Fire-fighting equipment intelligent inspection and fault early warning method based on Internet of Things

Abstract

The invention relates to the technical field of Internet of Things, in particular to a fire fighting equipment intelligent inspection and fault early warning method based on the Internet of Things. The method comprises the following steps: acquiring fire-fighting equipment data; performing spatial topological structure analysis according to the fire fighting equipment data to obtain spatial topological structure data; generating an intelligent inspection path according to the spatial topological structure data to obtain intelligent inspection path data; performing spray head structure feature extraction according to the fire-fighting equipment data to obtain spray head structure data; performing blockage detection according to the structural data of the spray header to obtain blockage data of the spray header; performing metal surface corrosion detection based on the spray header blockage data to obtain metal surface corrosion data; and fire hydrant water hammer effect simulation is performed according to the fire-fighting equipment data to obtain fire hydrant water hammer effect data. The intelligent management efficiency and the fault detection accuracy of fire fighting equipment maintenance are improved based on the Internet of Things technology.

Description

Technical Field

[0001] The present invention relates to the technical field of Internet of Things, and in particular to an intelligent inspection and fault warning method for fire-fighting equipment based on the Internet of Things. Background Art

[0002] Traditional intelligent inspection and fault warning systems for firefighting equipment focus on the collection and analysis of a single type of data, lacking the comprehensive integration of multi-dimensional firefighting equipment status data. This results in inaccurate and incomplete fault diagnosis. Inspection route planning often relies on static or fixed paths, lacking the ability to dynamically adjust inspection strategies based on equipment status, hindering efficient resource allocation and timely response. Fault warnings often rely on threshold alarms, making it difficult to detect potential early signs of equipment anomalies. They also lack support from deep learning and intelligent analysis, resulting in low alert timeliness and accuracy. Specific identification and analysis of hardware anomalies is crude, failing to fully integrate device structural characteristics with physical simulation results, limiting accurate assessment of equipment lifespan and risk. Overall system topology analysis and functional linkages are weak, preventing coordinated detection and coordinated warnings across firefighting equipment, impacting the overall level of intelligent fire safety management. These shortcomings hinder the practical effectiveness and widespread application of traditional IoT-based intelligent inspection and fault warning systems for firefighting equipment. Summary of the Invention

[0003] Based on this, it is necessary for the present invention to provide an intelligent inspection and fault warning method for fire-fighting equipment based on the Internet of Things to solve at least one of the above technical problems.

[0004] To achieve the above objectives, a firefighting equipment intelligent inspection and fault warning method based on the Internet of Things includes the following steps: Step S1: Obtain firefighting equipment data; perform spatial topology analysis based on the firefighting equipment data to obtain spatial topology data; generate an intelligent inspection path based on the spatial topology data to obtain intelligent inspection path data; Step S2: extracting sprinkler head structural features based on the firefighting equipment data to obtain sprinkler head structural data; performing blockage detection based on the sprinkler head structural data to obtain sprinkler head blockage data; performing metal surface corrosion detection based on the sprinkler head blockage data to obtain metal surface corrosion data; Step S3: Performing a fire hydrant water hammer effect simulation based on the fire-fighting equipment data to obtain fire hydrant water hammer effect data; performing a valve leakage test based on the fire hydrant water hammer effect data to obtain valve leakage data; performing a material aging analysis based on the valve leakage data to obtain material aging data; Step S4: identifying a fault hotspot area according to the metal surface corrosion data and the material aging data to obtain fault hotspot area data; performing risk priority sorting according to the fault hotspot area to obtain risk priority data; and dynamically adjusting an intelligent inspection path according to the risk priority to obtain dynamic inspection adjustment path data.

[0005] The application can accurately construct the spatial relationship between devices by analyzing the spatial topology of the fire-fighting equipment data, reasonably plan the intelligent inspection path, improve the inspection efficiency and reduce resource waste. The accurate feature extraction and blockage detection of the spray head structure can help to timely find the problems of abnormal spray hole diameter and internal flow passage blockage, and combined with the metal surface corrosion detection, the corrosion condition of the equipment hardware can be effectively monitored to prevent the performance degradation of the equipment. The water hammer effect simulation of the fire hydrant can truly reflect the stress condition of the pipe network, assist in identifying the valve leakage problem, and further evaluate the equipment life and potential failure risk through the material aging analysis. The fault hotspot area identification based on the metal corrosion and material aging is beneficial to accurately positioning the high-risk area, realizing the risk priority sorting, and ensuring the pertinence and scientificity of the inspection work. Finally, through the dynamic adjustment of the intelligent inspection path, the real-time optimization of the inspection work is realized, and the intelligent level and safety guarantee capability of the fire-fighting equipment maintenance are improved.

[0006] Preferably, the step S1 is specifically: Step S11: obtaining fire-fighting equipment data; Step S12: extracting spatial coordinate data according to the fire-fighting equipment data; Step S13: calculating a spatial distance matrix based on the spatial coordinate data to obtain spatial distance matrix data; and identifying a spatial connection relationship according to the spatial distance matrix data to obtain connection relationship data; Step S14: drawing a spatial topology graph according to the connection relationship data; annotating node attributes based on the spatial topology graph to obtain node attribute data; and analyzing the spatial topology structure according to the spatial topology graph and the node attribute data to obtain spatial topology structure data; Step S15: generating an intelligent inspection path according to the spatial topology structure data to obtain intelligent inspection path data.

[0007] The application realizes accurate analysis on the spatial position and connection relationship of the fire-fighting equipment by comprehensive collection and processing of multi-dimensional data, constructs detailed spatial topology, improves the visualization and understanding ability of the relationship between the equipment, and effectively solves the problem of weak topology analysis in the traditional method. The intelligent inspection path generated based on the spatial topology not only reasonably optimizes the inspection sequence, but also realizes the basis of dynamically adjusting the inspection strategy, improves the inspection efficiency and resource utilization, and overcomes the limitations of traditional fixed path planning. The process fully considers the node attributes and connection relationship, ensures comprehensive and prominent inspection coverage, provides solid data support for subsequent fault diagnosis and early warning, thereby significantly enhances the accuracy and timeliness of fault identification, and improves the intelligent level and practicability of the entire fire-fighting equipment management system.

[0008] Preferably, step S15 is specifically: Step S151: calibrating the inspection node according to the spatial topology structure data to obtain inspection node data; Step S152: identifying the node connectivity relationship based on the inspection node data to obtain node connectivity relationship data; Step S153: performing path reachability analysis according to the node connectivity relationship data to obtain a set of inspection feasible paths; Step S154: performing shortest path intelligent analysis based on the set of inspection feasible paths to obtain intelligent inspection path data.

[0009] The application realizes effective identification and management of key positions of fire-fighting equipment by accurate calibration of inspection nodes, ensures the comprehensiveness and pertinence of inspection coverage. Based on the identification of node connectivity relationship, the actual connection between the equipment is accurately reflected, providing a reliable data basis for path planning. Path reachability analysis ensures the rationality and feasibility of the inspection path, avoids the inspection blind area caused by path interruption or obstacles, and effectively improves the continuity and integrity of the inspection. The shortest path intelligent analysis maximizes the reduction of inspection cost and time while ensuring the inspection efficiency, improves the resource utilization, solves the problem of static and low efficiency of path planning in the traditional method. The whole path planning process realizes dynamic and scientific inspection strategy, enhances the intelligence and response speed of the inspection system, provides a solid guarantee for the state monitoring and fault warning of the fire-fighting equipment, and improves the fine level and safety of equipment management.

[0010] Preferably, step S2 is specifically: Step S21: extracting the spray head structure features according to the fire-fighting equipment data to obtain spray head structure data; Step S22: performing spray orifice blockage detection according to the spray head structure data to obtain spray orifice blockage data; Step S23: Perform internal flow channel impurity accumulation detection according to the shower head structure data to obtain flow channel impurity accumulation data; Step S24: Integrate the shower hole diameter blockage data and the flow channel impurity accumulation data to obtain shower head blockage data; Step S25: Perform metal surface corrosion detection based on the shower head blockage data to obtain metal surface corrosion data.

[0011] The present application improves the accuracy and comprehensiveness of fault diagnosis through comprehensive collection and fusion of multi-dimensional fire-fighting equipment data, realizes deep understanding and intelligent analysis of the equipment state. The accurate construction of spatial topology makes the connection relationship between equipment more clear, effectively supports the scientific planning and dynamic adjustment of intelligent inspection path, and significantly improves the inspection efficiency and resource utilization. For key hardware components such as shower heads and fire hydrants, combined with blockage detection, water hammer effect simulation and valve leakage identification, the physical state and potential abnormalities of the equipment can be monitored in detail, and the identification accuracy of hardware faults is improved. The introduction of material aging and metal corrosion data realizes the accurate evaluation of equipment life and risk, assists in identifying fault hotspots and risk priority, and ensures that key risks are paid attention to and handled in time. The overall scheme breaks through the limitations of traditional static inspection, enhances the function linkage and collaborative early warning capability between equipment, promotes the intelligent level of fire safety management, and improves the practical value and promotion potential of the system.

[0012] Preferably, step S22 specifically comprises: Step S221: Identify the shower hole position according to the shower head structure data to obtain shower hole position data; Step S222: Perform hot smoke shower simulation based on the shower hole position data to obtain hot smoke shower data; Step S223: Perform smoke diffusion analysis according to the hot smoke shower data to obtain smoke diffusion data; Step S224: Calculate the smoke dust particle size based on the smoke diffusion data to obtain smoke dust particle size data; Step S225: Calculate the shower hole diameter according to the shower hole position data to obtain shower hole diameter data; Step S226: Perform blockage analysis according to the smoke dust particle size data and the shower hole diameter data to obtain shower hole diameter blockage data.

