A method, device, equipment and medium for risk level assessment of a cable channel

Abstract

This invention discloses a method, apparatus, equipment, and medium for risk level assessment of cable channels, belonging to the field of equipment monitoring. The method comprises: acquiring a continuous monitoring factor dataset and a substitute factor dataset for the cable channel; calculating standardized risk indicators corresponding to the continuous monitoring factor dataset and the substitute factor dataset respectively, obtaining continuous risk indicator values ​​and substitute risk indicator values; mapping the continuous risk indicator values ​​and the substitute risk indicator values ​​to corresponding risk probability values, obtaining continuous risk probability values ​​and substitute risk probability values; performing a two-layer fusion of the continuous risk indicator values ​​and the substitute risk indicator values ​​to obtain a fused risk probability value; and determining the risk level of the cable channel based on the comparison result of the fused risk probability value and a preset risk threshold. This invention can accurately assess the actual conditions of the cable channel, reducing assessment bias and improving assessment accuracy.

Description

Technical Field

[0001] This invention relates to the technical field of cable channel assessment, and more particularly to a method, apparatus, equipment, and medium for assessing the risk level of cable channels. Background Technology

[0002] As the scale of urban power distribution networks and underground transmission cables continues to expand, cable tunnels, integrated utility corridors, cable trenches, and cable interlayers within stations are gradually exhibiting operational characteristics of "enclosed space, dense laying, complex environment, and strong risk coupling." Under dense laying conditions, the number of circuits is large, and limited heat dissipation in the channels easily leads to heat accumulation and localized hotspots. Coupled with factors such as load fluctuations and changes in ventilation conditions, the risk of abnormal temperature rise and insulation thermal aging increases significantly. Simultaneously, environmental factors such as humidity, water accumulation, corrosive media, and dust deposition can easily cause insulation dampness, metal corrosion, and interface defects. Therefore, risk assessments of cable channels are necessary.

[0003] To facilitate risk assessment of underground cable channels in complex environments, one common method is to pre-place different sensors at different locations within the cable channel. These sensors collect different data from the cable channel in real time and compare them with corresponding thresholds. If any detection data fails to meet the corresponding threshold, it is determined that the cable channel has a fault risk. The risk assessment is then performed based on the peak value of the detection data and the number of abnormal detection data.

[0004] However, the above method has the following technical problems: Although the single threshold method can intuitively determine whether there are abnormalities and risks, it is sensitive to operating conditions and is easily affected by seasonal temperature and humidity changes, load transfer, sensor drift and short-term interference, which may trigger false alarms or missed alarms, resulting in certain detection bias and low accuracy of risk assessment. Summary of the Invention

[0005] This invention provides a method, apparatus, equipment, and medium for assessing the risk level of cable channels, which can solve one or more technical problems existing in the prior art.

[0006] A first aspect of this invention provides a method for assessing the risk level of a cable channel, the method comprising: Obtain a continuous monitoring factor dataset and a replacement factor dataset for the cable channel, wherein the continuous monitoring factor dataset is a dataset obtained by detecting the real-time change characteristics of the cable channel, and the replacement factor dataset is a dataset obtained by detecting the periodic change characteristics of the cable channel; Calculate the standardized risk indicators corresponding to the continuous monitoring factor dataset and the alternative factor dataset respectively to obtain the continuous risk indicator value and the alternative risk indicator value; The continuous risk index value and the alternative risk index value are mapped to the corresponding risk probability values ​​respectively to obtain the continuous risk probability value and the alternative risk probability value; The risk level of the cable channel is determined by performing a two-layer fusion of the continuous risk index value and the alternative risk index value, and then comparing the fusion risk probability value with a preset risk threshold.

[0007] In conjunction with the first aspect, in one implementation, the step of obtaining a fused risk probability value through a two-layer fusion of the continuous risk index value and the alternative risk index value, and determining the risk level of the cable channel based on a comparison between the fused risk probability value and a preset risk threshold, includes: Based on adaptive weights, the continuous risk index value and the alternative risk index value are fused at the first layer to obtain the continuous fused index value and the alternative fused index value. The continuous fusion index value and the alternative fusion index value are fused in a second layer according to the gating coefficient to obtain the fusion risk probability value. The gating coefficient is determined based on the data completeness rate of the continuous monitoring factor dataset. The risk level of the cable channel is determined based on the comparison result between the fusion risk probability value and the preset risk threshold.

[0008] In conjunction with the first aspect, in one implementation, the first-level fusion of the continuous risk indicator value and the alternative risk indicator value based on adaptive weights to obtain the continuous fusion indicator value and the alternative fusion indicator value includes: The weight index parameters corresponding to the continuous monitoring factor dataset and the alternative factor dataset are determined respectively, wherein the weight index parameters include: reliability index value and discrimination index value; The continuous adaptive weight value corresponding to the continuous risk indicator value is determined based on the weight indicator parameters, and the alternative adaptive weight value corresponding to the alternative risk indicator value is determined based on the weight indicator parameters. The continuous risk index value is obtained by performing a first-layer fusion using the continuous adaptive weight value, and the alternative risk index value is obtained by performing a first-layer fusion using the alternative adaptive weight value.

[0009] In conjunction with the first aspect, in one implementation, the operation of calculating the standardized risk index corresponding to the continuously monitored factor dataset to obtain the continuous risk index value includes: Calculate the horizontal deviation index value, trend deterioration index value, and abnormal change index value of the continuous monitoring factor dataset, respectively. The continuous risk index value is obtained by weighted fusion of the horizontal deviation index value, the trend deterioration index value, and the abnormal mutation index value.

