount of thick smoke and reduced visibility. In view of the above conditions, tunnel fire detection equipment with certain functions is required to have certain temperature detection performance, flame radiation perception performance, and smoke perception performance; therefore, in order to efficiently and accurately inspect the operation failure problems of tunnel fire detection equipment, it is necessary to adapt the corresponding fault inspection targets and basic requirements for tunnel fire detection equipment with different functions. From the above examples, we can see that the tunnel environment will affect the state trend characteristics of the fire in the tunnel. Therefore, this method first analyzes the nonlinear structural influence of different tunnel environments on the benchmark time series change characteristics of the tunnel fire by constructing a local polynomial regression trend model to capture the state trend influence between the two. The local polynomial regression is more robust to outliers, and the selection of bandwidth parameters can be flexibly adjusted according to the actual situation of the data to balance the accuracy of smoothness and local fitting, improve the accuracy of the analysis of the state trend influence of the tunnel environment on the fire characteristics, and have high reliability. Then, according to the state trend influence, the fire characteristics that the tunnel environment will cause the fire to occur are determined, and the corresponding fault inspection targets are adapted for the tunnel fire detection equipment according to these fire characteristics. This method will cluster the determined multiple fire characteristics one by one and assign them to the equipment with the function of detecting the corresponding fire characteristics. Finally, the fault inspection clustering label of each tunnel fire detection equipment to be inspected for the tunnel fire can be determined according to the clustering results, so as to achieve the accurate allocation effect of fault inspection targets and requirements of different equipment. Compared with traditional fault inspection methods, this method ensures the adaptability and accuracy of fault inspection for different equipment functions, avoids false detection and missed detection of equipment failures, and greatly improves the efficiency and rationality of fault inspection for all tunnel fire detection equipment.

[0077] More specifically, the step S104 includes the following steps:

[0078] Obtaining an original engineering design drawing of the target tunnel space, constructing a multi-dimensional virtual domain of the target tunnel space according to the original engineering design drawing, and obtaining a layout and installation drawing of different tunnel fire detection equipment to be inspected in the target tunnel space;

[0079] The multi-dimensional virtual domain is divided into N sub-virtual domain grids based on the layout installation diagram, and the inspection log of the target tunnel space is obtained. The historical tunnel fire environment data text associated with the fault inspection clustering label of each tunnel fire detection device to be inspected for the tunnel fire is extracted from each sub-virtual domain grid at different time steps through the inspection log;

[0080] The Cellular Automata simulation method is introduced to subdivide each sub-virtual domain grid into a number of state simulation units, and the fire state of the state simulation unit at different time sequence steps is determined according to the historical tunnel fire environment data text, and the fire time sequence state of each state simulation unit is obtained;

[0081] Calculate the state transition probability matrix of the fire sequence state of the current state simulation unit based on the fire sequence state so that the next state simulation unit will synchronously appear the current fire sequence state, and design and define the fire state transition probability rules of the adjacent state simulation units according to the state transition probability matrix;

[0082] Taking different time sequence steps as the updating premise, the fire time sequence state of each state simulation unit is iteratively updated based on the fire state transition probability rule to generate a new fire time sequence state of each state simulation unit;

[0083] The Cellular Automata simulation method is used to simulate and express the new fire time sequence state of all state simulation units, and the tunnel fire environment simulation field where the tunnel fire detection equipment to be inspected is located in the sub-virtual domain grid is obtained;

[0084] Based on the tunnel fire environment simulation field where the tunnel fire detection equipment to be inspected is located in the sub-virtual domain grid, a fault inspection excitation signal is constructed, and the fault inspection excitation signal is uploaded to the fault inspection terminal corresponding to the tunnel fire detection equipment to be inspected to obtain a fault inspection strategy.