[0013] The application realizes accurate positioning of key spray units by accurately identifying the position of the spray hole, provides reliable basic data for subsequent hot smoke spraying simulation, and can truly reflect the distribution characteristics of smoke and heat at the initial stage of fire. Based on the data obtained by hot smoke spraying simulation, smoke diffusion analysis is carried out, which helps to reveal the diffusion path and concentration change of smoke in space, and enhances the understanding and prediction ability of fire environment. By calculating the particle size of smoke, the physical properties of smoke can be accurately evaluated, and scientific basis is provided for the judgment of blockage risk. Combined with the calculation of spray hole diameter, the matching degree of spray hole diameter and smoke particle size is accurately analyzed, the blockage of spray hole diameter is effectively identified, and the spraying effect and fire safety of the spray system are guaranteed. The overall method breaks through the limitations of traditional single data analysis, combines multi-dimensional physical simulation and structural feature analysis, significantly improves the accuracy and reliability of spray head blockage detection, provides solid technical support for intelligent inspection and fault warning, and improves the maintenance efficiency and operation safety of fire fighting equipment.

[0014] Preferably, step S23 is specifically: Step S231: identifying the flow passage cross section form according to the spray head structure data to obtain flow passage cross section form data; Step S232: performing flow passage expansion analysis based on the flow passage cross section form data to obtain flow passage expansion data; Step S233: calculating the flow velocity according to the flow passage expansion data to obtain flow velocity data; Step S234: identifying the flow velocity slowing down area according to the flow velocity data; Step S235: performing corrosion particle shedding detection according to the flow velocity slowing down area to obtain corrosion particle shedding data; Step S236: performing low-speed vortex area analysis based on the corrosion particle shedding data to obtain low-speed vortex area data; Step S237: determining the flow passage impurity accumulation according to the low-speed vortex area data to obtain flow passage impurity accumulation data.

[0015] The present application can accurately reflect the actual physical structure of the fluid channel inside the spray head by accurately identifying the cross-sectional shape of the flow channel and analyzing the expansion of the flow channel, improve the understanding of the change of the spray flow rate, help to find the flow rate slowing area in time, and warn the potential blockage risk. The corrosion particle shedding detection and low-speed vortex area analysis effectively reveal the impurity deposition and local fluid dynamics anomaly in the flow channel, and comprehensively reflect the hardware corrosion and pollution condition inside the flow channel. Through accurate determination of the impurity accumulation in the flow channel, scientific basis can be provided for maintenance of the spray system, guiding cleaning and repair work, and prolonging the service life of the equipment. The overall method improves the early identification ability of the hardware anomaly of the spray head, realizes the deep integration of multi-dimensional physical characteristics and fluid dynamics, enhances the comprehensiveness and accuracy of fault diagnosis, provides solid technical support for intelligent inspection and fault warning, and promotes the development of fire-fighting equipment maintenance management towards intelligence and precision.

[0016] Preferably, step S3 is specifically: Step S31: extracting fire-fighting pipe network features according to fire-fighting equipment data to obtain fire-fighting pipe network data; Step S32: identifying material elastic modulus according to the fire-fighting pipe network data; Step S33: determining shock wave speed data according to the material elastic modulus; Step S34: simulating fire hydrant water hammer effect according to the shock wave speed data to obtain fire hydrant water hammer effect data; Step S35: detecting valve leakage according to the fire hydrant water hammer effect data to obtain valve leakage data; Step S36: analyzing material aging according to the valve leakage data to obtain material aging data.

[0017] The present application extracts the features of the fire-fighting pipe network, comprehensively masters the structure and material information of the pipe network, provides accurate basic data for subsequent analysis, helps to accurately identify the elastic modulus of different materials, and further scientifically calculates the shock wave speed, truly simulates the water hammer effect, and reveals the dynamic response characteristics of the pipe network under sudden water pressure change. Based on the water hammer effect data, the valve leakage detection is realized, the sealing state of the valve is evaluated in detail, the leakage risk is effectively identified, and the water pressure loss and equipment damage caused by leakage are avoided. Through material aging analysis, the aging degree of the valve and pipe network components can be dynamically monitored, potential failure risks can be predicted, maintenance and replacement plans can be guided, the service life of the equipment can be prolonged, and the reliability and safety of the system operation can be improved. The overall scheme strengthens the fusion and dynamic simulation of multi-dimensional data, improves the accuracy of fault diagnosis and the timeliness of early warning, realizes the intelligent and fine management of fire-fighting equipment maintenance.

[0018] Preferably, step S35 is specifically: Step S351: Calculate the water hammer instantaneous impact load according to the fire hydrant water hammer effect data to obtain water hammer impact load data; Step S352: Identify the valve force sealing surface based on the water hammer impact load data to obtain valve force sealing surface data; Step S353: Perform sealing ring deformation detection based on the valve force sealing surface data to obtain sealing ring deformation data; Step S354: Calculate the edge lifting height according to the sealing ring deformation data to obtain edge lifting height data; Step S355: Perform valve leakage evaluation according to the edge lifting height data to obtain valve leakage data.

[0019] The present application comprehensively collects and fuses multi-dimensional fire-fighting equipment data, comprehensively improves the accuracy and comprehensiveness of fault diagnosis, realizes deep understanding and intelligent analysis of the equipment state. The spatial topology structure analysis clearly defines the connection relationship and spatial distribution between the equipment, optimizes the generation of intelligent inspection path, improves the inspection efficiency and resource utilization rate. The spray head blockage and metal corrosion detection finely monitors the hardware state of the key components, discovers the blockage and corrosion risk in advance, and guarantees the normal operation of the spray system. The fire hydrant water hammer effect simulation combined with valve leakage detection and material aging analysis realizes accurate evaluation of the mechanical damage and sealing performance of the key valve, improves the accuracy of fault identification. Through the identification of fault hot area and risk priority ranking, the dynamic inspection path is scientifically arranged, the reasonable allocation of inspection resources and the priority monitoring of key parts are realized, and the intelligentization of equipment management and the timeliness of early warning are significantly enhanced. The overall scheme effectively overcomes the shortcomings of single data analysis and static inspection of traditional methods, promotes the intelligent inspection and fault warning of fire-fighting equipment to be efficient, accurate and collaborative, and improves the practical value and safety protection ability of the system.

[0020] Preferably, step S36 specifically comprises: Step S361: Perform valve core wear detection according to the valve leakage data to obtain valve core wear data; Step S362: Perform coating peeling analysis based on the valve core wear data to obtain coating peeling data; Step S363: Calculate the valve core surface exposure area based on the coating peeling data; Step S364: Measure the valve core roughness according to the valve core surface exposure area to obtain valve core roughness data; Step S365: Perform material aging evaluation based on the valve core roughness data to obtain material aging data.

[0021] The application realizes accurate detection of the wear and surface state of the key components of the valve through comprehensive analysis of multi-dimensional data, effectively reveals the wear degree and coating peeling of the valve core, and further quantifies the exposed area and roughness of the valve core surface, comprehensively reflecting the aging state of the material. This detailed hardware state evaluation improves the accuracy and foresight of fault diagnosis, helps to discover potential risks in time, and prevents system failure caused by valve failure. By combining structural features and physical wear analysis, the scientific judgment ability of the service life and maintenance needs of the equipment is enhanced, supporting more reasonable maintenance plans and resource allocation. At the same time, the depth of understanding of the state of the fire-fighting equipment is strengthened, promoting the dynamic optimization of the intelligent inspection path and the accurate response of the fault warning, significantly improving the safety protection capability and operation efficiency of the fire-fighting system, effectively making up for the shortcomings of single data, rough analysis and slow response in traditional methods.

[0022] Preferably, step S4 specifically comprises: Step S41: demarcate the metal surface corrosion area according to the metal surface corrosion data; Step S42: demarcate the material aging area according to the material aging data; Step S43: perform intersection operation on the fault hot spot area according to the metal surface corrosion area and the material aging area, to obtain fault hot spot area data; Step S44: evaluate the fault severity according to the fault hot spot area, to obtain fault severity data; Step S45: perform risk scoring based on the fault severity data, to obtain fault risk data; perform risk priority sorting according to the fault risk data, to obtain risk priority data; Step S46: dynamically adjust the intelligent inspection path data according to the risk priority, to obtain dynamic inspection adjustment path data.

[0023] The application realizes scientific identification and positioning of the fault hot spot area through accurate demarcation of the metal surface corrosion and material aging area, effectively focuses on the high-risk parts of the equipment, and improves the pertinence and efficiency of fault diagnosis. Through intersection operation comprehensive analysis of different abnormal data, the identification ability of complex faults is enhanced, and misjudgment and omission caused by single index are avoided. The fault severity evaluation and risk scoring mechanism objectively quantifies the risk level of the equipment, providing a reliable basis for management decision-making. Based on the dynamic inspection path adjustment of the risk priority, the rational optimization allocation of the inspection resources is realized, the flexibility and response speed of the inspection work are improved, and the maintenance cost and sudden failure rate are effectively reduced. The overall method breaks through the limitations of the traditional static inspection mode, enhances the accuracy and real-time performance of the intelligent inspection and fault warning of the fire-fighting equipment, promotes the intelligent management level of the system, and improves the overall efficiency of the fire safety protection. BRIEF DESCRIPTION OF DRAWINGS

[0024] Other features, objects, and advantages of the application will become more apparent from the following detailed description when read in connection with the following drawings: Fig. 1 A step flowchart diagram of a fire-fighting equipment intelligent patrol and fault early warning method based on the Internet of Things according to the present application; Fig. 2 A detailed step flowchart diagram of step S1 in the present application; Fig. 3 A detailed step flowchart diagram of step S15 in the present application; The implementation, functional features and advantages of the present application will be further described with reference to the embodiments and the accompanying drawings. DETAILED DESCRIPTION

[0025] The technical method of the present application will be described clearly and completely below in combination with the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.

[0026] In addition, the accompanying drawings are only schematic illustrations of the present application, and are not necessarily drawn to scale. The same reference numerals in the drawings represent the same or similar parts, and thus repeated descriptions thereof will be omitted. Some of the block diagrams shown in the drawings are functional entities, which do not necessarily have to correspond to physically or logically independent entities. The functional entities can be implemented in the form of software, or in one or more hardware modules or integrated circuits, or in different network and / or processor methods and / or microcontroller methods.