[0010] In conjunction with the first aspect, in one implementation, the operation of calculating the standardized risk index corresponding to the alternative factor dataset to obtain the alternative risk index value includes: If the alternative factor dataset is a dataset obtained based on ledgers or low-frequency inspections, then a hierarchical function is used to calculate the standardized risk index corresponding to the alternative factor dataset to obtain the alternative risk index value. The standardized risk index includes the index value of the hierarchical mapping of any factor in the alternative factor dataset or the index value of the vulnerability combination mapping. If the substitution factor dataset is a dataset related to the external environment and vulnerability characteristics of cable channels, then the product of the exposure index value and the vulnerability index value corresponding to the substitution factor dataset is calculated to obtain the substitution risk index value.

[0011] In conjunction with the first aspect, in one implementation, mapping the continuous risk indicator value and the alternative risk indicator value to corresponding risk probability values ​​to obtain the continuous risk probability value and the alternative risk probability value includes: Determine the calibration parameters, which include benchmark calibration parameters and risk point calibration parameters; Based on the calibration parameters, the Logistic mapping function is called to map the continuous risk index value and the alternative risk index value to the corresponding risk probability value, thereby obtaining the continuous risk probability value and the alternative risk probability value.

[0012] In conjunction with the first aspect, in one implementation, acquiring the continuous monitoring factor dataset and the alternative factor dataset of the cable channel includes: Obtain a set of risk factors, which includes a set of continuous monitoring factors and a set of alternative factors. The set of continuous monitoring factors includes multiple feature factors used to characterize the real-time state changes of the cable channel, and the set of alternative factors includes multiple feature factors used to characterize the periodic changes of the cable channel. Based on the data collected from the factors in the continuously monitored factor set, a continuously monitored factor dataset is obtained. Based on the data collected from the factors in the alternative factor set, an alternative factor dataset is obtained.

[0013] A second aspect of the present invention provides a risk level assessment device for cable channels, the device comprising: The acquisition module is used to acquire a continuous monitoring factor dataset and a replacement factor dataset for the cable channel, wherein the continuous monitoring factor dataset is a dataset obtained by detecting the real-time change characteristics of the cable channel, and the replacement factor dataset is a dataset obtained by detecting the periodic change characteristics of the cable channel. The indicator calculation module is used to calculate the standardized risk indicators corresponding to the continuous monitoring factor dataset and the alternative factor dataset respectively, and obtain the continuous risk indicator value and the alternative risk indicator value. The probability calculation module is used to map the continuous risk index value and the alternative risk index value to the corresponding risk probability value, respectively, to obtain the continuous risk probability value and the alternative risk probability value. The risk assessment module is used to perform a two-layer fusion of the continuous risk index value and the alternative risk index value to obtain a fused risk probability value, and to determine the risk level of the cable channel based on the comparison result of the fused risk probability value and the preset risk threshold.

[0014] Compared to existing technologies, the present invention provides a method, apparatus, device, and medium for assessing the risk level of cable channels, which has the following advantages: The present invention can detect real-time changes in cable channel characteristics to obtain a continuous monitoring factor dataset and detect periodic changes in cable channel characteristics to obtain a substitution factor dataset; it calculates standardized risk indicators corresponding to the continuous monitoring factor dataset and the substitution factor dataset respectively, obtaining continuous risk indicator values ​​and substitution risk indicator values; it maps the continuous risk indicator values ​​and substitution risk indicator values ​​to corresponding risk probability values, obtaining continuous risk probability values ​​and substitution risk probability values; it performs a two-layer fusion of the continuous risk indicator values ​​and substitution risk indicator values ​​to obtain a fused risk probability value; and it determines the risk level of the cable channel based on the comparison result between the fused risk probability value and a preset risk threshold. The present invention can simultaneously acquire datasets with different characteristics of the cable channel, calculate and fuse risk probability values ​​using datasets with different characteristics, and then perform risk assessment based on the fused risk probability value. This can better reflect the actual condition of the cable channel, reduce assessment bias, and improve assessment accuracy. Attached Figure Description

[0015] Figure 1 This is a flowchart illustrating a risk level assessment method for cable channels according to an embodiment of the present invention; Figure 2 This is a comparative diagram showing the change of risk probability with key factors according to an embodiment of the present invention; Figure 3 This is a schematic diagram of the structure of a risk level assessment device for cable channels provided in an embodiment of the present invention. Detailed Implementation

[0016] 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 embodiments of the present invention, and not all embodiments. 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.

[0017] To address the aforementioned issues, the following specific embodiments will provide a detailed description and explanation of a risk level assessment method, apparatus, equipment, and medium for cable channels provided in this application.

[0018] To address the technical problems of bias and low accuracy in existing cable channel assessment technologies, this paper refers to... Figure 1 The diagram shows a flowchart of a risk level assessment method for cable channels provided in an embodiment of the present invention.

[0019] As an example, the risk level assessment method for the cable channel may include: S11. Obtain the continuous monitoring factor dataset and the alternative factor dataset of the cable channel, wherein the continuous monitoring factor dataset is a dataset obtained by detecting the real-time change characteristics of the cable channel, and the alternative factor dataset is a dataset obtained by detecting the periodic change characteristics of the cable channel.

[0020] In one embodiment, a continuous monitoring factor dataset and a substitution factor dataset for the cable channel can be acquired. The continuous monitoring factor dataset is a dataset obtained by combining data corresponding to real-time changes in the cable channel. The substitution factor dataset is a dataset obtained by combining data corresponding to periodic changes in the cable channel.

