A fault identification method for complex terrain-based radio and television wireless transmitting station

Abstract

This invention discloses a fault identification method for radio and television transmission stations in complex terrain, belonging to the field of fault identification technology for wireless transmission stations. It solves the problem of low accuracy in existing fault identification models under complex conditions such as terrain factors. This invention systematically collects environmental baseline data such as terrain, meteorology, geology, and historical fault records, and deploys a sensor network covering all elements of the infrastructure, fundamentally solving the problem of broken causal chains caused by the "environmental stripping" of traditional models. Based on multi-source data streams and calculations of fault transmission relationships and internal moisture risk using soil and structural parameters, a multi-physics coupled model of "rainfall-settlement-tilt-equipment anomaly" is constructed, overcoming the limitations of isolated monitoring and enabling timely quantification of hidden structural damage. Furthermore, this invention achieves adaptive adjustment of the early warning threshold according to environmental disturbances through dynamic weighting of terrain modulation factors and climate severity factors.

Description

Technical Field

[0001] This invention relates to the field of wireless transmission station fault identification technology, and in particular to a fault identification method for broadcast television wireless transmission stations based on complex terrain. Background Technology

[0002] Some radio and television transmission stations are deployed in high-altitude mountainous areas above 1,000 meters, with drastically undulating terrain and fragmented landforms, and are constantly exposed to a hot, rainy, and humid monsoon climate. The infrastructure of these stations is chronically exposed to the combined effects of complex geology and harsh weather, resulting in a significantly higher risk of systemic damage compared to sites in plains. The tower structures, due to the elevated terrain, bear stronger wind pressure, and the turbulence and topographic acceleration effects cause frequent abrupt changes in wind load, placing tower connection nodes in a state of fatigue cycle for extended periods. In high-humidity environments, the antenna and feeder systems are highly susceptible to condensation and electrochemical corrosion at connectors and feeder interfaces, leading to gradual deterioration of insulation performance. Power lines are laid along steep mountain ridges, and heavy rainfall can easily cause soil erosion in the tower foundations, resulting in tower tilting or even collapse. Conductive water films form on the surface of insulators under alternating hot and humid conditions, increasing leakage current and inducing flashover faults. Under continuous rainfall, the shear strength of the soil around the walls and slopes has decreased, leading to significant risks of landslides and collapses. Access roads to the station are mostly winding mountain paths, where rainwater erosion has caused roadbed hollowing and pavement cracking, limiting accessibility and maintenance. These hidden damages to infrastructure are characterized by both cumulative and sudden occurrences. Traditional manual inspection cycles are insufficient to keep up with the rapid pace of environmental changes, often resulting in minor defects developing into systemic failures before being discovered, posing a major risk source for broadcast interruptions.

[0003] Current mainstream fault identification models generally focus on monitoring the operating parameters of core equipment such as transmitters and exciters in the station's equipment room, including electrical indicators such as transmission power, reflection power, operating temperature, voltage, and current. Anomaly detection is achieved by setting fixed thresholds or training machine learning models based on historical data. This "equipment-centric" analytical paradigm may be effective in plains areas, but its accuracy systematically decreases in complex mountainous terrain. The fundamental reason is that such models simplify the equipment's operating environment to an ideal state, failing to incorporate external disturbances caused by terrain factors into the feature system. Its limitations are manifested in three aspects: First, relying solely on internal equipment data leads to a break in the causal chain. The model can only capture superficial phenomena such as "increased VSWR" and "power decrease," failing to connect them to the root causes such as tower subsidence due to continuous heavy rain, antenna shift, or feeder water seepage. This easily leads to misjudgments as equipment aging and the scheduling of ineffective maintenance. Second, complex terrain causes missing sensor data and transmission delays. The model receives degraded samples, resulting in a significant deviation between the training distribution and the actual scene, drastically reducing its generalization ability. Third, the lack of a multi-physics coupling mechanism prevents the construction of a quantitative transmission relationship of "rainfall → seepage → structural deformation → equipment anomaly," resulting in insufficient sensitivity to slowly evolving structural hazards and a significantly increased risk of missed detections. This isolated analysis model, which focuses solely on the equipment itself, is fundamentally unable to overcome the "causal black box" dilemma under the strong environmental disturbances of complex terrain.