[0085] It should be noted that after determining the specific fault inspection targets and requirements for each fire detection device in the tunnel, it is necessary to implement corresponding fault inspection tasks for each tunnel fire detection device. However, the traditional fault inspection method usually uses a uniform simulation method to globally verify the sensitive perception performance of the device to fire, which greatly reduces the response capabilities of different tunnel fire detection devices to fires in different environments. Therefore, there are large errors in the testing and positioning of its own faults, thereby reducing the quality and accuracy of equipment fault inspections, which is not conducive to the high-precision operation and management of subsequent equipment. Therefore, this method can build an exclusive tunnel fire environment simulation field for each device according to the detection targets and requirements of different fire characteristics of tunnel fire detection devices with different functions to provide accurate and targeted fire detection perception tests, thereby ensuring the stability and accuracy of the daily fault inspection of each fire detection device in the tunnel, effectively reducing the error degree of fault testing, and improving the credibility and reliability of fault inspection results. Among them, this method extracts the corresponding historical tunnel fire environment data text associated with the fault inspection clustering label according to the layout and installation of each device in the target tunnel as the simulation basis, which can improve the detection performance of the equipment for the actual fire situation in the target tunnel to perform high-quality daily inspections, and achieve accurate simulation and analysis effects based on the actual fire situation. Among them, the Chinese name of the Cellular Automata simulation method is the cellular automaton simulation method, which is a computational method that simulates the behavior of complex systems through feature rules. The Cellular Automata simulation method can be used to simulate a variety of different systems and phenomena, and has certain flexibility and wide applicability for the system simulation of different historical tunnel fire environment data texts, which improves the authenticity and reliability of the scene simulation of different tunnel fire environment simulation fields.

[0086] More specifically, the step S106 includes the following steps:

[0087] Obtain the actual detection delay rate and actual hysteresis amplitude of the fault inspection excitation signal corresponding to the detection of the current tunnel fire detection device to be inspected, and obtain the detection process and detection response logic of the current tunnel fire detection device to be inspected to execute the fault inspection strategy;

[0088] A discrete difference equation is constructed according to the detection process and the detection response logic, and the actual detection delay rate and the actual hysteresis amplitude are introduced into the discrete difference equation to update the hysteresis state, so as to obtain the hysteresis characteristic state equation of the tunnel fire detection equipment to be inspected;

[0089] Solving the hysteresis characteristic state equation of the tunnel fire detection device to be inspected to obtain an open-loop transfer function of the hysteresis characteristic, and determining the open-loop phase margin of the tunnel fire detection device to be inspected according to the open-loop transfer function of the hysteresis characteristic;

[0090] Obtaining an ideal tunnel fire detection requirement for the tunnel fire detection device to be inspected, extracting an expected detection index threshold when the tunnel fire detection device to be inspected detects a tunnel fire through the ideal tunnel fire detection requirement, and constructing a hysteresis compensation architecture based on the open-loop transfer function;

[0091] Presetting a phase margin threshold according to an expected detection indicator threshold, updating a hysteresis compensation zero point and a hysteresis compensation pole of a hysteresis compensation architecture based on the open-loop phase margin until the open-loop phase margin is greater than the phase margin threshold, and obtaining a hysteresis compensation parameter;

[0092] A detection fault sparse coding set is constructed based on the expected detection index threshold, and the hysteresis compensation parameter is converted into multiple sub-sparse codes. If multiple sub-sparse codes do not exist in the detection fault sparse coding set, the tunnel fire detection device to be inspected is calibrated as a normal tunnel fire detection device;

[0093] If at least one or more sub-sparse codes appear in the detection fault sparse code set, the tunnel fire detection device to be inspected is calibrated as a faulty tunnel fire detection device, and a fault inspection result is obtained.

[0094] It should be noted that the fault inspection excitation signal constructed by injecting the tunnel fire environment simulation field is used to test the fire detection perception performance of each corresponding tunnel fire detection equipment. However, the traditional fault inspection method still needs to rely on manual analysis and elimination to determine the fault results. This requires a lot of manpower and professional equipment to coordinate, which increases the workload and cost output of daily fault inspection and maintenance of tunnel fire detection equipment, and greatly reduces the efficiency of fault inspection. Therefore, this method first describes the actual hysteresis state transformation of the fault inspection excitation signal detected by the current tunnel fire detection equipment to be inspected corresponding to the hysteresis characteristic state equation. The hysteresis characteristic state equation expresses the hysteresis of the fire detection behavior and operation control state of the current tunnel fire detection equipment to be inspected, and reflects the performance of the tunnel fire detection equipment to be inspected; the open-loop transfer function of the tunnel fire detection equipment to be inspected is solved according to the hysteresis characteristic state equation. The open-loop transfer function describes the relationship between how the input signal of the tunnel fire detection equipment to be inspected is transmitted and processed when it passes through the system without feedback, which can further reflect the degree of equipment hysteresis and the gain required for compensation, and can provide a reliable data analysis basis for hysteresis compensation and fault judgment of each tunnel fire detection equipment to be inspected. The open-loop phase margin is an important parameter in the control system. It can evaluate the stability of the closed-loop system and can be determined by the open-loop transfer function. This method is based on the open-loop phase margin determined by the threshold compensation of each expected detection index, so as to design a series of lag compensation parameters for each tunnel fire detection device to be inspected. Finally, by judging whether the lag compensation parameters are reasonable, it is further determined whether the device has a fault. If there are no multiple sub-sparse codes in the detection fault sparse coding set, it means that the required lag compensation does not have detection fault compensation, so the device does not have a fault; otherwise, it means that there is detection fault compensation, so the device has a fault. This method can accurately inspect and compensate for the fault behavior of the tunnel fire detection equipment, replacing the tedious steps of traditional manual inspections one by one, saving time and effort, and greatly improving the accuracy and reliability of the equipment fault determination results.