[0021] As an example, the acquisition of the continuous monitoring factor dataset and the alternative factor dataset of the cable channel, wherein the continuous monitoring factor dataset is a dataset obtained by detecting the real-time change characteristics of the cable channel, and the alternative factor dataset is a dataset obtained by detecting the periodic change characteristics of the cable channel, may include the following sub-steps: S111. Obtain a set of risk factors, which includes a set of continuous monitoring factors and a set of alternative factors. The set of continuous monitoring factors includes multiple feature factors used to characterize the real-time state changes of the cable channel, and the set of alternative factors includes multiple feature factors used to characterize the periodic changes of the cable channel.

[0022] S112. Based on the data collected from the factors in the continuously monitored factor set, a continuously monitored factor dataset is obtained.

[0023] S113. Collect data corresponding to the factors in the alternative factor set to obtain the alternative factor dataset.

[0024] In one embodiment, a complete set of risk factors can be pre-constructed. Collection, of which the first The factors are denoted as ,satisfy: ; Where F represents the complete set of risk factors for dense cable channels, f j Let j represent the j-th risk factor, j=1,2,…,n represent the risk factor number, and n represent the total number of risk factors included in the evaluation in the current evaluation period.

[0025] Next, the entire set of risk factors can be stratified, and the entire set of risk factors can be divided into layers. The data is divided into a continuous monitoring factor set (set C) and a substitute factor set (set D). Both subsets must satisfy the following equation: ; Where C represents the continuously monitored factor set, D represents the alternative factor set, C∪D means that the two types of factors jointly cover all risk factors, and C∩D=⌀ means that the same risk factor is not repeatedly included in the two sets within the same assessment period.

[0026] Furthermore, the continuous monitoring factor set This includes factors that can reflect transient evolution, rapid deterioration, or early signs of accidents, with data sources being online monitoring devices or continuous acquisition systems; Furthermore, the set of substitution factors This includes factors that reflect long-term degradation, static vulnerability, or structural exposure, with data sources including asset ledgers, structural parameters, historical work orders, or low-frequency inspection records.

[0027] Preferably, for any factor Acquiring time series data The sampling period is ; Preferably, for any factor Acquire non-time-series or low-frequency update data The update cycle is and .

[0028] in, , , middle, Indicates continuous monitoring factor At any moment Time series observations, Indicates the first Each sampling time The original observations, This indicates the sampling period for the factor; for substitute factors, This indicates structural parameters, ledger information, or low-frequency inspection values. or The update period of the substitution factor usually satisfies .

[0029] The continuous monitoring factor set mainly reflects "accident precursors, abrupt evolution, and rapid deterioration." Typical examples include: joint temperature, partial discharge characteristics, sheath current / grounding current anomalies, ambient temperature and humidity, water level / water accumulation, flammable / toxic gas concentration, smoke / fire indications, vibration / intrusion / external disturbances, etc. These factors are characterized by "high-frequency time series, operating condition correlation, and significant fluctuations," requiring a combination of window statistics, trend detection, and anomaly detection to characterize them.

[0030] The alternative factor set primarily reflects "long-term degradation, static vulnerability, and structural exposure." Typical factors include: service life, laying density, number and type of joints, sheath grounding method, channel structural parameters (fire compartments, sealing integrity, ventilation capacity, drainage capacity, waterproofing rating), historical defects and maintenance records, and third-party construction intensity indices. These factors change slowly and can be obtained from records or infrequent inspections, making them suitable as constraints for "baseline risk" and "background vulnerability."

[0031] The continuously monitored factor set (C set) is a set of risk factors that must be continuously monitored, and the alternative factor set (D set) is a set of risk factors that can be replaced by structural parameters, ledger information and / or low-frequency inspection data.

[0032] In one embodiment, the continuous monitoring factor set (C set) includes at least one or more of the following risk factors: cable joint temperature, partial discharge characteristics, sheath current anomaly, grounding current anomaly, channel ambient temperature, channel ambient humidity, combustible gas concentration, toxic gas concentration, oxygen deficiency index, water level / water accumulation status, smoke / fire indication, vibration / external damage intrusion alarm.

[0033] Specifically, a continuous monitoring factor dataset can be obtained by collecting data corresponding to the factors in a continuous monitoring factor set. This continuous monitoring factor dataset can be used to characterize risk states that change in real time with operating conditions within dense cable channels and may trigger faults or accident precursors in a short period of time, in scenarios such as cable tunnels, integrated utility tunnel power compartments, cable trenches, and cable interlayers within stations.

[0034] The continuous monitoring factor dataset may include one or more of the following: cable joint surface temperature, terminal head temperature, distributed fiber optic temperature measurement data, partial discharge pulse amplitude and pulse count, partial discharge phase distribution characteristics, sheath current, grounding current, channel ambient temperature, channel ambient humidity, oxygen concentration, carbon monoxide concentration, concentration of combustible / toxic gases such as hydrogen sulfide or methane, water level, water accumulation switch status, smoke concentration, flame detection status, ventilation equipment operation status, manhole cover opening status, vibration intensity, and external intrusion alarm.