[0004] Therefore, a fault identification method for broadcast television wireless transmission stations in complex terrain is needed. Summary of the Invention

[0005] To address the issue of low accuracy in existing fault identification models under complex conditions such as terrain, this invention provides a fault identification method for broadcast television wireless transmission stations in complex terrain. This method can construct a three-in-one fault identification system of "data acquisition-linkage-early warning" for high-altitude, high-temperature, rainy, and humid environments, achieving high-precision fault identification through multi-physics data coupling analysis. The specific technical solution is as follows: A fault identification method for broadcast television wireless transmission stations in complex terrain includes the following steps: Collect station topographic maps, meteorological data from recent years, geological survey reports, and historical fault logs; and deploy sensors. Acquire sensor time-series data, digital elevation model (DEM), and network topology to form a synchronous multi-source data stream output; The fault transmission relationship is calculated based on multi-source data streams, soil physical parameters and tower structural parameters, and the risk of internal moisture is assessed. Based on this, a coupling feature vector is output. Based on coupled feature vectors, real-time meteorological data, and historical baseline data, the overall state is calculated, and a terrain-adaptive comprehensive health index is derived, as follows: In the formula, This indicates the terrain-adaptive comprehensive health index; Indicates the first i Real-time values ​​of each monitoring parameter; Indicates the first i Fault threshold for each parameter; Indicates the first i The historical maximum value of each parameter; Indicates the first i Each parameter has a basic weight; Indicates the terrain modulation factor; Indicates the climate severity modulation factor; Different warning levels are derived based on the scores of the terrain-adaptive comprehensive health index, and corresponding warning signals are then output.

[0006] Preferably, the formula for calculating the terrain modulation factor is as follows: In the formula, For reference tower height; The average slope angle of the station; The slope is for flat ground. The formula for calculating the climate severity modulation factor is as follows: In the formula, This represents the rainfall for the day. This represents the average daily rainfall during the local rainy season. Current humidity; This represents the average humidity.

[0007] Preferably, the deployment of sensors includes setting the monitoring point density based on terrain complexity. The formula for adaptive monitoring point density based on terrain complexity is as follows: In the formula, This indicates the sensor deployment density in the target area; This represents the baseline value for standard deployment density in plains areas; Indicates the terrain influence coefficient; Indicates the elevation of the highest point in the station area; Indicates the elevation of the lowest point in the station area; Indicates the longest diagonal distance of the station area; Indicates geological risk weight; Indicates the historical frequency of geological disasters; A function representing the severity of climate; This indicates the average annual rainfall over many years at the station. This indicates the multi-year average relative humidity of the station.

[0008] Preferably, the deployment of sensors also includes setting the number of backup sensors for critical locations, wherein the formula for calculating the number of backup sensors for critical locations is as follows: In the formula, Indicates the number of backup sensors that need to be configured; Indicates the mean time to repair (MTBT). Indicates the maximum allowable monitoring blind zone time; This indicates the required monitoring reliability probability of the system; This represents the terrain accessibility coefficient.

[0009] Preferably, the coupled feature vector F coupled At least the rainfall-tilt coupling coefficient should be included, and the calculation formula is as follows: In the formula, Indicates the rainfall-tilt coupling coefficient; This indicates the change in the tower's tilt angle; This indicates the change in effective rainfall. Indicates the soil type conversion coefficient; Indicates the area of ​​the tower base platform; Indicates rainwater density; Represents gravitational acceleration; Indicates the foundation drainage capacity coefficient; Indicates the equivalent compression modulus of soil; The tower's full height; This indicates the width of the foundation slab.

[0010] Preferably, the coupled feature vector F coupled At least including.