[0095] More specifically, the step S108 is as follows: Figure 2 As shown, the specific steps include:

[0096] S202: extracting all faulty tunnel fire detection devices based on the fault inspection results, obtaining a control network associated with each faulty tunnel fire detection device, and applying a hysteresis compensation parameter to the corresponding faulty tunnel fire detection device, thereby obtaining a control configuration parameter of the faulty tunnel fire detection device;

[0097] S204: Introducing a bus topology method to perform control topology on the control network associated with each faulty tunnel fire detection device, obtaining a control network topology map, obtaining a control execution node in each faulty tunnel fire detection device through the control network topology map, and defining active attributes and passive attributes of the control execution node;

[0098] S206: Analyze the feasibility of the control configuration parameters and the execution hysteresis compensation parameters of the active control execution node and the passive control execution node to list a number of feasible decision schemes, calculate the evaluation fuzzy numbers of each feasible decision scheme on the active attribute and the passive attribute one by one, and construct a decision matrix according to the evaluation fuzzy numbers;

[0099] S208: Introducing the hierarchical analysis method to calculate the attribute weights of the active attributes and the passive attributes to obtain attribute fuzzy weight values, calculating the fuzzy weighted average value of each feasible decision scheme in the decision matrix based on the evaluation fuzzy number and the attribute fuzzy weight value, and determining the fault membership function of the control execution node for generating current lag compensation for the faulty tunnel fire detection equipment according to the fuzzy weighted average value;

[0100] S210: Preset an active attribute fault membership function interval and a passive attribute fault membership function interval, and if the fault membership function is within the active attribute fault membership function interval, mark the control execution node as an active attribute fault node;

[0101] S212: If the fault membership function is within the fault membership function interval of the passive attribute, the control execution node is marked as a passive attribute fault node, and a fault tracing analysis result is obtained.

[0102] It should be noted that due to the wide variety of tunnel fire detection equipment, each device may need to use different control execution nodes to implement fire detection methods and behaviors. For example, some nodes control the active fire threat detection behavior of tunnel fire detection equipment through active attributes, while some nodes control the passive fire threat detection behavior of tunnel fire detection equipment through passive attributes. Improper control of these two attributes will cause tunnel fire detection equipment to malfunction. Therefore, after determining the fault condition of the tunnel fire detection equipment, it is crucial to trace the control fault of the equipment; this method derives the corresponding control execution node according to the control network topology of each faulty tunnel fire detection equipment. These control execution nodes have active attributes and passive attributes, depending on the function of the equipment itself; then, the fuzziness of the control configuration parameters of the control execution node in each faulty tunnel fire detection equipment that requires the execution of hysteresis compensation parameters is analyzed by fuzzy decision-making. Since the hysteresis compensation parameters are essentially the control performance of the compensation equipment, when hysteresis compensation needs to be executed, it means that a certain control execution node has a fault, which leads to an error in the operation control configuration parameters, and ultimately causes the equipment to be unable to detect and respond to tunnel fire behavior. Therefore, this method can be used to perform fuzzy decision-making fault tracing analysis on the control behavior and state rationality of one or more control execution nodes that cause the failure of the tunnel fire detection equipment, thereby achieving efficient fault troubleshooting effects of equipment-related control nodes, providing a more refined fault inspection planning basis for the subsequent daily operation and maintenance of the equipment, avoiding equipment failures caused by abnormal control execution nodes, and improving the control operation stability and reliability of the tunnel fire detection equipment.