[0035] Among them, the time-series monitoring data in the continuous monitoring factor dataset refers to the data sequence with timestamps that is continuously or quasi-continuously acquired by online monitoring devices, edge acquisition gateways or channel integrated online monitoring platforms according to a preset sampling period. Its data format includes at least one or more of the following: factor number, sampling time, original observation value, engineering unit, sensor location, channel segment number, and data quality identifier. For example, the time-series monitoring data of joint temperature can be represented as a temperature sequence collected every 5s or every 60s, the partial discharge characteristic can be represented as the number of discharge pulses, maximum discharge amount and PRPD statistical characteristics in each statistical window, and the water level data can be represented as a continuous water level height or a 0 / 1 state sequence of the water immersion switch.

[0036] In one embodiment, the alternative factor set (D set) includes at least one or more of the following risk factors: cable service life, insulation type and voltage level, laying method, laying density, historical statistics of circuit load, number of joints, joint type and process level, sheath grounding method, channel structure parameters, fire protection zone and flame retardant configuration level, drainage capacity and waterproof level, historical defect records, maintenance records, and third-party construction intensity index.

[0037] Specifically, alternative factor datasets can be obtained by collecting data corresponding to the factors in the alternative factor set. These alternative factor datasets can be used to characterize static or low-frequency state quantities that change slowly within the assessment period but affect the risk baseline, accident consequences, or channel vulnerability.

[0038] The alternative factor dataset includes one or more of the following: cable service life, insulation material type, voltage level, number of circuits, density of laying in the same trench or corridor, number of joints, joint spacing, joint type and construction technology level, sheath grounding method, cross-interconnection method, historical load peak and load rate statistics, number of historical defects, historical maintenance records, historical water ingress records, historical fire records, third-party construction intensity index, fire compartment level, fireproof sealing integrity, flame retardant configuration level, ventilation capacity, drainage capacity, waterproofing level, and inspection defect level.

[0039] Structural parameters can refer to geometric, configuration, material, and protection parameters obtained from design data, as-built drawings, asset ledgers, BIM models, passageway surveys, or low-frequency inspections. These parameters include passageway length, passageway width, clear height, cross-sectional shape, number of support layers, number of cables per layer, distance between cables and walls / floors, coordinates of joints, length of fire compartments, location of fire-resistant sealing, slope of drainage ditches, capacity of sump pits, number of water pumps and rated drainage capacity, location of ventilation openings, capacity of fans, configuration of fire protection facilities, waterproofing rating, and installation of flame-retardant partitions and fireproof wrapping tape. Continuous sampling of these parameters is not required, but they should be updated after modifications, repairs, defect elimination, or inspection verification.

[0040] S12. Calculate the standardized risk indicators corresponding to the continuous monitoring factor dataset and the alternative factor dataset respectively to obtain the continuous risk indicator value and the alternative risk indicator value.

[0041] In one embodiment, since both the continuous monitoring factor dataset and the alternative factor dataset contain multiple data points, a standardized risk indicator corresponding to the continuous monitoring factor dataset can be calculated to quantify these multiple data points, thus obtaining a continuous risk indicator value. Similarly, a standardized risk indicator corresponding to the alternative factor dataset can also be calculated to obtain an alternative risk indicator value.

[0042] In one optional operating method, monitoring and evaluation criteria of different dimensions can be constructed. In order to unify the objectives of the indicators, standardized risk indicator values ​​corresponding to continuous monitoring factor datasets and alternative factor datasets can be constructed. This allows data with different dimensions, distributions, and frequencies to be compared, while also ensuring directional consistency that "the larger the value, the higher the risk."

[0043] The standardized risk indicators corresponding to the continuously monitored factor dataset and the alternative factor dataset can be in a general form. Preferably, the original observations of the continuously monitored factor dataset and the alternative factor dataset are used... Convert to standardized risk indicators If one of the following conditions is met: ; in, Indicates the first Standardized risk index values ​​for each risk factor Represents the original observations or statistical values. This represents the baseline mean for health. Indicates the standard deviation of the healthy period. This indicates the engineering threshold or alarm threshold. This indicates the tolerance bandwidth, gradation interval, or empirical scale parameter.

[0044] Because indicators such as temperature, current, partial discharge characteristics, gas concentration, water level, and structural defect level have different dimensions, distributions, and dynamic characteristics, directly performing weighted summation can easily lead to "dimensional dominance" and "frequency dominance" problems. This invention addresses this by constructing a unified, standardized risk index. To solve the problem of cross-dimensional comparability, all factors fall within a unified risk scale space.

[0045] In some embodiments, the standardized risk indicator value can be normalized using statistical benchmarks or engineering thresholds, satisfying one of the following expressions: ; in, The health baseline statistical parameters (which can be estimated from historical health period data). For engineering thresholds, These are scale parameters (tolerance bandwidth, gradation interval, or empirical scale).

[0046] For factors where "the smaller the value, the higher the risk" (e.g., excessively low oxygen content, lower drainage capacity level is more dangerous), inverse standardization can be used, for example: ; At the same time, it can be used for Apply lower limit cutoff To avoid negative risk values ​​affecting the probability mapping.

[0047] In an optional embodiment, the operation of calculating the standardized risk index corresponding to the continuously monitored factor dataset to obtain the continuous risk index value may include the following sub-steps: S21. Calculate the level deviation index value, trend deterioration index value, and abnormal mutation index value of the continuous monitoring factor dataset respectively.

[0048] S22. The horizontal deviation index value, the trend deterioration index value, and the abnormal mutation index value are weighted and fused to obtain the continuous risk index value.

[0049] In one embodiment, the level deviation index, trend deterioration index, and anomalous change index values ​​of the continuous monitoring factor dataset can be calculated separately. These three values ​​are then weighted and fused to obtain the continuous risk index value.