[0011] Preferably, the coupled feature vector F coupled It should include at least the moisture penetration depth, calculated using the following formula: In the formula, Indicates the depth of moisture penetration; This represents the diffusion coefficient of water vapor in the sealing material; Indicates the duration of continuous exposure to a high temperature and high humidity environment; Equipment enclosure sealing rating coefficient; Indicates the temperature difference between day and night; This represents the coefficient for increasing the gap size due to thermal expansion and contraction.

[0012] Preferably, the coupled feature vector F coupled It should include at least the fatigue damage accumulation index, calculated using the following formula: In the formula, Indicates the cumulative fatigue damage index; This indicates the number of vibration cycles at wind speed level i. Indicates the first i The median wind speed of each wind speed range; Indicates the terrain-corrected SN curve index; Indicates the fatigue strength coefficient of the material; Indicates the weight of turbulence influence; Indicates the standard deviation of wind speed; This indicates the average wind speed.

[0013] A computer-readable storage medium includes a stored program, wherein, when the program is executed, it controls the device on which the computer-readable storage medium is located to perform the fault identification method for a broadcast television wireless transmission station based on complex terrain as described above.

[0014] A processor for running a program, wherein the program, when running, executes the fault identification method for broadcast television wireless transmission stations based on complex terrain as described above.

[0015] Compared with the prior art, the beneficial effects of the present invention are as follows: This invention fundamentally solves the problem of causal chain breakage caused by the "environmental stripping" of traditional models by systematically collecting environmental baseline data such as topography, meteorology, geology, and historical fault records, and deploying a sensor network covering all elements of infrastructure. Based on multi-source data streams and soil and structural parameters, it calculates fault transmission relationships and internal moisture risk, and constructs a multi-physics coupling model of "rainfall-settlement-tilt-equipment anomaly," breaking through the limitations of isolated monitoring and enabling timely quantification of hidden structural damage. The original Terrain Adaptive Comprehensive Health Index (TA-CHI) dynamically weights terrain modulation factors and climate severity factors, realizing adaptive adjustment of warning thresholds with environmental disturbances. This significantly improves the identification accuracy and early warning lead time under complex terrain, effectively overcoming the shortcomings of fixed threshold models in terms of data degradation and insufficient scene generalization ability, forming a complete closed loop from environmental perception → causal modeling → dynamic evaluation → graded early warning. Attached Figure Description

[0016] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the accompanying drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. In all the drawings, similar elements or parts are generally identified by similar reference numerals. In the drawings, the elements or parts are not necessarily drawn to scale.

[0017] Figure 1This is a flowchart of the method of the present invention. Detailed Implementation The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0018] It should be understood that, when used in this specification and the appended claims, the terms "comprising" and "including" indicate the presence of the described features, integrals, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or collections thereof.

[0019] It should also be understood that the terminology used in this specification is for the purpose of describing particular embodiments only and is not intended to limit the invention. As used in this specification and the appended claims, the singular forms “a,” “an,” and “the” are intended to include the plural forms unless the context clearly indicates otherwise.

[0020] It should also be further understood that the term "and / or" as used in this specification and the appended claims refers to any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.

[0021] In one embodiment of the present invention, a fault identification method for broadcast television wireless transmission stations based on complex terrain is provided, such as... Figure 1 As shown, it includes the following steps: Step 1: Collect topographic maps of the stations, meteorological data from the past few years (e.g., 5 years), geological survey reports, and historical fault logs; and deploy the sensors.