[0103] More specifically, the step S110 includes the following steps:

[0104] If the fault tracing analysis result of the control execution node shows that it is an active attribute fault node, multiple detection activity trajectories of the control fault tunnel fire detection device when the active attribute fault node fails are obtained, and an actual detection activity trajectory model of the fault tunnel fire detection device is constructed according to the multiple detection activity trajectories;

[0105] Obtaining a preset detection activity control strategy for controlling a faulty tunnel fire detection device when an active attribute fault node fails, and creating an accurate detection activity trajectory model according to the preset detection activity control strategy;

[0106] A hash attention mechanism is introduced to calculate a first hash attention score of the actual detection activity trajectory model compared to the accurate detection activity trajectory model, and an operation and maintenance difficulty coefficient of an active attribute failure of the faulty tunnel fire detection equipment is determined according to the first hash attention score, which is defined as a first operation and maintenance difficulty coefficient;

[0107] If the fault tracing analysis result of the control execution node shows that it is a passive attribute fault node, the disturbance factor that causes the passive attribute fault of the control execution node under the current tunnel environment state is obtained based on the big data network, and the historical migration disturbance area of ​​the disturbance factor is extracted through the operation and maintenance log;

[0108] Obtaining a control criterion for a passive attribute fault node, obtaining a preset detection execution range of the passive attribute fault node according to the control criterion, and calculating an interference area value between a historical migration disturbance region of a disturbance factor and the preset detection execution range;

[0109] A threshold of allowable interference area is preset, and a second hash attention score of the interference area value compared with the threshold of allowable interference area is calculated through a hash attention mechanism. The operation and maintenance difficulty coefficient of the passive attribute failure of the faulty tunnel fire detection equipment is determined according to the second hash attention score, which is defined as the second operation and maintenance difficulty coefficient;

[0110] Acquire several operation and maintenance decisions associated with the faulty tunnel fire detection equipment and the operation and maintenance difficulty level solved by each operation and maintenance decision, construct a hash index table of several operation and maintenance decisions based on the operation and maintenance difficulty level, import the first operation and maintenance difficulty coefficient and the second operation and maintenance difficulty coefficient into the hash index table for index query, obtain the final operation and maintenance decision and upload it to the inspection and operation and maintenance robot terminal.

[0111] It should be noted that when the fault phenomenon of the tunnel fire detection equipment is generated by the control execution node running the detection instruction of the active attribute, it means that the tunnel fire detection equipment fails when it detects fire signs in its autonomous activity. Therefore, it is necessary to conduct a standardized analysis of the actual detection activity trajectory of the tunnel fire detection equipment, so that the degree of fault damage of the tunnel fire detection equipment can be judged according to the degree of deviation of the actual detection activity trajectory compared with the accurate detection activity trajectory, so as to further judge the difficulty index of operating and maintaining the tunnel fire detection equipment with active attribute faults. When the fault phenomenon of the tunnel fire detection equipment is generated by the control execution node running the detection instruction of the passive attribute, it means that the tunnel fire detection equipment fails when it detects fire signs in passive control, and the main reason for the passive attribute fault of the control execution node is the disturbance factor that interferes with its normal control execution, wherein the disturbance factor includes electromagnetic interference, ventilation flow, natural airflow, noise interference and vibration interference. Therefore, it is necessary to analyze the interference degree between the preset detection execution range and the migration disturbance area of ​​the disturbance factor when the control execution node passively controls the tunnel fire detection equipment to detect fire signs, so as to judge the difficulty index of operating and maintaining the tunnel fire detection equipment with passive attribute faults. This method can realize the targeted matching effect of the operation and maintenance difficulty index of the tunnel fire detection equipment under different fault conditions, improve the operation and maintenance quality and feasibility of the tunnel fire detection equipment with active attribute faults, and effectively reduce the frequency of tunnel fire detection equipment failures caused by active and passive control of the control execution nodes.

[0112] In addition, the fault inspection and operation and maintenance method of the tunnel fire detection equipment further includes the following steps:

[0113] Obtain the fault inspection results of all tunnel fire detection equipment in the target tunnel space, and at the same time obtain the inspection channel map of the target tunnel space;

[0114] According to the fault inspection results, the location of the faulty tunnel fire detection device is marked in the inspection channel map, and the operation and maintenance decision and the operation and maintenance difficulty level of each faulty tunnel fire detection device are obtained;

[0115] The planning constraint threshold of each faulty tunnel fire detection device is set based on the operation and maintenance decision and the operation and maintenance difficulty level, and each faulty tunnel fire detection device is regionally planned in the inspection channel map until the corresponding planning constraint threshold is reached, thereby generating an expected inspection and operation and maintenance area for each faulty tunnel fire detection device;