[0050] Specifically, the horizontal deviation index value can be determined as follows: ; in, Represents a sliding window The average observed value within, Indicates the window length. Represents the integral variable. Indicates the horizontal threshold. This indicates a horizontal deviation from the scale.

[0051] Optionally, the trend deterioration index value can be determined by the linear fitting slope 𝑘𝑗(𝑡) within the window and the reference slope 𝑘0𝑗, specifically as shown in the following formula: ; in, This represents the measured slope of change obtained by linear fitting within the sliding window. This indicates the reference degradation slope or the engineering-given slope. The larger the value, the more pronounced the continuous deterioration trend of the factor.

[0052] Optionally, the abnormal mutation index value is determined by residual normalization, specifically as shown in the following formula: ; in, This represents the exponentially weighted moving average. This represents the smoothing reference at the previous sampling time. This represents the standard deviation of residuals during the healthy period. Used to characterize the strength of short-term mutations, where , This represents the standard deviation of the residuals during the healthy period.

[0053] Furthermore, for any factor in the continuously monitored factor set (set C) In the sliding window The horizontal deviation index, trend deterioration index, and abnormal change index are calculated separately and then synthesized into a continuous risk index value. Specifically, the continuous risk index value can be expressed by the following formula: ; in, Indicates continuous monitoring factor At any moment The comprehensive standardized risk indicator, i.e., the continuous risk indicator value, , , These represent the values ​​of the horizontal deviation index, the trend deterioration index, and the abnormal change index, respectively. , , For the corresponding composite weights.

[0054] It should be noted that different types of continuous factors can be configured with different... For example, joint temperature emphasizes level and trend, partial discharge emphasizes trend and anomaly, and water level / smoke emphasizes anomalous sudden changes to ensure consistency between model and mechanism.

[0055] In another optional embodiment, the operation of calculating the standardized risk index corresponding to the alternative factor dataset to obtain the alternative risk index value may include the following sub-steps: S31. If the alternative factor dataset is a dataset obtained based on ledgers or low-frequency inspections, then a tiered function is used to calculate the standardized risk index corresponding to the alternative factor dataset to obtain the alternative risk index value. The standardized risk index includes the index value of the tiered mapping of any factor in the alternative factor dataset or the index value of the vulnerability combination mapping.

[0056] S32. If the substitution factor dataset is a dataset related to the external environment and vulnerability characteristics of the cable channel, then calculate the product of the exposure index value and the vulnerability index value corresponding to the substitution factor dataset to obtain the substitution risk index value.

[0057] In one embodiment, for any factor in the alternative factor dataset Alternative risk indicator values ​​can be obtained by using hierarchical mapping or exposure-vulnerability combination mapping. .

[0058] Preferably, the hierarchical mapping to alternative risk indicator values ​​can be expressed as follows: ; in, Indicates substitution factor The hierarchical evaluation results Indicates the total number of levels or the highest level. For hierarchical functions, The alternative risk index value for the alternative factor dataset represents the normalized static vulnerability risk level.

[0059] Optionally, if the alternative factor dataset is a dataset concerning the external environment and vulnerability characteristics of cable tunnels, the alternative risk index value for the exposure-vulnerability combination mapping can be expressed as follows: ; in, Exposure index values ​​(such as third-party construction intensity, joint density) are used to characterize the probability of risk events occurring; The vulnerability index value (such as insufficient protection level or insufficient drainage capacity) is used to characterize the insufficient carrying capacity of the channel or equipment; the product form is used to reflect the relationship of simultaneous risk amplification when there is high exposure and high vulnerability, thus coupling "risk opportunity" and "carrying capacity" into a unified risk index.

[0060] S13. Map the continuous risk index value and the alternative risk index value to the corresponding risk probability value to obtain the continuous risk probability value and the alternative risk probability value.

[0061] In one embodiment, continuous risk index values ​​can be mapped to corresponding risk probability values ​​to obtain continuous risk probability values. Similarly, alternative risk index values ​​can be mapped to corresponding risk probability values ​​to obtain alternative risk probability values.

[0062] Here, mapping can be a way to apply standardized risk indicators. Mapped to risk probability or risk membership degree ,make .

[0063] In one embodiment, mapping the continuous risk index value and the alternative risk index value to corresponding risk probability values ​​to obtain the continuous risk probability value and the alternative risk probability value may include the following sub-steps: S131. Determine the calibration parameters, which include benchmark calibration parameters and risk point calibration parameters.

[0064] S132. Based on the calibration parameters, call the Logistic mapping function to map the continuous risk index value and the alternative risk index value to the corresponding risk probability value, thereby obtaining the continuous risk probability value and the alternative risk probability value.

[0065] Reference Figure 2 The diagram illustrates a comparison of risk probability as a function of key factors according to an embodiment of the present invention.

[0066] To quantify risk and facilitate cross-scenario comparisons, this invention standardizes risk indicator values. (Including continuous risk indicator values ​​and substitute risk indicator values) are mapped to risk probabilities (or risk membership degrees). This yields the probability values ​​of continuous risk and the probability values ​​of alternative risk.

[0067] Preferably, the Logistic mapping function can be used to map the continuous risk index value and the alternative risk index value to the corresponding risk probability value, respectively, to obtain the continuous risk probability value and the alternative risk probability value.