[0027] It should be understood that although the terms "first", "second" and the like can be used herein to describe various elements, these elements should not be limited by these terms. These terms are only used to distinguish one element from another. For example, without departing from the scope of the exemplary embodiments, a first element can be called a second element, and similarly a second element can be called a first element. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0028] To achieve the above-mentioned purpose, please refer to Figs. 1 to 3 The present application provides a fire-fighting equipment intelligent patrol and fault early warning method based on the Internet of Things, which comprises the following steps: Step S1: Obtain fire-fighting equipment data; perform spatial topology analysis according to the fire-fighting equipment data to obtain spatial topology data; generate an intelligent inspection path according to the spatial topology data to obtain intelligent inspection path data; In this embodiment, the fire-fighting equipment data is collected by various Internet of Things sensors deployed on various fire-fighting facilities. The collected data includes spatial coordinates (X, Y, Z three-dimensional coordinates with an accuracy of within 0.1 meters), equipment type, installation height, pipe network connection relationship, etc. The spatial coordinates are obtained by a high-precision RTK (Real-Time Kinematic) system, and the pipe network connection information is obtained from a pre-established fire-fighting facility database and verified in combination with the actual layout on site. The collected data is first cleaned to remove invalid and abnormal values, and a moving average filter is used to reduce sensor noise. According to the cleaned spatial coordinate data, a spatial distance matrix is constructed to calculate the Euclidean distance between any two equipment nodes, and devices with a distance less than or equal to 50 meters are considered as adjacent nodes. The spatial connection relationship is generated by judging the distance between nodes and the pipe network database connection information, and the connection relationship is stored in the form of an adjacency matrix. Based on the adjacency matrix, a spatial topology graph is drawn using the Depth-First Search (DFS) algorithm in graph theory, and the node attributes include equipment type, state, and pipe diameter parameters. The intelligent inspection path is generated based on the spatial topology data. First, all equipment nodes are determined as inspection nodes, and the Dijkstra algorithm is used to calculate the shortest path between nodes. The total distance of the inspection path and the importance weight of the equipment are considered comprehensively. The importance weight of the equipment is assigned according to the historical failure rate of the equipment (range 0.1 to 1.0), and the weight of important equipment is higher. According to the above path planning results, a sequential inspection path list is generated. The path data format includes node serial number, path length, and estimated inspection time. The inspection time is set to 2 minutes per node based on the historical inspection speed, and the moving speed is set to 1.2 meters per second.

[0029] Step S2: Extract the spray head structure features according to the fire-fighting equipment data to obtain spray head structure data; perform clogging detection according to the spray head structure data to obtain spray head clogging data; perform metal surface corrosion detection based on the spray head clogging data to obtain metal surface corrosion data; In this embodiment, the spray head structure feature extraction is based on high-definition industrial endoscope image acquisition, which adopts multi-angle and multi-scale shooting, and the image resolution requirement is not less than 1920x1080 pixels. Through image preprocessing, including denoising (using a median filter with a filter window of 3x3 pixels), enhancing contrast (histogram equalization), and improving subsequent detection accuracy. The edge detection operator (such as Canny algorithm, low threshold 0.1, high threshold 0.3) is used to identify the spray hole edge morphology and measure the spray hole diameter size, and the hole diameter size error is controlled within ±0.05 millimeters. The flow sensor data is used for clogging detection, and the flow threshold is less than 70% of the normal design flow to determine potential clogging. Combined with the area of the clogging material (such as dust, rust, impurities) in the image, which accounts for more than 15% of the spray hole area, the clogging state is confirmed. The clogging data includes clogging position, clogging area percentage, and clogging material type (classified according to image color and texture, mainly divided into rust, silt, and organic matter). Metal surface corrosion detection relies on electrochemical impedance spectroscopy (EIS) technology, which collects impedance values of different areas on the surface of the spray head, and areas with impedance threshold values below 5kΩ·cm² are determined as corrosion areas. High-definition image analysis is used to analyze the morphology of corrosion spots, and spots with an area greater than 2% of the spray head surface area are considered significant corrosion. The corrosion data includes corrosion location coordinates, corrosion degree (quantified by impedance value), and corrosion area percentage.

[0030] Step S3: simulate the water hammer effect of the fire hydrant according to the fire-fighting equipment data to obtain fire hydrant water hammer effect data; detect valve leakage according to the fire hydrant water hammer effect data to obtain valve leakage data; and analyze material aging according to the valve leakage data to obtain material aging data; In this embodiment, the simulation of the water hammer effect of the fire hydrant is based on the actual material and structure parameters of the fire-fighting pipe network, with a pipe diameter range of 0.1 to 0.3 meters and materials including ductile cast iron, steel pipe and PVC. Pressure sensors are used to collect pressure changes in the pipe network, with a sampling frequency of 1 kHz, and a pressure change amplitude exceeding 0.2 MPa / s triggers water hammer event recording. Combined with the elastic modulus of the pipe material (steel pipe about 2.0x10 9Pa) and pipe wall thickness data, calculate the shock wave propagation speed (according to the pipe wave speed formula), simulate the water hammer pressure wave propagation process by numerical method (finite difference method), and generate the water hammer pressure curve. The valve leakage detection uses an ultrasonic sensor to capture the leakage sound signal. When the leakage sound pressure level exceeds 60 dB and the duration exceeds 10 seconds, it is determined to be in a leakage state. Combined with the valve opening angle sensor data (error ± 1 degree), the stress state of the sealing surface is analyzed to further confirm the leakage cause. The leakage data includes leakage location, leakage sound pressure level, leakage duration, and current valve opening degree. The material aging analysis comprehensively utilizes the environmental exposure time of the valve and pipe material, working temperature (measured by on-site sensor, accuracy ± 0.5°C), water quality pH value (collected by online water quality monitor, error ± 0.1), and other parameters, combined with the laboratory accelerated aging test database, to calculate the material fatigue degree. Using the fatigue life model, set the safety life threshold to 80% of the design life, and if it exceeds, it is determined to be severely aged. The aging data specifically includes fatigue life percentage, corrosion rate, and estimated remaining life.

[0031] Step S4: According to the metal surface corrosion data and the material aging data, the fault hotspot area is identified to obtain fault hotspot area data; according to the fault hotspot area, the risk priority is sorted to obtain risk priority data; according to the risk priority, the dynamic inspection path of the intelligent inspection path data is adjusted to obtain dynamic inspection adjustment path data.

[0032] In this embodiment, the fault hotspot area identification is based on spatial coordinate data. First, the metal surface corrosion data and the material aging data are superimposed in three-dimensional space, and the overlapping area of corrosion and aging is automatically identified by clustering algorithm (DBSCAN, neighborhood radius 0.5 meters, minimum point number 5) to delimit the fault hotspot area. The fault hotspot area data includes area center coordinates, coverage range, and cumulative corrosion and aging index (the index calculation formula is the weighted average of corrosion area percentage and aging percentage, with each weight being 50%). The risk priority sorting adopts a weighted scoring mechanism to calculate the risk score of each fault hotspot area. The scoring formula is: risk score = 0.6 x corrosion index + 0.4 x aging index + 0.2 x historical fault frequency (score after normalization processing), and the score range is 0-1. According to the score from high to low, a risk priority list is generated. The dynamic inspection path adjustment is based on the risk priority data, and the high-risk areas are preferentially included in the inspection path. The improved heuristic genetic algorithm is used to recalculate the inspection path, and the path length change is controlled within ± 10% of the original path to ensure the balance between inspection efficiency and risk coverage rate. The adjusted path data includes new node order, total path length, and estimated inspection time.

[0033] Preferably, step S1 specifically comprises: Step S11: Obtain fire-fighting equipment data; In this embodiment, the acquisition of fire-fighting equipment data is completed by various sensors installed on various types of fire-fighting facilities, including but not limited to positioning sensors (such as high-precision GPS modules, RTK differential positioning devices), pressure sensors, flow sensors, temperature and humidity sensors, and equipment state monitoring sensors. The accuracy of the positioning sensor should reach centimeter level, and the sampling frequency is set to 1 Hz to ensure real-time updating of spatial data. The data collected by all sensors is transmitted to the data processing center through wired or wireless communication modules, and the data format is unified to the standardized JSON or XML structure, including device unique identifier, timestamp, sensor reading and device category fields. The data receiving end uses multi-thread asynchronous processing technology to receive and store data in real time, and the storage database uses a time series database (such as InfluxDB) to support efficient time series data query and management. In the data cleaning link, threshold filtering method is used to eliminate abnormal data points, such as data points with positioning error exceeding 5 meters will be discarded, and sensor readings exceeding the physical measurement range are also judged as abnormal.

[0034] Step S12: extracting spatial coordinate data according to the fire-fighting equipment data; In this embodiment, spatial coordinate data is extracted from the acquired fire-fighting equipment data, including X, Y, Z coordinate values in a three-dimensional coordinate system. The coordinate system uses the national geodetic coordinate system (such as WGS-84) to ensure compatibility with geographic information system (GIS) data. The positioning information field is parsed from the original sensor data and converted to floating point format, and the coordinate value accuracy is required to reach four decimal places, ensuring meter-level positioning accuracy. After extraction, all spatial coordinate data is summarized into a spatial database (such as PostGIS) for subsequent spatial analysis. Data verification includes comparing historical equipment installation locations, screening out abnormal coordinates outside the preset geographic range (such as ±0.001 degrees of latitude and longitude), and ensuring the spatial continuity and accuracy of the data.

[0035] Step S13: calculating a spatial distance matrix based on the spatial coordinate data to obtain spatial distance matrix data; identifying a spatial connection relationship based on the spatial distance matrix data to obtain connection relationship data; In this embodiment, based on the spatial coordinate data extracted in step S12, the spatial distance between any two device nodes is calculated using the Euclidean distance formula, and the specific calculation formula is d=√[(x2-x1)²+(y2-y1)²+(z2-z1)²]. The calculation process is batch executed through an efficient matrix calculation library (such as NumPy), and a spatial distance matrix is generated. The matrix is a symmetric matrix, and the matrix element d(i,j) represents the spatial distance between the i-th and j-th devices, and the distance unit is unified as meters. After the calculation is completed, the connection relationship is identified according to the preset spatial connection threshold, and the threshold standard is 50 meters. The device nodes with a distance less than or equal to the threshold are considered to have a connection relationship. The threshold is determined according to the fire pipe network design specification, and can cover the effective communication and maintenance range between normal devices. The connection relationship result is stored in the form of an adjacency matrix, and the matrix element value 1 represents that the two nodes are connected, and 0 represents no connection. All distance and connection data are stored in a relational database to support subsequent query and analysis.