[0022] For example, meteorological data for the past n years should include at least rainfall, humidity, and wind speed, and geological survey reports should include at least soil type and bearing capacity. For example, the sensor deployment at the monitoring station is obtained using a terrain complexity adaptive monitoring point density formula, including the number, type, and installation coordinates of sensors in each monitoring area. Furthermore, backup sensors are configured for key nodes (such as tower bases and antenna feeder interfaces), with the number of backup sensors configured as follows: Nbackup ; The formula for adaptive monitoring point density based on terrain complexity is as follows: In the formula, Indicates the sensor deployment density in the target area (number per 100m²). This represents the baseline value for standard deployment density in plains areas (usually taken as 0.5 units / 100m²). This represents the topographic influence coefficient (dimensionless), ranging from 1.5 to 3.0. A larger value indicates a more complex topography. This indicates the elevation (m) of the highest point in the station area. This indicates the lowest point elevation (m) in the station area. Indicates the longest diagonal distance of the station area; The geological risk weight (dimensionless) is determined based on the soil and rock type: 0.8 for stable bedrock, 1.5 for weathered soil, and 2.5 for landslide hazard zones. Indicates the historical frequency of geological disasters (times / year); The function representing the severity of the climate is , fclimate =1+0.02·( R annual -1200)+0.01·( RH avg -75%); This indicates the average annual rainfall over many years at the station. This indicates the multi-year average relative humidity of the station.

[0023] The specific formula for calculating the number of backup sensors for critical components is as follows: In the formula, Indicates the number of backup sensors that need to be configured (rounded up); This represents the mean time to repair faults (h), which is typically taken as 48~72h for complex mountainous terrain. This indicates the maximum allowable monitoring blind zone time (h), which is 2~6h depending on the fault development rate; This represents the required monitoring reliability probability of the system (usually taken as 0.95~0.99). The topographic accessibility coefficient (dimensionless) is determined by road conditions: 1.0 for paved roads, 1.5 for dirt roads, and 3.0 for roads that are inaccessible.

[0024] By employing the above methods, we can reduce the cost waste caused by over-deployment while ensuring monitoring coverage. Furthermore, the establishment of redundant backups for key monitoring points reduces the risk to data terminals.

[0025] Step Two: Acquire raw sensor time-series data (sensors deployed in Step One), Digital Elevation Model (DEM), and network topology. Evaluate the reliability of each data point using the terrain occlusion data reliability attenuation formula, and unify the time series using the multi-path delay synchronization correction formula. Finally, output a synchronized data stream with reliability labels. Fi , Cdata ,tsync} and low-reliability data tagging queue (for subsequent cleaning or retransmission requests).

[0026] For example, the formula for reducing the reliability of terrain occlusion data is as follows: In the formula, Indicates the data credibility score (0~1); This indicates the reliability of the line-of-sight transmission reference (taken as 0.95). This represents the terrain shading attenuation coefficient (dimensionless), with values ​​based on obstacle type: dense forest 0.3, ridge 0.5, canyon 0.7; This represents the actual transmission distance (m) from the sensor to the aggregation node. Indicates the reference transmission distance (taken as 1000m); This indicates the angle (°) between the transmission path and the horizontal plane, reflecting the difficulty of transmission while climbing a slope; The value represents the elevation angle influence index (ranging from 2 to 4). The larger the value, the more significant the impact of the elevation angle on the signal.

[0027] For example, the multipath delay synchronization correction formula is as follows: In the formula, This represents the total time deviation (s) that needs to be compensated. This indicates the sensor clock drift deviation (s), which is typically <0.001s after synchronization via BeiDou time synchronization. This represents the signal transmission path length (m), and the actual path is calculated taking into account terrain undulations. The speed of signal propagation (m / s) is represented by 3 × 10⁻⁶ m / s for wireless signals. 8 m / s, wired connection takes 2×10 8 m / s; This represents the relay node queuing delay (s), which is related to network load. Indicates the altitude (m) of the relay equipment; This indicates the tower's altitude (m), used to correct for transmission delays caused by altitude differences.

[0028] By using the above methods, data synchronization accuracy can reach ±0.1 seconds, and unreliable data can be automatically downweighted or eliminated to reduce the false alarm rate.

[0029] Step 3: Obtain the synchronized multi-source data stream output from Step 2, and acquire soil physical parameters and tower structural parameters; then calculate the fault propagation relationship, assess the internal moisture risk, and calculate structural damage, thereby outputting the coupling feature vector F. coupled =[ Krain - tilt ,dpenetration , Dfatigue [,...] and the fault propagation probability matrix of each subsystem.