[0068] Specifically, the probability values ​​of continuous risk and alternative risk can be represented by the following formulas: ; The interpretability constraints of the Logistic mapping are as follows: when hour, The larger The larger it is, the more it satisfies monotonicity; Determine the baseline state risk level ( (Prior risks at the time) Determines the sensitivity to risk escalation. This mapping exhibits saturation under extreme conditions, preventing risk values ​​from amplifying indefinitely, making it suitable for engineering alarms and classification.

[0069] In some embodiments, the calibration parameters can be determined as follows: set up and Then the benchmark calibration parameter b j With risk point calibration parameter a j It can be represented as follows: ; in A value of 0.01 to 0.10 can be used to characterize baseline health risk. A value of 0.70 to 0.95 can be used to characterize significant risk. A value of 1 to 4 can be used to represent the standardized range of "crossing significant risks". This method does not rely on large sample failure data and is suitable for engineering scenarios.

[0070] S14. A fusion risk probability value is obtained by performing a two-layer fusion based on the continuous risk index value and the alternative risk index value, and the risk level of the cable channel is determined based on the comparison result between the fusion risk probability value and the preset risk threshold.

[0071] After calculating the continuous risk index value and the alternative risk index value, a two-level fusion can be performed on the continuous risk index value and the alternative risk index value to obtain the fused risk probability value. The two-level fusion can be performed by first fusing the continuous risk index value and the alternative risk index value, and then fusing the two fusion results a second time to obtain the fused risk probability value.

[0072] Finally, the risk probability value of the fusion can be compared with the preset risk threshold, and the risk level of the cable channel can be determined based on the comparison result.

[0073] In one embodiment, the step of obtaining a fused risk probability value by performing a two-layer fusion based on the continuous risk index value and the alternative risk index value, and determining the risk level of the cable channel based on the comparison result of the fused risk probability value and a preset risk threshold, may include the following sub-steps: S141. Based on adaptive weights, the continuous risk index value and the alternative risk index value are fused at the first layer to obtain the continuous fused index value and the alternative fused index value.

[0074] S142. Perform a second-level fusion of the continuous fusion index value and the alternative fusion index value based on the gating coefficient to obtain the fusion risk probability value, wherein the gating coefficient is determined based on the data completeness rate of the continuous monitoring factor dataset.

[0075] S143. Determine the comparison result between the fusion risk probability value and the preset risk threshold, and determine the risk level of the cable channel based on the comparison result.

[0076] In one embodiment, the continuous risk indicator values ​​can be fused at the first layer based on adaptive weights to obtain continuous fused indicator values, and then the alternative risk indicator values ​​can be fused at the first layer based on adaptive weights to obtain alternative fused indicator values.

[0077] Next, a second-level fusion can be performed on the continuous fusion index value and the alternative fusion index value according to the gating coefficient to obtain the fusion risk probability value, wherein the gating coefficient is determined based on the data completeness rate of the continuous monitoring factor dataset.

[0078] In one embodiment, the fusion risk probability value can be expressed as follows: ; in, To integrate the risk probability values, representing the overall risk probability of the channel; The value of the continuous fusion indicator represents the risk probability of the continuous monitoring layer; The replacement fusion index value represents the risk probability of the replacement factor layer; Represents the cross-set gating coefficient; The larger the value, the greater the impact of the continuous monitoring layer on the overall risk. The larger the value, the stronger the background constraint of the substitution factor layer.

[0079] In one embodiment, the gating coefficient The explicit calculation rules are as follows: Preferably, the completeness rate of continuous monitoring data Defined as the percentage of valid sampling points within the evaluation period: ; in, This indicates the completeness rate of continuous monitoring data. Indicates the number of valid sampling points. This indicates the theoretical number of sampling points. This represents the minimum available completeness threshold.

[0080] Preferably, the gating coefficient From completeness With the lowest available threshold Sure: ; Furthermore, when consistency indicators exist (For example, consistency among multiple sensors, consistency among similar locations) When the gating coefficient is corrected, it is as follows: ; in, This represents the corrected gating coefficient. This indicates the consistency of multiple sensors or the consistency of similar locations. The larger the value, the higher the degree of mutual verification between continuous monitoring data.

[0081] Furthermore, risk probability values ​​will be integrated. Mapped to risk level The levels include at least three levels, preferably four levels: I low, II medium, III high, and IV very high, and the threshold values ​​of each range can be set according to engineering requirements.

[0082] In some embodiments, risk probability values ​​will be integrated. Mapped to risk level (For example, I Low, II Medium, III High, IV Very High), the threshold is configurable.

[0083] The specific steps for risk level mapping are as follows: (1) Threshold setting steps: Based on the importance of the channel, voltage level, load nature, historical fault statistics and operation and maintenance management requirements, set three thresholds τ1, τ2 and τ3, satisfying 0<τ1<τ2<τ3<1; when there are no special requirements, τ1=0.25, τ2=0.50 and τ3=0.75.

[0084] (2) Probability preprocessing steps: perform boundary verification on the fusion risk probability value P obtained by the two-layer fusion. If P is less than 0, set it to 0; if P is greater than 1, set it to 1. For scenarios with obvious short-term fluctuations, the sliding average, maximum value or quantile can be taken as the level discrimination input within M consecutive evaluation windows.

[0085] (3) Level determination steps: When 0≤P<τ1, it is determined to be Level I low risk, and routine inspections are maintained; when τ1≤P<τ2, it is determined to be Level II medium risk, and key attention is paid and inspections are appropriately intensified; when τ2≤P<τ3, it is determined to be Level III high risk, triggering an early warning and arranging on-site verification, defect location or load optimization; when τ3≤P≤1, it is determined to be Level IV extremely high risk, triggering emergency response, special investigation, and, if necessary, load reduction or shutdown for maintenance.