[0036] Step S14: drawing a spatial topology graph according to the connection relationship data; labeling node attributes based on the spatial topology graph to obtain node attribute data; and performing spatial topology structure analysis based on the spatial topology graph and the node attribute data to obtain spatial topology structure data; In this embodiment, according to the connection relationship data identified in step S13, a graph drawing library (such as Graphviz or Gephi) is used to construct a spatial topology graph. The nodes in the topology graph represent fire-fighting devices, the node positions are determined by spatial coordinates, the edges represent the connection relationship between devices, and the attributes of the edges include connection strength (calculated based on the reciprocal of distance), pipe diameter specification and other information. Node attribute labeling includes device type (such as spray head, fire hydrant, valve), device state (normal, fault, maintenance), installation time, etc., which are loaded in batches through a database query interface. Spatial topology structure analysis uses graph theory-based index calculation methods, including node degree, node connectivity, network density, etc. The analysis results are stored as structured data for further path planning. All processing steps are automatically completed through scripts to ensure timely data updating. The topology graph data format adopts standard GraphML or JSON format for cross-platform calling.

[0037] Step S15: generating an intelligent inspection path according to the spatial topology structure data to obtain intelligent inspection path data.

[0038] In this embodiment, based on the spatial topology data obtained in step S14, an intelligent inspection path is generated using a path search algorithm. Specifically, Dijkstra's shortest path algorithm or A* algorithm is used to cover all inspection nodes with path weights taking into account the importance of the device and the spatial distance as comprehensive consideration factors. The importance of the device is assigned according to the historical failure rate and the criticality level of the device, with a numerical range of 0.1 to 1.0, and the spatial distance is directly taken from the distance matrix. The path cost function is designed as the weighted sum of path length and device importance. The maximum path length is set to 5 kilometers during path planning to avoid fatigue of the inspection personnel. The planning result is output as an ordered node list, including the length of each path and the estimated inspection time, with a fixed inspection speed of 1.2 meters per second and a fixed node stay time of 2 minutes. The data format is stored in CSV or JSON. The path planning module is implemented in C++, ensuring the efficiency of the algorithm, and the path information is updated in real time through the interface.

[0039] Preferably, step S15 specifically comprises: Step S151: calibrate the inspection nodes according to the spatial topology data to obtain inspection node data; In this embodiment, according to the spatial topology data, all fire-fighting device nodes are first selected from the spatial topology data. The fire-fighting device nodes include device unique identifier, spatial coordinates (X, Y, Z, unit: meter, local plane coordinates converted from WGS-84 coordinate system), device category (such as fire hydrant, sprinkler, valve) and device state information. For each node, according to the device category and the preset inspection requirement rule, it is judged whether it needs to be included in the inspection node range. The rule is specifically that all fire hydrants and valves are marked as inspection nodes by default, and only when the pipe diameter of a sprinkler is greater than 25 mm and the last maintenance is more than 30 days, the sprinkler is marked as an inspection node. The spatial position of the node is detected repeatedly, and if the distance between two nodes is less than 0.5 meters, it is considered as a repeated record and merged into a single node. The final output inspection node data includes node number, spatial coordinates, device category and inspection priority level (priority 1 to 5, 1 highest), which is calculated based on the importance of the device and the failure history statistics. All data structures are stored in a relational database for subsequent query and processing.

[0040] Step S152: identify the node connectivity based on the inspection node data to obtain node connectivity data; In this embodiment, the three-dimensional Euclidean distance between any two nodes is calculated using the node spatial coordinates, and the calculation formula is d = V [(x2-x1)2+(y2-y1)2+(z2-z1)2], unit: meter. Set the maximum threshold of node connectivity to 60 meters, and all nodes with a distance less than or equal to this threshold are considered connected. In addition to the spatial distance, the actual pipe connection relationship between nodes is checked in combination with the fire pipe network design parameters. The pipe diameter needs to be greater than or equal to 20 mm and the pipe material is steel pipe or high-density polyethylene (HDPE), and the connection relationship data is obtained through the pipe network design drawing electronic data or GIS database query. Use the above information to build a node connectivity relationship matrix, and the matrix element value 1 represents that two nodes are connected, and 0 represents that they are not connected. The connectivity relationship data format is an adjacency matrix, stored in a graph database, which supports efficient path query.

[0041] Step S153: Path reachability analysis is performed according to the node connectivity relationship data to obtain a set of inspection feasible paths; In this embodiment, a graph traversal algorithm (depth-first search DFS or breadth-first search BFS) is used to traverse the inspection node graph, and the starting point is set as the fire control center position node, and all connected nodes are traversed. For each path, the path length is calculated as the sum of the spatial distances corresponding to the connection edges, with the unit being meters. Set the maximum reachable path length threshold to 5000 meters, and exclude paths exceeding the threshold. According to the path length and node coverage rate, an effective path set is selected to ensure that each path covers at least 80% or more high-priority inspection nodes. The path data includes path node sequence, corresponding path length, and expected inspection time (calculated based on a fixed inspection speed of 1.2 meters / second). The path set is stored in JSON format, including path ID, node list, path length, and time, which supports subsequent path optimization algorithm calls.

[0042] Step S154: Intelligent analysis of the shortest path based on the set of inspection feasible paths is performed to obtain intelligent inspection path data.

[0043] In this embodiment, intelligent shortest path analysis is performed based on the set of feasible inspection paths obtained in step S153. This analysis method is based on the Dijkstra algorithm, calculating the shortest total path cost for each path. The path cost function is defined as a weighted sum, with weights including spatial distance (weight 0.6), node importance (weight 0.3), and equipment failure risk level (weight 0.1). The equipment failure risk level is statistically derived from historical failure data and ranges from 0 to 1. During the path search, high-risk and high-priority nodes are prioritized to ensure that the total path length does not exceed 5000 meters and that all important nodes are visited at least once. The algorithm is implemented in C++, utilizing a priority queue for efficient edge relaxation. The path results include the node visit order, the dwell time at each node (default 2 minutes), and the estimated total inspection duration. Intelligent inspection path data is exported in CSV format, with fields including path sequence number, node number, node coordinates, dwell time, and cumulative inspection duration, for the inspection system to access and adjust in real time.

[0044] Preferably, step S2 is specifically as follows: Step S21: extracting sprinkler head structural features based on firefighting equipment data to obtain sprinkler head structural data; In this embodiment, sprinkler head structural feature extraction is performed based on firefighting equipment data. First, the equipment parameters of the fire sprinkler system are collected, including the sprinkler head model, manufacturer, installation location, nozzle aperture, material, and installation angle. High-definition industrial endoscope cameras are used to capture images of the sprinkler head's internal structure, with a resolution of 1920×1080 pixels and a frame rate of 30 fps. The images are transmitted in real time to the image processing unit via a wired high-speed transmission interface. Image processing uses edge detection algorithms (such as the Canny operator) to identify the sprinkler aperture contour. Combined with image geometry measurement techniques, the actual nozzle aperture size is calculated using a calibration plate correction, with an accuracy of ±0.05 mm. Combining equipment manufacturing standards and design data, a three-dimensional structural model of the sprinkler head is generated, including parameters such as aperture size, flow channel morphology, and structural thickness. The data format uses a CAD-standard STEP file to facilitate subsequent analysis and processing.

[0045] Step S22: performing a spray aperture blockage detection based on the spray head structure data to obtain spray aperture blockage data; In this embodiment, according to the spray head structure data, the spray aperture blockage detection is performed, and an ultrasonic detection instrument is selected to detect the blockage of the inner cavity of the aperture. The ultrasonic frequency is set in the range of 5 MHz to 10 MHz, the ultrasonic pulse echo technology is used to measure the echo signal of the inner wall of the nozzle, and the attachment of impurities in the aperture is analyzed by the change of the echo amplitude. The echo amplitude threshold is set to -30 dB, and the amplitude below the threshold is determined as the blockage area. The echo signal is mapped to the position by combining the three-dimensional model of the aperture, and the spatial distribution diagram of the blockage area is output. The detection data record includes the aperture position coordinates, the blockage percentage area and the blockage thickness, and the thickness measurement error is controlled within ±0.1 mm. All detection data is saved in a structured format for easy database management.

[0046] Step S23: According to the spray head structure data, the internal flow channel impurity accumulation detection is performed to obtain flow channel impurity accumulation data; In this embodiment, according to the spray head structure data, the internal flow channel impurity accumulation detection is performed, a high-precision laser scattering sensor is arranged along the flow channel, the laser wavelength is set to 650 nm, the power is 10 mW, and the measurement angle is fixed at 45 degrees. The sensor real-time collects the concentration and size distribution data of suspended particulate matter in the flow channel, the detection particle size range is 0.1 to 10 microns, and the concentration measurement accuracy is ±2%. Combined with the spray head flow channel geometric model, three-dimensional positioning technology is used to realize spatial mapping of particle distribution, and a three-dimensional map of impurity accumulation in the flow channel is generated. The data acquisition frequency is 10 times per second, and the data is transmitted to the analysis server through industrial Ethernet. The impurity accumulation degree is calculated according to the ratio of the cumulative concentration of particles to the cross-sectional area of the flow channel, and the accumulation threshold is set to 5% of the cross-sectional area of the flow channel. If the threshold is exceeded, it is determined that the accumulation is serious.

[0047] Step S24: Integrate the spray aperture blockage data and the flow channel impurity accumulation data to obtain the spray head blockage data; In this embodiment, the spray aperture blockage data and the flow channel impurity accumulation data are integrated, and a weighted fusion algorithm is used for data fusion. The aperture blockage weight is set to 0.7, the flow channel impurity accumulation weight is set to 0.3, and the fusion calculation formula is: blockage comprehensive index = 0.7 x aperture blockage percentage + 0.3 x flow channel impurity accumulation index. The fusion result updates the spray head blockage state through the database interface, and the state classification includes normal (index <10%), light blockage (10% ≤ index <30%), medium blockage (30% ≤ index <60%) and heavy blockage (≥60%). All data formats are unified to JSON structure for easy system call and subsequent fault diagnosis analysis.

[0048] Step S25: Based on the spray head blockage data, the metal surface corrosion detection is performed to obtain the metal surface corrosion data.