[0030] Among them, soil physical parameters include at least the permeability coefficient. kdrain Compression modulus Esoil The structural parameters of the iron tower include at least the tower height. Htower Basic dimensions.

[0031] For example, the fault transmission relationship is calculated using the formula for the linkage coefficient of rainfall-tower foundation settlement-tilt, as follows: In the formula, This represents the rainfall-tilt coupling coefficient (° / mm), reflecting the change in tower tilt caused by a unit effective rainfall. This indicates the change in the tower's tilt angle (°). This indicates the change in effective rainfall. The soil type conversion coefficient (dimensionless) is represented by 0.8~1.2 for clay, 0.3~0.6 for sand, and 0.1~0.2 for rock. This indicates the area of ​​the tower base platform (m²). This represents the density of rainwater (taken as 1.0 × 10³ kg / m³). Represents gravitational acceleration; This represents the foundation drainage capacity coefficient (m / s), which is related to the soil permeability coefficient and drainage slope. It represents the equivalent compressibility modulus of soil (Pa), reflecting the stiffness of the foundation. Total height of the tower (m); This indicates the width of the foundation slab (m).

[0032] For example, the risk of internal moisture ingress can be assessed using the temperature and humidity penetration depth formula for the device casing, as follows: In the formula, The depth of moisture penetration (mm) indicates that if it exceeds the thickness of the sealing ring, it is considered a high-risk condition. This represents the diffusion coefficient of water vapor in the sealing material (mm² / h), with silicone taking 0.02~0.05 and rubber taking 0.01~0.03. Indicates the duration of continuous exposure to a high temperature and high humidity environment (h); The sealing rating coefficient (dimensionless) for the equipment enclosure is 1.0 for IP65, 0.5 for IP54, and 0.2 for IP40. This represents the temperature difference between day and night (°C), and in high-altitude areas it is usually taken as 15~25°C; This represents the coefficient for increasing the gap due to thermal expansion and contraction (taken as 0.05~0.1℃⁻¹).

[0033] For example, the structural damage is calculated using the cumulative formula of wind field turbulence-tower vibration fatigue, as follows: In the formula, The fatigue damage cumulative index (dimensionless) indicates that fatigue failure has occurred; ≥1 indicates that fatigue failure has occurred. This indicates the number of vibration cycles at wind speed level i. Indicates the first i Median wind speed (m / s) for each wind speed range. This represents the dimensionless exponent for terrain-corrected SN curves, with 3.0 for plains, 3.5 for mountain peaks, and 4.0 for mountain passes. This represents the fatigue strength coefficient of the material, which is related to the type of tower material. The turbulence influence weight (dimensionless) is taken as 0.2~0.4; It represents the standard deviation of wind speed (m / s) and reflects the intensity of turbulence; This represents the average wind speed (m / s).

[0034] Step 4: Obtain the coupled feature vector, real-time meteorological data, and historical baseline data, then calculate the overall state and correct the triggering conditions, and predict secondary faults.

[0035] Historical baseline data includes Fi , threshold , R 6 h , max .

[0036] For example, the overall state is calculated using the terrain-adaptive integrated health index formula, as follows: In the formula, This represents the terrain-adaptive comprehensive health index (0~100). Indicates the first i Real-time values ​​of each monitoring parameter; Indicates the first i Fault threshold for each parameter; Indicates the first i The historical maximum value of each parameter; Indicates the first i The basic weights of each parameter (after normalization, ∑) wi =1); Represents the terrain modulation factor (dimensionless). This represents the climate severity modulation factor (dimensionless).

[0037] The formula for calculating the terrain modulation factor is as follows: In the formula, For reference tower height (taken as 50m); The average slope angle of the station (°); The slope is for flat ground (take 5°).

[0038] The formula for calculating the climate severity modulation factor is as follows: In the formula, This represents the rainfall for the day. This represents the average daily rainfall during the local rainy season. Current humidity; This represents the average humidity.