[0116] After the planning is completed, all inspection channels included in each expected inspection and operation area are stripped out through the inspection channel diagram, and several optional inspection channels for the inspection and operation of each fault tunnel fire detection equipment are obtained. The non-repeated optional inspection channels in each expected inspection and operation area are eliminated, and the remaining repeated optional inspection channels in each expected inspection and operation area are obtained, which are defined as the necessary inspection channels;

[0117] A heuristic path search algorithm is introduced to calculate and screen the multiple necessary inspection channels to obtain the optimal necessary inspection channel, and the optimal necessary inspection channel is uploaded to the navigation terminal of the inspection and operation robot.

[0118] It should be noted that the faulty tunnel fire detection equipment in the tunnel may be unevenly distributed, and there are also multiple dedicated channels in the tunnel for inspection drones to perform inspection tasks. If the traditional global inspection method is used for each faulty device, it may cause the inspection and operation robot to inspect a faulty device multiple times or walk the same inspection channel multiple times, thereby reducing the inspection efficiency, causing false detection and missed detection, and increasing the error rate of the inspection results. Therefore, this method can ensure that the inspection robot accurately inspects each faulty device and reduces the frequency of repeated walking in the inspection channel, and select the optimal inspection channel for the inspection robot to complete the predetermined fault inspection task in the shortest time, which greatly improves the inspection efficiency and inspection quality, ensures the accuracy of the fault inspection results, and reduces the occurrence of false detection and missed detection.

[0119] In addition, the fault inspection and operation and maintenance method of the tunnel fire detection equipment further includes the following steps:

[0120] Obtain the station tunnel in the current target tunnel space that is in operation, and obtain all movable areas in the station tunnel and corresponding functional information;

[0121] Based on the functional information, the tunnel fire cause of each movable area is obtained in the big data network, and the danger level index of the tunnel fire cause is obtained. If the danger level index is greater than the preset danger level index, the movable area is marked as an movable fire risk area;

[0122] Real-time monitoring of one or more active fire risk areas is performed through a monitoring system in a target tunnel space, and a monitoring interval step is preset, and multiple groups of real-time dynamic image features of the active fire risk areas in a continuous time sequence are extracted based on the monitoring interval step;

[0123] Constructing a model architecture of an input layer, an LSTM layer, and an output layer of the LSTM dynamic model, and obtaining a trained LSTM real-time dynamic model of a movable fire risk area by training and fitting multiple sets of real-time dynamic image features in the model architecture of the input layer, the LSTM layer, and the output layer;

[0124] Based on the big data network, several suspected fire activity behavior characteristics that lead to the fire inducement of each tunnel are obtained, and the actual fire activity behavior feature set in the active fire risk area is extracted through the LSTM real-time dynamic model;

[0125] If there is at least one or more standard fire activity behavior characteristics in the actual fire activity behavior characteristic set, the monitoring intensity of the fire activity behavior in the active fire risk area will be increased, and the tunnel fire detection equipment in the active fire risk area will be called as a temporary task detection equipment.

[0126] It should be noted that there are many areas in the station tunnel for public activities, and these areas have different functions. For example, the station entrances and exits are open for the public to enter and exit the station or get on and off the train. There are also many causes of tunnel fires at the station entrances and exits, such as static electricity, open flame sources, and electronic device batteries. Therefore, the station will implement monitoring measures for these tunnel fire causes. Therefore, this method can perform real-time monitoring and identification based on the suspected fire activity behavior characteristics that may cause tunnel fire causes in the monitoring measures. If there is at least one or more standard fire activity behavior characteristics in the actual fire activity behavior characteristics, it means that there are high-risk fire behaviors that cause tunnel fire causes in the movable fire risk area. Therefore, the tunnel fire detection equipment in the movable fire risk area should be controlled to start in advance, and respond to possible temporary fire accidents at any time, improve the temporary operation safety factor of the station tunnel, and reduce the probability and threat of fire safety accidents.

[0127] A second aspect of the present invention provides a fault inspection and maintenance device for tunnel fire detection equipment, such as Figure 3 As shown, the fault inspection and operation and maintenance device includes a memory 31 and a processor 32. The memory 31 stores a fault inspection and operation and maintenance method program for tunnel fire detection equipment. When the fault inspection and operation and maintenance method program for tunnel fire detection equipment is executed by the processor 32, any one of the steps of the fault inspection and operation and maintenance method for tunnel fire detection equipment is implemented.