[0049] In this embodiment, based on the clogging data of the spray head, the metal surface corrosion detection is carried out, and the electrochemical impedance spectroscopy (EIS) technology is used to measure the corrosion state of the metal surface of the spray head. The detection frequency range is set to 0.01 Hz to 100 kHz, and the alternating voltage amplitude is 10 mV. By measuring the changes of the electrode surface resistance and capacitance, the corrosion rate is calculated, and the corrosion rate threshold is set to 0.1 mm / year, and if the corrosion rate exceeds the threshold, it indicates that the corrosion is serious. Combined with the clogging data, the corrosion key area is marked by using the corrosion map drawing technology. The corrosion data includes the corrosion rate, corrosion area and position coordinates, and all the data are stored in a special metal maintenance database for subsequent maintenance strategy formulation and life prediction.

[0050] Preferably, step S22 specifically comprises: Step S221: identifying the spray hole part according to the spray head structure data to obtain spray hole part data; In this embodiment, first, the detailed structure design data of the spray head is obtained from the fire-fighting equipment database, including the overall size of the spray head, the nozzle arrangement diagram, the aperture distribution and the installation angle, etc. A high-precision three-dimensional scanner (resolution 0.01 mm) is used to physically scan the spray head to obtain the point cloud data of the surface and internal structure of the spray head. After denoising processing, the spatial geometric feature recognition algorithm is used to locate the nozzle hole part, and the spatial coordinates and shape parameters of the spray hole are extracted. The spray hole position data is stored in the form of XYZ three-dimensional coordinate system, and the aperture size measurement accuracy is controlled within ±0.02 mm. The hole position recognition process is filtered by setting the aperture size threshold (minimum aperture 0.5 mm, maximum aperture 5 mm) and the hole spacing threshold (not less than 1 mm), to ensure that the identified spray hole data is accurate and without overlap. The spray hole part data is saved in a structured table form, including hole number, spatial coordinates, aperture size, hole shape and other attribute information.

[0051] Step S222: based on the spray hole part data, a hot smoke spray simulation is carried out to obtain hot smoke spray data; In this example, a hot smoke shower simulation was performed using fire scene environmental parameters and sprinkler hole location data. Field environmental parameters, including temperature distribution (range: 20°C to 1200°C, measurement accuracy: ±1°C), airflow velocity (0 to 10 m / s, accuracy: ±0.1 m / s), and smoke concentration (0 to 5000 ppm, accuracy: ±5 ppm), were collected in real time by fixed environmental monitoring sensors. A physical spray test device was used to simulate the hot smoke flow field from the sprinkler holes within a control room, with the spray pressure set to 0.2 MPa and the spray duration set to 30 seconds. The sprinkler hole location data was used to locate the spray nozzles, and the spray device automatically aligned the sprinkler holes to ensure that the spray angle and aperture direction were consistent. The smoke concentration and temperature were dynamically measured using an infrared thermal imager and a laser light scattering instrument, with a data sampling frequency of 20 times per second. The hot smoke shower data was stored as a time series, containing the three-dimensional coordinate data of the smoke concentration, temperature, and spatial diffusion range.

[0052] Step S223: performing smoke diffusion analysis based on the hot smoke spray data to obtain smoke diffusion data; In this example, smoke diffusion analysis was performed using aerosol light scattering technology based on hot smoke spray data. A smoke diffusion channel with wind speed adjustment was set in the laboratory, with wind speeds set between 0 and 5 m / s and a resolution of 0.01 m / s. A laser particle counter was used to measure the smoke particle size distribution, ranging from 0.1 to 10 microns with an accuracy of ±0.1 micron. The smoke diffusion data included the temporal and spatial variations of smoke particle concentration. The spatial coordinates were based on a three-dimensional coordinate system at the spray orifice location, with a time resolution of 0.05 seconds. Multi-point sampling in the diffusion channel was used to obtain smoke concentration gradients and diffusion velocities. A data interpolation algorithm was used to generate a continuous three-dimensional smoke diffusion field from the sampling point data. The data was stored as a volume grid, with each grid cell measuring 1 cm³.

[0053] Step S224: Calculating smoke particle size based on the smoke diffusion data to obtain smoke particle size data; In this embodiment, the smoke particle size is calculated based on the smoke diffusion data using a laser diffraction particle size analyzer with an instrument wavelength of 632.8 nm, a measurement angle range of 5° to 175°, and a resolution of 0.1°. The particle size distribution is calculated by analyzing the scattered light intensity distribution generated when the laser beam passes through the smoke, combined with the Lambert-Beer law. The particle size calculation uses the Mie scattering theory formula, and the results show that the smoke particle size range is concentrated between 0.5 and 5 microns, with a data accuracy of ±0.05 microns. The analysis results are presented in the form of a particle size frequency distribution curve and a cumulative volume distribution, and the data timestamp is synchronized with the smoke diffusion data. The particle size data storage format is a CSV file, which contains time, position coordinates and corresponding particle size frequency.

[0054] Step S225: Calculate the spray hole diameter according to the spray hole position data to obtain spray hole diameter data; In this embodiment, the spray hole diameter is calculated according to the spray hole position data, and a numerical control microscope is used to measure the spray hole with high precision, with a magnification of 1000 times and a measurement error of less than 0.01 millimeters. Through digital image processing technology, the profile recognition of the aperture edge is performed, and the minimum inner diameter, maximum outer diameter and shape parameters (roundness, ellipticity) of the aperture are extracted. The spray hole diameter data is exported as a two-dimensional contour graph by using CAD software, and combined with the three-dimensional coordinates of the hole position, the three-dimensional positioning of the aperture is realized. The aperture data includes aperture size, shape parameters and spatial coordinates, and all data are stored in XML format for subsequent blockage analysis and calling.

[0055] Step S226: Perform blockage analysis according to the smoke particle size data and the spray hole diameter data to obtain spray hole diameter blockage data.

[0056] In this embodiment, the blockage analysis process first calculates the ratio of the smoke particle size to the spray hole diameter, and sets the threshold value 0.7 as the blockage determination standard, that is, when the smoke particle size accounts for more than 70% of the spray hole diameter, it is determined that the hole has a blockage risk. The high-precision microscope is used to collect the internal image of the spray hole, and the image segmentation technology is used to identify the area proportion of the blocking material, and the area proportion exceeding 30% is determined as serious blockage. The blockage analysis combines the spatial distribution of the spray hole to quantify the blockage probability of each spray hole, and the data is expressed in percentage form. The blockage data output includes spray hole number, blockage rate, blockage risk level (low, medium, high) and spatial position, and the data format is JSON, which is convenient for system calling and early warning processing.

[0057] Preferably, step S23 specifically comprises: Step S231: Identify the flow passage cross-sectional shape according to the spray head structure data to obtain flow passage cross-sectional shape data; In this embodiment, the flow passage cross-sectional shape is identified according to the spray head structure data. First, a three-dimensional high-precision image of the internal flow passage of the spray head is obtained by using an industrial CT scanner, and the scanning resolution is controlled to be 0.01 millimeters to ensure that the microstructure of the flow passage can be captured. The obtained three-dimensional body data is extracted by using the image segmentation algorithm, and the flow passage cross-sectional contour is extracted by using the region growing method based on threshold value, and the threshold value is set to be 80 to 120 (set according to the X-ray absorption characteristics of the material), and the noise and non-flow passage area are removed. The extracted flow passage cross-sectional shape data is measured for geometric parameters, including cross-sectional area, perimeter, minimum inner diameter, maximum inner diameter and cross-sectional shape coefficient, and the measurement accuracy is 0.01 millimeters. The cross-sectional shape data is stored in the form of two-dimensional coordinate point set, and the data format uses the DXF file compatible with CAD, which is convenient for subsequent flow passage analysis.

[0058] Step S232: Based on the flow passage cross-section shape data, flow passage expansion analysis is performed to obtain flow passage expansion data; In this embodiment, based on the flow passage cross-section shape data, the flow passage expansion analysis is performed, and the numerical geometry calculation method is used to calculate the cross-section change rate. Specifically, for each cross-section of the flow passage, the percentage change of the area of the adjacent cross-section is calculated, and the expansion threshold is set to 10%. When the area of the adjacent cross-section increases by more than the threshold, it is determined to be a flow passage expansion region. The length of the flow passage is calculated according to the three-dimensional scanning data, and the precision is controlled within 0.05 millimeters. Through the curve fitting method, the flow passage center line is extracted, and combined with the cross-section change rate, the flow passage expansion space distribution map is established. The output flow passage expansion data includes the start and end position coordinates of the flow passage expansion, the expansion percentage and the expansion length, and the data format is JSON, which is convenient for subsequent flow rate calculation and calling.

[0059] Step S233: Calculate the flow rate according to the flow passage expansion data to obtain the flow rate data; In this embodiment, the flow rate is calculated according to the flow passage expansion data, and the continuity equation in the basic equation of fluid mechanics is used for calculation. Given the flow passage inlet flow rate Q, the flow rate v = Q / A is calculated by the flow passage cross-sectional area A. The flow rate Q is measured by an ultrasonic flow meter installed at the inlet of the spray head, with a measurement range of 0.1 to 10 L / min, an accuracy of ±0.01 L / min, and a sampling frequency of 20 times per second. The flow passage cross-sectional area A uses the cross-sectional area data measured in step S231, with an accuracy of 0.01 millimeters. During the flow rate data calculation process, for the flow passage expansion region, the local cross-sectional area corresponding to the instantaneous flow rate is dynamically calculated, and the data is stored in the form of time series, with a sampling frequency of 20 Hz, a format of CSV, and containing time stamp, position coordinates and flow rate value.

[0060] Step S234: Identify the flow rate deceleration region according to the flow rate data; In this embodiment, the flow rate deceleration region is identified according to the flow rate data, and the flow rate threshold is set to 70% or less of the normal flow rate, which is defined as the flow rate deceleration region. The normal flow rate is calculated by the flow meter and the normal operating parameters, and is set to 3.0 m / s. The flow rate data is processed by a sliding window filtering algorithm, with a window size of 5 seconds to remove the influence of instantaneous noise. The spatial interval with a flow rate lower than the threshold in a continuous time period is considered as a deceleration region, and the spatial position is marked by the corresponding flow passage coordinates. The output flow rate deceleration region data includes the start and end position coordinates, the time interval and the flow rate statistical characteristics (mean, maximum and minimum), and the data format is JSON.