[0039] For example, the triggering conditions are corrected using a dynamic early warning threshold adjustment formula, as follows: In the formula, Dynamic tilt warning threshold (°); This represents the static baseline threshold (°), typically taken as 0.5% of the tower height; The seasonal impact amplitude (dimensionless) is 0.3 for the rainy season and -0.1 for the dry season. Indicates the day of the year (1-365); This represents the real-time rainfall attenuation coefficient (dimensionless), ranging from 0.2 to 0.4. This indicates the cumulative rainfall (mm) over the past 6 hours. This represents the maximum rainfall (mm) in the same period in history over a 6-hour period.

[0040] For example, secondary faults can be predicted using a fault propagation path probability formula, as follows: In the formula, Indicates a fault originating from the subsystem i propagation to subsystems j The probability of; This represents the probability of direct causation, derived from a knowledge graph knowledge base. Indicates the number of intermediate nodes in the propagation path; Indicates the first k The terrain barrier coefficient (dimensionless) of the route segment ranges from 0.1 to 0.8. This represents the physical distance (m) from the fault source to the intermediate node. The characteristic distance of fault propagation (m) is represented, ranging from 50 to 200 m; Indicates the elevation difference (m) between the fault source and the intermediate node; The elevation attenuation scale (m) is represented, ranging from 100 to 300m.

[0041] Through the above process, the system can output a terrain-adaptive comprehensive health index, warning levels (blue / yellow / orange / red) and dynamic thresholds, and a fault propagation path probability map (used to predict cascading failures). This improves the accuracy of warnings and enables early warnings (such as slow-moving subsidence faults), and automatically identifies high-risk cascading failure paths to prevent systemic collapse.

[0042] Step 5: Obtain the TA-CHI index and warning level from Step 4, and combine them with network status data to maintain personnel location information (which can be obtained through a mobile app). Based on this, determine the localized decision-making scope and output an alarm signal.

[0043] For example, the localization decision scope can be determined using a dynamic allocation formula for edge computing resources: In the formula, This represents the computational complexity (FLOPS) of edge node allocation. This represents the total computing requirements per station (FLOPS). This indicates the maximum allowable response delay (in seconds), typically 5 to 10 seconds. This represents the expected fault repair time (s), and for complex terrain, it is taken as 172800~259200s (48~72h). This represents the relay node delay penalty coefficient (ranging from 0.1 to 0.3). This indicates the number of relay hops (dimensionless).

[0044] Through the above process, real-time alarm signals, local control commands (such as starting backup power and switching antenna feeders) can be output at the edge, along with compressed feature vector z (uploaded to the cloud). Furthermore, the cloud can output multi-site correlation analysis results, maintenance work order optimized paths (integrated with GIS), and model parameter update packages ΔW. Based on this, a second-level response can be achieved at the edge, meeting the needs of emergency fault handling, and global optimization can be performed in the cloud, improving maintenance path efficiency.

[0045] Step Six: Obtain all warning times output in Step Four, along with maintenance feedback data and terrain-climate labels. Then, calculate the actual performance and trigger online incremental learning for false alarm / missed alarm samples to further update the causal chain probabilities in the knowledge graph. Pdirect and terrain barrier coefficient δterrain .

[0046] The maintenance feedback data includes the actual causes of the faults reported by maintenance personnel through the APP, as well as on-site photos.

[0047] For example, the actual performance is calculated using the terrain complexity-corrected early warning accuracy evaluation formula, as follows: In the formula, This indicates the overall early warning accuracy rate after terrain correction; This represents the original early warning accuracy of the s-th station; Indicates the total number of platforms; This represents the combined topographic and climatic weight of the s-th station; This represents the terrain complexity index of the s-th station (normalized from 0 to 1). This represents the climate severity index of the s-th station (normalized from 0 to 1). This represents the average severity index of all stations (used for normalization).