[0086] (4) Anti-shake verification steps: In order to avoid frequent level jumps caused by small fluctuations of P near the threshold, a continuous condition or hysteresis interval can be set. For example, when the risk level rises, P is required to exceed the threshold for at least M windows in N consecutive windows, and when the risk level falls, P is required to be lower than the adjacent threshold and continue for a preset time, or a hysteresis band of δ=0.03 to 0.05 can be set.

[0087] (5) Single-factor upgrade verification steps: When the fusion risk probability value P has not yet reached the high-level threshold, but there are strong accident precursor factors such as smoke / fire, toxic gas exceeding the limit, rapid rise in water level, and sharp increase in partial discharge, the risk level can be upgraded by one level or a special alarm can be directly output according to the preset rules in order to avoid the dilution of extreme single-factor risks by overall fusion.

[0088] (6) Results output steps: While outputting the final risk level L, simultaneously output the trigger threshold, judgment basis, ranking of the contribution of the dominant factors, original evidence data and recommended treatment measures to make the risk level interpretable and verifiable.

[0089] In an optional embodiment, to achieve interpretable traceability, the contribution of the dominant factor and the chain of evidence can be output. Specifically, the contribution of the dominant factor is calculated as follows: ; In the formula for calculating the contribution of dominant factors, Γj or γj represents the normalized contribution of the j-th risk factor to the overall risk probability P, with a value ranging from 0 to 1; wj represents the fusion weight of the j-th factor; pj represents the single-factor risk probability of the j-th factor; logit(pj) represents the risk evidence strength of the j-th factor; η represents the gating coefficient between the continuous monitoring layer and the alternative factor layer; and K represents the number of dominant factors output.

[0090] When the j-th factor belongs to set C, its contribution to the overall risk is preferably determined by η, the weight wj within the set, and logit(pj); when the j-th factor belongs to set D, its contribution to the overall risk is preferably determined by 1-η, the weight wj within the set, and logit(pj); finally, the contribution values ​​of all factors are normalized so that the sum of the contributions of each factor is 1, thereby obtaining a sortable and verifiable sequence of the dominant factor contributions.

[0091] Output the top contributors by contribution ranking. The factors are used as the dominant factor set to form the chain of evidence for risk interpretation and the priority of treatment.

[0092] Specifically, the risk interpretation evidence chain is formed in the order of "data source—original observation—standardized evaluation—probability mapping—fusion contribution—engineering implications—disposal recommendations." For each factor in the set of dominant factors, the system records its data source, collection time or update time, monitoring point, original value, threshold or health benchmark, standardized risk indicator zj, single-factor risk probability pj, fusion weight wj, data reliability rj, contribution γj, and triggering reason, thereby forming a risk interpretation entry that can be traced back to the original data.

[0093] For example, when the dominant factor is water level / water accumulation status, the evidence chain includes the section where the water level sensor is located, the maximum water level within the assessment window, the duration, drainage capacity parameters, historical inflow records, standardized risk indicators, single-factor risk probabilities and their contribution to the overall risk; when the dominant factor is a partial discharge characteristic quantity, the evidence chain includes the partial discharge monitoring point, the number of discharge pulses, the maximum discharge quantity, phase distribution characteristics, the joint location, historical maintenance records, standardized risk indicators, probability mapping results and their contribution.

[0094] The priority of handling is determined comprehensively based on risk contribution, single-factor risk probability, accident development speed, data reliability, and handling accessibility. Factors that simultaneously meet the criteria of high contribution, high single-factor risk probability, rapid change speed, and high data reliability are given priority. For factors in cluster C such as sudden temperature rise, rapid enhancement of partial discharge, sudden increase in water level, and smoke or toxic gas alarms, priority is given to generating suggestions for immediate verification, on-site retesting, ventilation and drainage, load reduction, or emergency response. For factors in cluster D such as long service life, high joint density, fireproofing defects, and insufficient drainage capacity, priority is given to generating suggestions for planned maintenance, structural rectification, record review, and long-term remediation.

[0095] Furthermore, a priority score can be constructed: Oj = ω1γj + ω2pj + ω3uj + ω4rj - ω5cj, where uj represents urgency, cj represents difficulty of handling or expected power outage impact, and ω1 to ω5 are configurable weights. The handling list is output according to Oj from high to low, so that maintenance personnel can not only obtain the overall risk level, but also clarify "where to check first, why to check first, and what measures to take".

[0096] In one embodiment, the first-level fusion of the continuous risk index value and the alternative risk index value based on adaptive weights to obtain the continuous fused index value and the alternative fused index value may include the following sub-steps: S1411. Determine the weight index parameters corresponding to the continuous monitoring factor dataset and the alternative factor dataset respectively, wherein the weight index parameters include: reliability index value and discrimination index value.

[0097] S1412. Determine the continuous adaptive weight value corresponding to the continuous risk indicator value based on the weight indicator parameters, and determine the alternative adaptive weight value corresponding to the alternative risk indicator value based on the weight indicator parameters.

[0098] S1413. The continuous risk index value is obtained by performing a first-layer fusion using the continuous adaptive weight value, and the alternative risk index value is obtained by performing a first-layer fusion using the alternative adaptive weight value.