[0061] Step S235: Corrosion particle shedding detection is performed according to the flow rate deceleration region to obtain corrosion particle shedding data; In this embodiment, corrosion particle shedding detection is performed according to the flow rate deceleration area, and a high-sensitivity laser particle counter is used to detect the particle concentration and particle size distribution at the outlet of the flow channel. The laser wavelength is set to 650 nm, the power is 5 mW, the detection particle size range is 0.1 to 10 microns, and the measurement accuracy is ±0.1 microns. The counter sampling frequency is 10 times per second, and the collected data includes particle concentration (particle number / m³) and particle size. By mapping the position of the flow rate deceleration area, the particle release source is determined. The threshold for corrosion particle shedding is set to a particle concentration of more than 1000 particles / m³, and the particle size is concentrated in the interval of 0.5 to 5 microns. The output corrosion particle shedding data includes the particle concentration peak, the shedding timestamp, and the corresponding flow rate deceleration area coordinates, and the storage format is CSV.

[0062] Step S236: low-speed vortex area analysis based on corrosion particle shedding data, to obtain low-speed vortex area data; In this embodiment, low-speed vortex area analysis is performed based on corrosion particle shedding data, and a micro three-axis acceleration sensor array is arranged on the inner wall of the flow channel. The sensor sensitivity is ±2g, and the sampling frequency is 1 kHz. By collecting the vibration and vortex fluctuation signals generated by the local fluid in the flow channel, the typical vortex frequency (set frequency range 100 to 500 Hz) is identified using frequency spectrum analysis method. The vibration signal is processed by time domain envelope demodulation to extract the vortex characteristic parameters. Combined with the time and space information of corrosion particle shedding, the low-speed vortex area is determined by positioning the vortex frequency peak area. The output low-speed vortex area data includes vortex center coordinates, vortex intensity and frequency characteristics, and the data format is JSON.

[0063] Step S237: flow channel impurity accumulation determination according to low-speed vortex area data, to obtain flow channel impurity accumulation data.

[0064] In this embodiment, flow channel impurity accumulation determination is performed according to low-speed vortex area data, and a micro camera is used to collect impurity deposition images in the flow channel. The camera resolution is 1920x1080 pixels, and it is installed on the inner wall position of the low-speed vortex area. The image is processed by edge detection and texture analysis algorithm to identify the impurity coverage area and thickness. The impurity area threshold is set to 5% of the flow channel cross-sectional area, and the thickness threshold is 0.5 mm. The impurity accumulation amount is calculated by accumulating multiple images, and the image data and vortex intensity data are superimposed to determine the impurity accumulation degree. Finally, the flow channel impurity accumulation data is generated, including the accumulation position, area, thickness and accumulation time, and the data storage format is XML file, which is convenient for subsequent maintenance management system to call.

[0065] Preferably, step S3 specifically comprises: Step S31: fire-fighting equipment data is used to extract fire-fighting pipe network characteristics, to obtain fire-fighting pipe network data; In this embodiment, the detailed layout information of the fire protection pipe network in the building is first collected through the data interface of the fire protection Internet of Things system, including pipe diameter, pipe type, pipe section length, connection method and pipe network node coordinates. The data source is the built-in intelligent sensor and the building fire protection design database. The sensor measurement accuracy is ±0.5 mm for the pipe diameter, and the node coordinate positioning accuracy is ±5 cm. After collection, the two-dimensional layout diagram of the pipe network is converted into digital pipe network topology data, using a coordinate system specification (such as the local coordinate system of the building or the global coordinate system WGS84). The pipe network data is stored in a structured database, and the fields include pipe diameter (mm), material code (such as steel, ductile iron, PVC, etc.), pipe length (m), node ID and spatial coordinates (x, y, z). After collecting the data, the pipes are marked using a preset material code table to ensure the accuracy of subsequent attribute recognition.

[0066] Step S32: Identify the material elastic modulus based on the fire protection pipe network data; In this example, the elastic modulus of steel pipes is set to 2.0×10^11Pa, the elastic modulus of ductile iron pipes is set to 1.8×10^11Pa, and the elastic modulus of PVC pipes is set to 3.0×10^9Pa. Material elastic modulus data is in Pa and accurate to two decimal places. During elastic modulus identification, precise matching of pipe material codes ensures a one-to-one correspondence between materials and corresponding elastic moduli. This creates a table of pipe segment and elastic modulus correspondences. The data is stored in a table structure with fields including pipe segment ID, material type, and corresponding elastic modulus.

[0067] Step S33: determining shock wave velocity data according to the elastic modulus of the material; In this embodiment, the shock wave velocity is calculated based on the elastic modulus of the identified material. The shock wave velocity c is calculated using the formula c=√(E / ρ), where E is the elastic modulus (Pa) and ρ is the pipe density (kg / m^3). The density data is also determined based on the material standard database, which is 7850kg / m^3 for steel, 7200kg / m^3 for ductile iron, and 1400kg / m^3 for PVC. The calculation process uses precise floating-point operations, and the result is in meters per second (m / s), with three decimal places. To ensure accuracy, the calculation is performed independently for each pipe segment, and the results are recorded in the pipe segment shock wave velocity data table, including the pipe segment ID and the corresponding shock wave velocity. The data is exported in CSV format for subsequent simulation use.

[0068] Step S34: simulating the fire hydrant water hammer effect according to the shock wave velocity data to obtain fire hydrant water hammer effect data; In this embodiment, the water hammer effect of fire hydrant is simulated based on shock wave speed data. First, a hydraulic model of the fire pipe network is constructed, including pipe length, diameter, material elastic modulus and shock wave speed parameters. Real-time pressure sensors are used to collect flow rate, pressure and valve switching state in the fire pipe network, with a sampling frequency of 100 Hz. According to the water hammer theory, the pressure wave propagation speed and pressure change in the pipe are calculated. The finite difference method is used to calculate the instantaneous pressure wave, with a time step of 0.001 seconds and a spatial grid division based on pipe diameter and length to ensure numerical stability. The actual flow rate curve is input during simulation to calculate the instantaneous pressure peak value and pressure gradient. The simulation results generate a fire hydrant water hammer effect data file containing pressure time series, position coordinates and wave speed values in HDF5 format, which is used for subsequent fault analysis.

[0069] Step S35: Valve leakage detection according to fire hydrant water hammer effect data, obtaining valve leakage data; In this embodiment, the pressure sensor and flow sensor data at the valve are used to compare the deviation between the actual pressure waveform and the simulated pressure waveform. The leakage judgment threshold is set to a pressure wave peak value reduction of more than 5% or a flow fluctuation anomaly of more than 10%. Signal processing algorithms are used to extract pressure waveform feature points, combined with valve switching state, to identify abnormal leakage signs. The leakage detection results include leakage time, leakage location, leakage degree (percentage) and leakage duration, with a JSON data format. The leakage data is uploaded to the monitoring system in real time for material aging analysis.

[0070] Step S36: Material aging analysis according to valve leakage data, obtaining material aging data.

[0071] In this embodiment, the valve leakage duration, leakage degree and valve opening and closing frequency are used as main parameters, combined with the corresponding aging curve model of the valve material, to calculate the aging degree. The aging curve model is based on experimental data, with an aging index range of 0 to 1, where 0 represents new and 1 represents complete failure. The threshold is set to an aging index of 0.7 as a warning limit. The calculation process uses a mathematical regression model, with parameter weights of 50% for leakage duration, 30% for leakage degree and 20% for opening and closing frequency. The material aging data is stored in the form of aging index and corresponding timestamp, with a CSV format, including valve ID, aging index, time and analysis notes.

[0072] Preferably, step S35 specifically comprises: Step S351: Calculate water hammer instantaneous impact load according to fire hydrant water hammer effect data, obtaining water hammer impact load data; In this embodiment, the water hammer instantaneous impact load is calculated according to the fire hydrant water hammer effect data. First, the pressure sensor data in the fire hydrant network is collected. The pressure sensor is a high-precision piezoelectric sensor with a measurement range of 0 to 10 MPa, a response time of less than 1 ms, and a sampling frequency of 1 kHz, ensuring that the instantaneous pressure fluctuations in the water hammer event can be captured. Using the collected pressure time series data, combined with the pipe diameter and fluid density, the instantaneous impact load is calculated using the basic equations of fluid dynamics. The impact load F is calculated by the formula F = P x A, where P is the instantaneous pressure value (Pa) and A is the pipe cross-sectional area (m²). The cross-sectional area is accurately calculated according to the pipe diameter (mm), and the cross-sectional area = π x (d / 2)². The pipe diameter data is obtained from the pipe network digital model, with an accuracy of ±0.5 mm. The calculation results are output in time series with a time step of 0.001 seconds, generating a water hammer impact load data file containing time stamp, pressure value, pipe diameter, cross-sectional area, and corresponding impact load. The file format is CSV, which is convenient for subsequent processing.

[0073] Step S352: Identify the valve force-sealing surface based on the water hammer impact load data, and obtain the valve force-sealing surface data; In this embodiment, the sealing surface size and position are obtained based on the structural parameters of the valve, including the sealing surface diameter, sealing ring thickness, and material. The sealing surface size data is obtained from the technical manual provided by the valve manufacturer, with a diameter range of 50 to 300 mm and a thickness range of 2 to 10 mm. The force distribution on the sealing surface is calculated using the water hammer impact load data. According to the principles of fluid mechanics, combined with the valve body geometry, the static force balance formula is used to quantitatively analyze the force on the sealing surface. This calculation process involves decomposing the impact load to the action area of the sealing surface, using finite element meshing technology to divide the sealing surface into several units, and calculating the local force on each unit. Finally, the valve force-sealing surface data is output, including the sealing surface force distribution diagram and the maximum force point coordinates, with a data format of a two-dimensional force field matrix file.