[0048] For example, triggering online incremental learning for false positive / false negative samples is as follows: In the formula, This represents the parameter matrix (weights and biases) of the new model after incremental learning. This represents the parameter matrix of the old model currently in use; This represents the online learning rate (dimensionless), typically ranging from 0.001 to 0.01, and controls the step size for parameter updates. This represents the gradient operator of the loss function with respect to the model parameters, and indicates the direction of parameter optimization. This represents the loss function, which calculates the error between the model's predicted values ​​and the true labels. This represents the coupled feature vector of a sample that was marked as a false positive or false negative (source: output of step 3). The true label of the sample (0=normal, 1=fault) is determined by the maintenance personnel after on-site verification and uploading via the APP.

[0049] Through the above process, the overall early warning accuracy rate after terrain correction is obtained. A adjusted The system includes monthly reports, model parameter update packages (released quarterly), and sensor deployment optimization suggestions (annual assessment). This avoids simply masking the problems of complex terrain sites with basic site data, resulting in a more objective assessment. Furthermore, the model iterates every 30 days to adapt to seasonal climate changes. After running these steps for a year, the false alarm rate will further decrease, forming a continuous optimization loop.

[0050] In summary, this invention fundamentally solves the problem of causal chain breakage caused by the "environmental stripping" of traditional models by systematically collecting environmental baseline data such as topography, meteorology, geology, and historical fault records, and deploying a sensor network covering all elements of infrastructure. Based on multi-source data streams and soil and structural parameters, it calculates fault transmission relationships and internal moisture risk, and constructs a multi-physics coupling model of "rainfall-settlement-tilt-equipment anomaly," breaking through the limitations of isolated monitoring and enabling timely quantification of hidden structural damage. The original Terrain Adaptive Comprehensive Health Index (TA-CHI) dynamically weights terrain modulation factors and climate severity factors, achieving adaptive adjustment of warning thresholds with environmental disturbances. This significantly improves the identification accuracy and early warning lead time under complex terrain, effectively overcoming the shortcomings of fixed threshold models in terms of data degradation and insufficient scene generalization ability. It forms a complete closed loop from environmental perception → causal modeling → dynamic evaluation → graded early warning. At the same time, this invention also reveals hidden fault chains such as "rainfall → settlement → tilt → antenna offset" and transforms isolated monitoring data into causal correlation information, improving fault location accuracy.

[0051] Those skilled in the art will recognize that the units of the various examples described in connection with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of the invention.

[0052] In the embodiments provided by the present invention, it should be understood that the division of units is only a logical functional division. In actual implementation, there may be other division methods, such as multiple units can be combined into one unit, one unit can be split into multiple units, or some features can be ignored.

[0053] Furthermore, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0054] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, read-only memory (ROM), random access memory (RAM), portable hard drives, magnetic disks, or optical disks.

[0055] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features therein. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention, and they should all be covered within the scope of the claims and specification of the present invention.

Claims

1. A fault identification method for radio and television transmission stations in complex terrain, characterized in that, Includes the following steps: Collect station topographic maps, meteorological data from recent years, geological survey reports, and historical fault logs; and deploy sensors. Acquire sensor time-series data, digital elevation model (DEM), and network topology to form a synchronous multi-source data stream output; The fault transmission relationship is calculated based on multi-source data streams, soil physical parameters and tower structural parameters, and the risk of internal moisture is assessed. Based on this, a coupling feature vector is output. Based on coupled feature vectors, real-time meteorological data, and historical baseline data, the overall state is calculated, and a terrain-adaptive comprehensive health index is derived, as follows: In the formula, This indicates the terrain-adaptive comprehensive health index; Indicates the first i Real-time values ​​of each monitoring parameter; Indicates the first i Fault threshold for each parameter; Indicates the first i The historical maximum value of each parameter; Indicates the first i Each parameter has a basic weight; Indicates the terrain modulation factor; Indicates the climate severity modulation factor; Different warning levels are derived based on the scores of the terrain-adaptive comprehensive health index, and corresponding warning signals are then output.