[0128] A third aspect of the present invention provides a computer-readable storage medium, which stores a computer detection program. When the computer detection program is executed by at least one processor, it implements any one of the fault inspection and operation and maintenance methods for tunnel fire detection equipment.

[0129] The above are only specific embodiments of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art who is familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims

1. A fault inspection and operation and maintenance method for tunnel fire detection equipment, characterized in that: The following steps are involved: S102: Obtain and analyze the influence of the migration trend vectors of different tunnel environments in the target tunnel space within a preset time period on the state trend of the tunnel fire benchmark time series change characteristics to determine multiple tunnel fire characteristics, and cluster inspection targets based on the multiple tunnel fire characteristics to obtain the fault inspection clustering label of each tunnel fire detection device to be inspected for the tunnel fire; S104: extracting historical tunnel fire environment data text associated with the fault inspection clustering label in the multi-dimensional virtual domain, creating a tunnel fire environment simulation field describing each tunnel fire detection device to be inspected through the historical tunnel fire environment data text and allocating a fault inspection excitation signal to obtain a fault inspection strategy; S106: constructing a hysteresis characteristic state equation of the tunnel fire detection equipment to be inspected based on the fault inspection strategy, designing a hysteresis compensation parameter for compensating the open-loop phase margin in the hysteresis characteristic state equation according to the ideal tunnel fire detection requirements, analyzing the fault state through the hysteresis compensation parameter, and obtaining a fault inspection result; S108: based on the fault inspection result, the control configuration parameters and control execution nodes of the faulty tunnel fire detection device are obtained, the control execution node execution hysteresis compensation parameters are fuzzy evaluated by the control configuration parameters for the fault membership function of the faulty tunnel fire detection device, and the fault membership function is analyzed to obtain the fault tracing analysis result; S110: Determine the first operation and maintenance difficulty coefficient of the active fault and the second operation and maintenance difficulty coefficient of the passive fault of the faulty tunnel fire detection equipment according to the fault tracing analysis result, match the operation and maintenance decision of corresponding difficulty according to the first operation and maintenance difficulty coefficient and the second operation and maintenance difficulty coefficient, and obtain the final operation and maintenance decision.

2. The fault inspection and operation and maintenance method of tunnel fire detection equipment according to claim 1 is characterized in that: The step S102 specifically includes the following steps: Obtain model specification parameter information of all the tunnel fire detection devices to be inspected in the target tunnel space, and obtain the established functions of each tunnel fire detection device to be inspected according to the model specification parameter information; Obtain migration trend vectors of different tunnel environments of the target tunnel space within a preset time period, and retrieve from a big data network several tunnel fire benchmark time series variation characteristics when a fire occurs in the target tunnel space under different tunnel environment conditions based on the migration trend vectors of the different tunnel environments; Preset multiple local regression points and bandwidth parameters according to the migration trend vectors of different tunnel environments, calculate the kernel function of the benchmark time series change characteristics of several tunnel fires and each local regression point with the bandwidth parameter as the local range, obtain multiple kernel function weights, and construct a local polynomial regression trend model according to the multiple kernel function weights; By solving the weighted least squares equation group included in the local polynomial regression trend model, the local polynomial regression coefficient is obtained, and the influence degree of different tunnel environments on the state trend of the tunnel fire benchmark time series change characteristics is determined according to the local polynomial regression coefficient; updating each tunnel fire benchmark time series variation feature based on the state trend influence degree to obtain multiple tunnel fire characteristics corresponding to each tunnel fire benchmark time series variation feature after update, and setting access points and access point areas based on the multiple tunnel fire characteristics; The OPTICS algorithm and the preset minimum number of access points are introduced, and the domain radius of the current access point is calculated according to the established function to obtain the core distance of the current access point. If the number of access points within the core distance is greater than the minimum number of access points, a distance check is performed on all neighboring access points of the current access point in the access point domain to output the reachable distance of each neighboring access point. Update the neighborhood access points based on the reachable distance and the core distance until the preset threshold is reached, add the current access point after the domain access point is updated to the cluster order, and continue to explore the remaining access points in the domain of the access point with distance check to generate a cluster order graph of all current access points; The cluster sequence diagrams of all current access points are analyzed to obtain clustering results, and a fault inspection clustering label for the tunnel fire of each tunnel fire detection device to be inspected is determined according to the clustering results.