[0074] Step S353: Perform sealing ring deformation detection based on the valve force-sealing surface data, and obtain the sealing ring deformation data; In this embodiment, a non-contact laser scanning measurement system is used to collect the surface topography of the valve seal ring. The laser scanning precision is 0.005 mm, and the scanning range covers the entire seal ring surface. The collected three-dimensional point cloud data is filtered to remove noise, and a surface reconstruction algorithm is used to generate a seal ring surface topography model. Combined with the coordinates of the maximum stress point in the stress seal surface data, the local deformation of the seal ring in this area is calculated. The deformation is calculated by the height difference of the point cloud surface, and the calculation accuracy is 0.01 mm. The deformation detection also considers the elastic modulus of the seal ring material, which is usually 5 to 20 MPa, and the data is obtained from material testing experiments. The final output seal ring deformation data includes deformation region coordinates, deformation depth, and corresponding timestamp, and the data format is XYZ coordinate points and deformation matrix.

[0075] Step S354: Calculate the edge lifting height according to the seal ring deformation data to obtain the edge lifting height data; In this embodiment, the edge lifting height is calculated according to the seal ring deformation data, and the edge lifting is defined as the protruding height of the edge region of the seal ring relative to the normal seal surface. The seal ring three-dimensional topography model generated in step S353 is used, combined with the pre-set normal seal surface reference surface, and a geometric fitting algorithm is used to determine the edge lifting boundary. The edge lifting height is obtained by measuring the vertical distance between the edge lifting boundary and the reference surface, and the measurement accuracy reaches 0.01 mm. In order to avoid the influence of abnormal data, median filtering is used to smooth the edge lifting height data, and the filter window size is set to 5 measurement points. The calculation results include the maximum height of the edge lifting, the average height, and the area percentage of the edge lifting, and the data format is stored in a structured table, including the edge lifting boundary coordinates, edge lifting height values, and statistical characteristics, for valve leakage evaluation.

[0076] Step S355: Perform valve leakage evaluation according to the edge lifting height data to obtain valve leakage data.

[0077] In this embodiment, a leakage judgment threshold is set, and the threshold is determined according to relevant industry standards, for example, the maximum height of the edge lifting exceeding 0.1 mm is considered as a mild leakage risk, and exceeding 0.3 mm is considered as a serious leakage risk. The proportion of the edge lifting area to the total seal surface area is used as an auxiliary evaluation index, and the threshold is set to 5%. The edge lifting height data is compared with the threshold, combined with the leakage history data and the valve working pressure, and the rule engine is used to judge the leakage level. The evaluation results include the leakage risk level (no leakage, mild, moderate, and serious), the leakage position, and the corresponding timestamp. The data is saved in JSON format, and the fields include valve ID, leakage level, maximum edge lifting height, edge lifting area ratio, and evaluation time, which is convenient for integration into intelligent inspection and fault warning system.

[0078] Preferably, step S36 specifically comprises: Step S361: Perform valve core wear detection according to the valve leakage data to obtain valve core wear data; In this embodiment, valve leakage data, including leakage flow rate, leakage pressure change, and leakage time series, are collected through the Internet of Things system. The leakage flow rate is collected using an ultrasonic flow meter with a measurement range of 0.1 to 1000 L / min and an accuracy of ±1%. The pressure change is obtained through pressure resistance pressure sensors installed at the inlet and outlet of the valve, with a measurement range of 0 to 10 MPa, a resolution of 0.01 MPa, and a sampling frequency of 10 Hz. Combined with the leakage data and the valve design structure parameters (valve core material, size, and shape), the wear degree is calculated using the internal fluid mechanics formula of the valve. The wear detection mainly uses the fluid shear force estimation based on the leakage flow rate and the valve opening degree. The valve core surface wear thickness is in units of microns (pm), and the threshold is set to 0.05 mm as the lower limit of detection, and more than 0.3 mm is judged as severe wear. Time series analysis is used for data processing, and the wear thickness change trend under different working conditions is calculated, and the valve core wear data file is output, including time stamp, wear thickness, valve ID, and corresponding working condition parameters, in CSV format.

[0079] Step S362: coating spalling analysis based on valve core wear data, to obtain coating spalling data; In this embodiment, coating spalling is achieved by detecting the degree of metal exposure on the valve core surface and the integrity of the coating. An electromagnetic eddy current detection instrument is used, with a frequency range of 100 kHz to 2 MHz, a probe spacing of 5 mm, a scanning speed of 10 mm / s, and a resolution of 0.01 mm. The valve core surface is divided into multiple detection areas, and the coating thickness of each area is determined by the change in eddy current signal amplitude and phase, with a measurement threshold of 0.02 mm. The coating spalling judgment standard is based on the signal amplitude being lower than the set reference value by 20%, which is considered as coating spalling. Using wear data as auxiliary parameters, a multivariate analysis model is used to correlate wear depth and coating spalling probability. Finally, the coating spalling data is output, including spalling area (mm²), spalling position coordinates, and spalling thickness, in the format of a two-dimensional coordinate matrix and a spalling thickness mapping table.

[0080] Step S363: calculate the exposed area of the valve core surface based on the coating spalling data; In this embodiment, image processing technology is used to image the coating spalling detection area at high resolution, with a resolution not less than 5 pm / pixel. Image preprocessing includes grayscale conversion, filter noise reduction, and edge detection. The coating spalling area profile is determined through binary processing, and the coating spalling area is accurately calculated using a contour area calculation algorithm. The area calculation unit is square millimeters (mm²), and the threshold is defined as areas less than 1 mm² are ignored. The total exposed area of the valve core surface is the sum of the spalling area, and the exposed area percentage is calculated in combination with the total surface area of the valve core. The data is stored in table format, including the area, position, and exposed area percentage of each spalling area, for subsequent roughness measurement.

[0081] Step S364: Measure the valve core roughness according to the valve core surface exposure area, and obtain valve core roughness data; In this embodiment, a three-dimensional laser scanning measuring instrument is used, with a scanning accuracy of 0.005 mm and a scanning range covering the entire exposed area. A digital elevation model of the valve core exposed surface is generated from the point cloud data. The arithmetic average roughness Ra and root mean square roughness Rq of the exposed area are calculated using surface roughness analysis software. The calculation is based on the ISO4287 standard, with a calculation length of 10 mm and a sampling interval of 0.01 mm. The roughness parameters are used to describe the degree of surface fluctuation, with a precision of 0.001 μm. The measurement results are stored in the form of roughness curves and statistical data, including timestamp, area ID, Ra value, Rq value, for material aging assessment.

[0082] Step S365: Material aging assessment based on valve core roughness data, and obtain material aging data.

[0083] In this embodiment, material aging assessment is based on valve core roughness data. An aging assessment model is established, with input parameters including roughness indicators Ra and Rq, valve core working pressure, fluid medium type and temperature. The threshold is set as Ra exceeding 1.5 μm is considered as significant material aging, and Rq exceeding 2.0 μm is considered as severe aging. An empirical formula is used to convert roughness to aging index, with a range of 0~1, and the calculation formula is aging index=(Ra / 3)+(Rq / 4). If the aging index exceeds 0.6, a warning is triggered. The model calculation results are combined with the valve core usage time to output the aging degree and remaining life prediction. The data is stored in CSV format, including valve core ID, aging index, remaining life prediction, analysis time and evaluation notes, for intelligent inspection and fault warning system to call.

[0084] Preferably, step S4 specifically comprises: Step S41: Mark the metal surface corrosion area according to the metal surface corrosion data; In this embodiment, the corrosion depth, corrosion area and corrosion type data collected by the aforementioned metal surface corrosion detection module are used, with a data format of two-dimensional coordinate point set and corresponding corrosion depth (unit: millimeter, precision 0.01 mm). A threshold filtering method is used, with a corrosion depth threshold of 0.05 mm. Any point set with a corrosion depth exceeding this threshold is considered as a valid corrosion area. A spatial clustering algorithm (based on Euclidean distance, with a clustering radius of 5 mm) is used to divide the corrosion point set into multiple corrosion clusters. The clustering results are used to fit the polygon boundary of the corrosion clusters, forming the geometric boundary data of the corrosion area. The center coordinates, area (unit: square millimeter), and average corrosion depth of each corrosion area are recorded. The final output corrosion area data includes region ID, boundary coordinate sequence, area, average depth, maximum depth, and is stored in JSON format.

[0085] Step S42: Calibrate material aging region according to material aging data; In this embodiment, the material aging data includes the aging index, aging degree grade and spatial distribution information of the valve core or pipe network components. The data is divided into two categories of aging and non-aging by using an aging index threshold of 0.4. The aging index data corresponds to the spatial coordinates of the equipment, and the points with aging index exceeding the threshold are marked as aging points. A spatial neighborhood search algorithm (neighborhood radius of 10 mm) is used to gather the aging points to form an aging region. The boundary of the aging region is extracted to generate the polygon boundary of the aging region, and the area and the maximum aging index of the region are calculated. The aging region data structure includes the region identifier, the spatial boundary point coordinates, the region area, the maximum and average aging index, and the data is stored in XML format for subsequent analysis and calling.

[0086] Step S43: Perform intersection operation on the fault hot spot region data according to the metal surface corrosion region and the material aging region; In this embodiment, the boundary polygon data of the two types of regions is imported into a spatial relationship calculation module, and a computational geometry algorithm (such as Sutherland-Hodgman polygon clipping algorithm) is used to calculate the intersection of the two regions. The intersection polygon represents the fault hot spot region. The intersection area, position and shape characteristics of the intersection region are calculated. If the intersection area is less than 10 mm², it is considered as an invalid hot spot region and is removed. The result is output in the form of a table containing the hot spot region ID, boundary coordinate sequence, intersection area and related metadata. During the entire calculation process, the coordinate precision of the boundary points is not less than 0.01 mm, and the operation error is controlled within 1%.

[0087] Step S44: Evaluate the fault severity according to the fault hot spot region to obtain the fault severity data; In this embodiment, the fault severity is evaluated according to the fault hot spot region. According to the area, average corrosion depth and average aging index of the hot spot region, a severity index is constructed. The specific calculation formula is: severity = 0.5 × (average corrosion depth / maximum allowed corrosion depth 0.5 mm) + 0.5 × (average aging index / 1.0). The severity range is limited to 0 to 1, and the larger the value, the more serious the fault. This calculation is performed independently for each hot spot region, and the input data includes the hot spot region area (mm²), corrosion depth data (mm) and aging index. Real-time measurement data is called through the database to ensure that the calculation is based on the latest detection information. The severity result is saved in the form of a floating point number, with the hot spot region ID, timestamp and calculation parameters, and is output as a CSV format file.