2. The fault identification method for a radio and television transmission station based on complex terrain according to claim 1, characterized in that, The formula for calculating the terrain modulation factor is as follows: In the formula, For reference tower height; The average slope angle of the station; The slope is for flat ground. The formula for calculating the climate severity modulation factor is as follows: In the formula, This represents the rainfall for that day. This represents the average daily rainfall during the local rainy season. Current humidity; This represents the average humidity.

3. The fault identification method for a radio and television transmission station based on complex terrain according to claim 1, characterized in that, Sensor deployment includes setting the monitoring point density based on terrain complexity. The formula for adaptive monitoring point density based on terrain complexity is as follows: In the formula, Indicates the sensor deployment density in the target area; This represents the baseline value for standard deployment density in plains areas; Indicates the terrain influence coefficient; Indicates the elevation of the highest point in the station area; Indicates the elevation of the lowest point in the station area; Indicates the longest diagonal distance of the station area; Indicates geological risk weight; Indicates the historical frequency of geological disasters; A function representing the severity of climate; This indicates the average annual rainfall over many years at the station. This indicates the multi-year average relative humidity of the station.

4. The fault identification method for a radio broadcasting and television transmitting station based on complex terrain according to claim 3, characterized in that, Sensor deployment also includes setting the number of backup sensors for critical locations. The specific formula for calculating the number of backup sensors for critical locations is as follows: In the formula, Indicates the number of backup sensors that need to be configured; Indicates the mean time to repair (MTBT). Indicates the maximum allowable monitoring blind zone time; This indicates the required monitoring reliability probability of the system; This represents the terrain accessibility coefficient.

5. The fault identification method for a radio and television transmission station based on complex terrain according to claim 1, characterized in that, Coupled eigenvector F coupled At least the rainfall-tilt coupling coefficient should be included, and the calculation formula is as follows: In the formula, Indicates the rainfall-tilt coupling coefficient; This indicates the change in the tower's tilt angle; This indicates the change in effective rainfall. Indicates the soil type conversion coefficient; Indicates the area of ​​the tower base platform; Indicates rainwater density; Represents gravitational acceleration; Indicates the foundation drainage capacity coefficient; Indicates the equivalent compression modulus of soil; The tower's full height; This indicates the width of the foundation slab.

6. The fault identification method for a radio broadcasting and television transmitting station based on complex terrain according to claim 1, characterized in that, Coupled eigenvector F coupled At least including.

7. The fault identification method for a radio broadcasting and television transmitting station based on complex terrain according to claim 1, characterized in that, Coupled eigenvector F coupled It should include at least the moisture penetration depth, calculated using the following formula: In the formula, Indicates the depth of moisture penetration; This represents the diffusion coefficient of water vapor in a sealing material; Indicates the duration of continuous exposure to a high temperature and high humidity environment; Equipment enclosure sealing rating coefficient; Indicates the temperature difference between day and night; This represents the coefficient for increasing the gap size due to thermal expansion and contraction.

8. The fault identification method for a radio broadcasting and television transmitting station based on complex terrain according to claim 1, characterized in that, Coupled eigenvector F coupled It should include at least the fatigue damage accumulation index, calculated using the following formula: In the formula, Indicates the cumulative fatigue damage index; This indicates the number of vibration cycles at wind speed level i. Indicates the first i The median wind speed of each wind speed range; Indicates the terrain-corrected SN curve index; Indicates the fatigue strength coefficient of the material; Indicates the weight of turbulence influence; Indicates the standard deviation of wind speed; This indicates the average wind speed.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium includes a stored program, wherein, when the program is executed, it controls the device containing the computer-readable storage medium to perform the fault identification method for a broadcast television wireless transmission station based on complex terrain as described in any one of claims 1 to 8.

10. A processor, characterized in that, The processor is used to run a program, wherein the program executes the fault identification method for broadcast television wireless transmission stations based on complex terrain as described in any one of claims 1 to 8.

Images