3. The fault inspection and operation and maintenance method of tunnel fire detection equipment according to claim 1 is characterized in that: The step S104 specifically includes the following steps: Obtaining an original engineering design drawing of the target tunnel space, constructing a multi-dimensional virtual domain of the target tunnel space according to the original engineering design drawing, and obtaining a layout and installation drawing of different tunnel fire detection equipment to be inspected in the target tunnel space; The multi-dimensional virtual domain is divided into N sub-virtual domain grids based on the layout installation diagram, and the inspection log of the target tunnel space is obtained. The historical tunnel fire environment data text associated with the fault inspection clustering label of each tunnel fire detection device to be inspected for the tunnel fire is extracted from each sub-virtual domain grid at different time steps through the inspection log; The Cellular Automata simulation method is introduced to subdivide each sub-virtual domain grid into a number of state simulation units, and the fire state of the state simulation unit at different time sequence steps is determined according to the historical tunnel fire environment data text, and the fire time sequence state of each state simulation unit is obtained; Calculate the state transition probability matrix of the fire sequence state of the current state simulation unit based on the fire sequence state so that the next state simulation unit will synchronously appear the current fire sequence state, and design and define the fire state transition probability rules of the adjacent state simulation units according to the state transition probability matrix; Taking different time sequence steps as the updating premise, the fire time sequence state of each state simulation unit is iteratively updated based on the fire state transition probability rule to generate a new fire time sequence state of each state simulation unit; The Cellular Automata simulation method is used to simulate and express the new fire time sequence state of all state simulation units, and the tunnel fire environment simulation field where the tunnel fire detection equipment to be inspected is located in the sub-virtual domain grid is obtained; Based on the tunnel fire environment simulation field where the tunnel fire detection equipment to be inspected is located in the sub-virtual domain grid, a fault inspection excitation signal is constructed, and the fault inspection excitation signal is uploaded to the fault inspection terminal corresponding to the tunnel fire detection equipment to be inspected to obtain a fault inspection strategy.

4. The fault inspection and operation and maintenance method of tunnel fire detection equipment according to claim 1 is characterized in that: The step S106 specifically includes the following steps: Obtain the actual detection delay rate and actual hysteresis amplitude of the fault inspection excitation signal corresponding to the detection of the current tunnel fire detection device to be inspected, and obtain the detection process and detection response logic of the current tunnel fire detection device to be inspected to execute the fault inspection strategy; A discrete difference equation is constructed according to the detection process and the detection response logic, and the actual detection delay rate and the actual hysteresis amplitude are introduced into the discrete difference equation to update the hysteresis state, so as to obtain the hysteresis characteristic state equation of the tunnel fire detection equipment to be inspected; Solving the hysteresis characteristic state equation of the tunnel fire detection device to be inspected to obtain an open-loop transfer function of the hysteresis characteristic, and determining the open-loop phase margin of the tunnel fire detection device to be inspected according to the open-loop transfer function of the hysteresis characteristic; Obtaining an ideal tunnel fire detection requirement for the tunnel fire detection device to be inspected, extracting an expected detection index threshold when the tunnel fire detection device to be inspected detects a tunnel fire through the ideal tunnel fire detection requirement, and constructing a hysteresis compensation architecture based on the open-loop transfer function; Presetting a phase margin threshold according to an expected detection indicator threshold, updating a hysteresis compensation zero point and a hysteresis compensation pole of a hysteresis compensation architecture based on the open-loop phase margin until the open-loop phase margin is greater than the phase margin threshold, and obtaining a hysteresis compensation parameter; A detection fault sparse coding set is constructed based on the expected detection index threshold, and the hysteresis compensation parameter is converted into multiple sub-sparse codes. If multiple sub-sparse codes do not exist in the detection fault sparse coding set, the tunnel fire detection device to be inspected is calibrated as a normal tunnel fire detection device; If at least one or more sub-sparse codes appear in the detection fault sparse code set, the tunnel fire detection device to be inspected is calibrated as a faulty tunnel fire detection device, and a fault inspection result is obtained.