[0088] Step S45: Risk score based on fault severity data, get fault risk data; according to the risk priority ranking data of fault risk data; In this embodiment, the numerical range of fault severity is limited between 0 and 1, and the value 0 represents no fault and the value 1 represents the most serious fault. The risk score is calculated by a linear mapping function, and the specific mapping formula is: risk score = severity x 100, the calculation result is an integer score between 0 and 100, and the higher the score represents the greater the risk of the fault area. The scoring standard is strictly set according to the national fire equipment safety specification and industry technical standard to ensure the consistency and comparability of the scoring range and risk level. After the risk score of all fault hot spot areas is calculated, all scoring results are input into the sorting module. The sorting module uses the QuickSort algorithm, which has an average time complexity of O(nlogn) and is suitable for processing fault hot spot data of more than one million, ensuring the efficiency and real-time performance of the sorting. When sorting, the risk score is arranged in descending order, and the fault area with high priority is arranged in the front row. The sorting result generates a list containing the unique identification of the fault area (area ID), the corresponding risk score and the ranking. This list is stored in a relational database, and the table structure design includes the fields: area ID (string type), risk score (integer type), priority ranking (integer type) and timestamp (record sorting time). The database index is established based on the risk score field to support fast retrieval when called later. The risk priority list provides direct data basis for subsequent dynamic adjustment of inspection path and early warning decision, ensures that the inspection resources are allocated to high-risk areas first, and realizes scientific and efficient equipment maintenance management.

[0089] Step S46: Dynamic inspection path adjustment based on risk priority for intelligent inspection path data, get dynamic inspection adjustment path data.

[0090] In this embodiment, the path planning algorithm is based on node connection relationship and node weight (assigned according to device importance, range 1 to 10). The heuristic shortest path algorithm is used to solve the Traveling Salesman Problem (TSP) to calculate the shortest path covering all inspection nodes. The path length unit is meter, and the total inspection time is calculated based on the path length and the preset inspection speed (1.2 meters / second). The algorithm sets the upper limit of path time to 8 hours, and if it exceeds, the path is divided into multiple inspection sections. The final generated inspection path data is stored in the form of node sequence and path distance matrix, and the format supports XML and JSON, which is convenient for downstream scheduling system to call.

[0091] Therefore, the embodiments should be regarded, at any point, as being exemplary and not limiting, the scope of the application being defined by the appended claims and not by the above description, and all changes which come within the meaning and range of equivalency of the claims are therefore intended to be embraced therein.

[0092] The foregoing is considered as illustrative only of the principles of the application. Numerous modifications and changes will readily occur to those skilled in the art, and the generic principles defined herein can be applied to other embodiments without departing from the spirit or scope of the application. Therefore, the application is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A fire-fighting equipment intelligent inspection and fault warning method based on the Internet of Things, characterized in that: The following steps are involved: Step S1: Obtain firefighting equipment data; Perform spatial topological structure analysis based on firefighting equipment data to obtain spatial topological structure data; Generate an intelligent inspection path according to the spatial topological structure data to obtain intelligent inspection path data; Step S2: extracting sprinkler head structural features based on the firefighting equipment data to obtain sprinkler head structural data; performing blockage detection based on the sprinkler head structural data to obtain sprinkler head blockage data; performing metal surface corrosion detection based on the sprinkler head blockage data to obtain metal surface corrosion data; Step S3: Performing a fire hydrant water hammer effect simulation based on the fire-fighting equipment data to obtain fire hydrant water hammer effect data; performing a valve leakage test based on the fire hydrant water hammer effect data to obtain valve leakage data; performing a material aging analysis based on the valve leakage data to obtain material aging data; Step S4: identifying the fault hotspot area based on the metal surface corrosion data and the material aging data to obtain the fault hotspot area data; Risk priorities are sorted according to fault hotspot areas to obtain risk priority data; intelligent inspection path data are dynamically adjusted according to risk priorities to obtain dynamic inspection adjustment path data.

2. The method for intelligent inspection and fault warning of firefighting equipment based on the Internet of Things according to claim 1 is characterized in that: Step S1 is specifically as follows: Step S11: Obtain firefighting equipment data; Step S12: extracting spatial coordinate data based on firefighting equipment data; Step S13: Calculating a spatial distance matrix based on the spatial coordinate data to obtain spatial distance matrix data; identifying spatial connection relationships based on the spatial distance matrix data to obtain connection relationship data; Step S14: drawing a spatial topology map based on the connection relationship data; Node attribute annotation is performed based on the spatial topology map to obtain node attribute data; spatial topology structure analysis is performed based on the spatial topology map and node attribute data to obtain spatial topology structure data; Step S15: Generate an intelligent inspection path according to the spatial topology structure data to obtain intelligent inspection path data.

3. The method for intelligent inspection and fault warning of firefighting equipment based on the Internet of Things according to claim 2 is characterized in that: Step S15 is specifically as follows: Step S151: calibrating inspection nodes according to spatial topological structure data to obtain inspection node data; Step S152: Identify node connectivity relationships based on the inspection node data to obtain node connectivity relationship data; Step S153: performing path reachability analysis based on the node connectivity data to obtain a set of feasible inspection paths; Step S154: Perform shortest path intelligent analysis based on the inspection feasible path set to obtain intelligent inspection path data.

4. The method for intelligent inspection and fault warning of firefighting equipment based on the Internet of Things according to claim 1 is characterized in that: Step S2 is specifically as follows: Step S21: extracting sprinkler head structural features based on firefighting equipment data to obtain sprinkler head structural data; Step S22: performing a spray aperture blockage detection based on the spray head structure data to obtain spray aperture blockage data; Step S23: performing an internal flow channel impurity accumulation detection based on the shower head structure data to obtain flow channel impurity accumulation data; Step S24: Integrate the spray aperture blockage data and the flow channel impurity accumulation data to obtain the spray head blockage data; Step S25: performing metal surface corrosion detection based on the shower head blockage data to obtain metal surface corrosion data.

5. The method for intelligent inspection and fault warning of firefighting equipment based on the Internet of Things according to claim 4 is characterized in that: Step S22 is specifically as follows: Step S221: Identify the spray hole location according to the sprinkler head structure data to obtain the spray hole location data; Step S222: performing hot smoke spray simulation based on the spray hole position data to obtain hot smoke spray data; Step S223: performing smoke diffusion analysis based on the hot smoke spray data to obtain smoke diffusion data; Step S224: Calculating smoke particle size based on the smoke diffusion data to obtain smoke particle size data; Step S225: Calculate the spray hole diameter according to the spray hole position data to obtain the spray hole diameter data; Step S226: performing blockage analysis based on the smoke particle size data and the spray aperture data to obtain spray aperture blockage data.

6. The method for intelligent inspection and fault warning of firefighting equipment based on the Internet of Things according to claim 4 is characterized in that: Step S23 is specifically as follows: Step S231: Identify the flow channel cross-sectional shape according to the shower head structure data to obtain flow channel cross-sectional shape data; Step S232: performing flow channel expansion analysis based on the flow channel cross-sectional morphology data to obtain flow channel expansion data; Step S233: Calculating the flow velocity based on the flow channel expansion data to obtain flow velocity data; Step S234: Identifying a flow velocity slowdown area based on the flow velocity data; Step S235: performing rust particle shedding detection according to the flow rate slowdown area to obtain rust particle shedding data; Step S236: performing low-speed vortex region analysis based on the rust particle shedding data to obtain low-speed vortex region data; Step S237: Determine the flow channel impurity accumulation based on the low-speed vortex area data to obtain the flow channel impurity accumulation data.

7. The method for intelligent inspection and fault warning of firefighting equipment based on the Internet of Things according to claim 1 is characterized in that: Step S3 is specifically as follows: Step S31: extracting fire pipe network features based on fire equipment data to obtain fire pipe network data; Step S32: Identify the material elastic modulus based on the fire protection pipe network data; Step S33: determining shock wave velocity data according to the elastic modulus of the material; Step S34: simulating the fire hydrant water hammer effect according to the shock wave velocity data to obtain fire hydrant water hammer effect data; Step S35: performing valve leakage detection based on the fire hydrant water hammer effect data to obtain valve leakage data; Step S36: Perform material aging analysis based on the valve leakage data to obtain material aging data.

8. The method for intelligent inspection and fault warning of firefighting equipment based on the Internet of Things according to claim 7 is characterized in that: Step S35 is specifically as follows: Step S351: Calculate the instantaneous impact load of water hammer according to the fire hydrant water hammer effect data to obtain water hammer impact load data; Step S352: Identify the valve's stress-bearing sealing surface based on the water hammer impact load data to obtain valve's stress-bearing sealing surface data; Step S353: performing sealing ring deformation detection based on the valve stress sealing surface data to obtain sealing ring deformation data; Step S354: Calculating the warping height according to the deformation data of the sealing ring to obtain warping height data; Step S355: performing valve leakage assessment based on the warping edge height data to obtain valve leakage data.

9. The method for intelligent inspection and fault warning of firefighting equipment based on the Internet of Things according to claim 7 is characterized in that: Step S36 is specifically as follows: Step S361: performing valve core wear detection according to valve leakage data to obtain valve core wear data; Step S362: performing coating peeling analysis based on the valve core wear data to obtain coating peeling data; Step S363: Calculating the exposed area of ​​the valve core surface based on the coating peeling data; Step S364: measuring the roughness of the valve core according to the exposed area of ​​the valve core surface to obtain valve core roughness data; Step S365: Perform material aging assessment based on the valve core roughness data to obtain material aging data.

10. The method for intelligent inspection and fault warning of firefighting equipment based on the Internet of Things according to claim 1 is characterized in that: Step S4 is specifically as follows: Step S41: calibrating the metal surface corrosion area according to the metal surface corrosion data; Step S42: calibrating the material aging area according to the material aging data; Step S43: performing an intersection operation of the fault hotspot area according to the metal surface corrosion area and the material aging area to obtain the fault hotspot area data; Step S44: Evaluate the fault severity according to the fault hotspot area to obtain fault severity data; Step S45: performing risk scoring based on the fault severity data to obtain fault risk data; Sort risk priorities according to fault risk data to obtain risk priority data; Step S46: Dynamically adjust the intelligent inspection path data according to the risk priority to obtain dynamic inspection adjustment path data.

Images