5. The fault inspection and operation and maintenance method of tunnel fire detection equipment according to claim 1 is characterized in that: The step S108 specifically includes the following steps: Extract all faulty tunnel fire detection devices based on the fault inspection results, obtain the control network associated with each faulty tunnel fire detection device, apply the hysteresis compensation parameters to the corresponding faulty tunnel fire detection device, and obtain the control configuration parameters of the faulty tunnel fire detection device; The bus topology method is introduced to control the topology of the control network associated with each faulty tunnel fire detection device to obtain a control network topology map. The control execution node in each faulty tunnel fire detection device is obtained through the control network topology map, and the active and passive attributes of the control execution node are defined. Analyze the feasibility of the control configuration parameters and the execution hysteresis compensation parameters of the active control execution node and the passive control execution node to list several feasible decision schemes, calculate the evaluation fuzzy numbers of each feasible decision scheme on the active attribute and the passive attribute one by one, and construct a decision matrix according to the evaluation fuzzy numbers; A hierarchical analysis method is introduced to calculate the attribute weights of the active attributes and the passive attributes to obtain attribute fuzzy weight values, and a fuzzy weighted average value of each feasible decision scheme in the decision matrix is ​​calculated based on the evaluation fuzzy number and the attribute fuzzy weight value, and a fault membership function of the control execution node for generating current lag compensation for the faulty tunnel fire detection equipment is determined according to the fuzzy weighted average value; Preset the fault membership function interval of active attributes and the fault membership function interval of passive attributes, and if the fault membership function is within the fault membership function interval of active attributes, mark the control execution node as an active attribute fault node; If the fault membership function is within the fault membership function interval of the passive attribute, the control execution node is marked as a passive attribute fault node, and the fault tracing analysis result is obtained.

6. The fault inspection and operation and maintenance method of tunnel fire detection equipment according to claim 1 is characterized in that: The step S110 specifically includes the following steps: If the fault tracing analysis result of the control execution node shows that it is an active attribute fault node, multiple detection activity trajectories of the control fault tunnel fire detection device when the active attribute fault node fails are obtained, and an actual detection activity trajectory model of the fault tunnel fire detection device is constructed according to the multiple detection activity trajectories; Obtaining a preset detection activity control strategy for controlling a faulty tunnel fire detection device when an active attribute fault node fails, and creating an accurate detection activity trajectory model according to the preset detection activity control strategy; A hash attention mechanism is introduced to calculate a first hash attention score of the actual detection activity trajectory model compared to the accurate detection activity trajectory model, and an operation and maintenance difficulty coefficient of an active attribute failure of the faulty tunnel fire detection equipment is determined according to the first hash attention score, which is defined as a first operation and maintenance difficulty coefficient; If the fault tracing analysis result of the control execution node shows that it is a passive attribute fault node, the disturbance factor that causes the passive attribute fault of the control execution node under the current tunnel environment state is obtained based on the big data network, and the historical migration disturbance area of ​​the disturbance factor is extracted through the operation and maintenance log; Obtaining a control criterion for a passive attribute fault node, obtaining a preset detection execution range of the passive attribute fault node according to the control criterion, and calculating an interference area value between a historical migration disturbance region of a disturbance factor and the preset detection execution range; A threshold of allowable interference area is preset, and a second hash attention score of the interference area value compared with the threshold of allowable interference area is calculated through a hash attention mechanism. The operation and maintenance difficulty coefficient of the passive attribute failure of the faulty tunnel fire detection equipment is determined according to the second hash attention score, which is defined as the second operation and maintenance difficulty coefficient; Acquire several operation and maintenance decisions associated with the faulty tunnel fire detection equipment and the operation and maintenance difficulty level solved by each operation and maintenance decision, construct a hash index table of several operation and maintenance decisions based on the operation and maintenance difficulty level, import the first operation and maintenance difficulty coefficient and the second operation and maintenance difficulty coefficient into the hash index table for index query, obtain the final operation and maintenance decision and upload it to the inspection and operation and maintenance robot terminal.

7. A fault inspection and operation and maintenance device for tunnel fire detection equipment, characterized in that: The fault inspection and operation and maintenance device includes a memory and a processor, and the memory stores a fault inspection and operation and maintenance method program for tunnel fire detection equipment. When the fault inspection and operation and maintenance method program for tunnel fire detection equipment is executed by the processor, the fault inspection and operation and maintenance method steps for the tunnel fire detection equipment as described in any one of claims 1 to 6 are implemented.

8. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer detection program, and when the computer detection program is executed by at least one processor, it implements the fault inspection and operation and maintenance method of the tunnel fire detection equipment according to any one of claims 1 to 6.

Images