[0099] For any set Collection layer risk probability Log-probability weighted fusion is preferred to obtain continuous fusion index values ​​and alternative fusion index values, as shown in the following formula: ; Represents any set to be merged. Indicates the probability of risk at the set level. Indicates the first The fusion weights of the factors.

[0100] in: ; in, Indicates any location probability variables, Represents the natural logarithm. Used to transform probability values ​​into a linearly weighted log-odds space.

[0101] This method ensures that the fusion result still belongs to [the original source]. It can be interpreted as a weighted geometric average of multiple evidence odds, avoiding the probability distortion caused by simple linear averaging.

[0102] To avoid non-reproducibility due to subjective weighting, this invention uses a reliability index value. Discrimination index value By constructing weights with relevant penalties, we obtain the weight index parameters, which can be expressed as follows: ; in, This represents the initial weights before normalization. Indicates reliability. Indicates the degree of discrimination. Indicates the first The factor and the first The correlation coefficient between the factors This represents the relevant penalty coefficient. This is the penalty coefficient.

[0103] The discrimination index value can be obtained from the historical health / risk sample effect size, as shown in the following formula: ; in, , They represent the first and second risk samples, respectively. The mean and standard deviation of each factor. , These represent the first and second digits of the healthy or normal sample, respectively. The mean and standard deviation of each factor. The larger the value, the better the factor can distinguish between risky and normal states.

[0104] When there are insufficient samples, the discrimination can be replaced by expert scoring or information entropy index, which will not affect the implementation of the framework of this invention.

[0105] In some embodiments, to improve the feasibility and traceability of the project, a data quality label is generated for each factor simultaneously for subsequent adaptive determination of fusion weights. The data quality label includes at least one or more of the following: missing rate, outlier rate, effective sampling percentage / self-check status, and a reliability index value is constructed. , Preferably, a data quality index is generated for each factor, including the missing rate. Outlier rate Self-check status At least one of them, and construct a reliability index value. .

[0106] Preferably, the reliability index value can be expressed as follows: ; in, Indicates the first Data reliability of each factor Indicates the missing rate. Indicates the outlier rate. This indicates the sensor self-test pass rate, effective sampling rate, or confidence level of the ledger record review; The closer This indicates that the factor is more reliable during fusion.

[0107] The following uses a densely populated cable tunnel section as an example to present an evaluation calculation process. For ease of manual calculation and demonstration, four factors from set C and three factors from set D are selected; in actual engineering projects, this can be extended to more factors without changing the essence of this invention.

[0108] Set C (continuous monitoring factors) includes: : Intermediate joint temperature (°C); Partial discharge intensity index (normalized); Water level (cm); CO concentration (ppm).

[0109] Set D (substitution factors) includes: Service years (in years); : Joint density (pieces / 100m); Drainage capacity level (1–5, 5 is the best, with reverse risk).

[0110] Set a sliding window h. Obtained by threshold normalization or level normalization. To unify the probability mapping, a two-point calibration is used: taking... , , .

[0111] The calibration parameters, including the benchmark calibration parameters and the risk point calibration parameters, are obtained from two-point calibration, as shown in the following formula: ; therefore: ; The standardized risk indicator is calculated (example value) as follows: (1) Joint temperature: horizontal deviation ,trend ,abnormal ,Pick : ; (2) Partial discharge: horizontal ,trend ,abnormal ,Pick : ; (3) Water level: Abnormally significant (sudden increase), take .

[0112] (4) CO: Take .

[0113] (5) Service period: 18 years, corresponding to grade level .

[0114] (6) Joint density: 14 joints / 100m, corresponding to graded levels .

[0115] (7) Drainage capacity: Level 2 (poor, reverse), corresponding to .

[0116] Risk probability calculation Substitute each item one by one: ; ; ; ; ; The adaptive weight calculation can be performed as follows: In one embodiment, the reliability index value (example) is as follows: ; The discrimination index values ​​(examples) are as follows: ; Related penalties: set The correlation and approximate values ​​within set C are given: temperature - partial emission 0.4, temperature - CO 0.2, water level - CO 0.1.

[0117] Thus, the adaptive weights for set C normalization are obtained (after example calculation), as follows: ; The adaptive weights for D-set normalization are as follows: ; Next, the first-level fusion within each set and the second-level fusion of the two sets as a whole can be performed.

[0118] Calculate logit: ; We obtain (approximately): , , , , ; The first layer of fusion within each set is as follows: ; ; Cross-set fusion: Continuous monitoring of completeness , ,have to: ; If a consistency coefficient is introduced ,but .

[0119] ; Interpretation of results and recommendations for handling them.

[0120] Overall fusion risk probability value If a four-level threshold example is used (Ⅰ: II: 0.25–0.50; III: 0.50–0.75; IV: If the risk level is high, it falls under Level I (low risk). However, due to the relatively high single-factor probability of water level and localized discharge (…), the risk level is lower. Given its significant weight, it is recommended to trigger a "focused attention" strategy: verify the drainage system and leak points, review the status of partial discharge monitoring points and joints, and, if necessary, increase the frequency of inspections or arrange window maintenance. This result demonstrates the advantages of this invention: it can still output the dominant factor evidence chain when the overall risk is low, avoiding the inoperability of "only giving a single score".

[0121] In some embodiments, the preferred ranges for the key parameters are as follows: Sampling period : 1s~300s, preferably 5s~60s; Sliding window : 10min~24h, preferably 30min~6h; Three-element evaluation weights: satisfying a sum of 1; event-type factors are preferred to increase the weight of anomalies; Logistic calibration: The preferred value is 0.01 to 0.10. The preferred value is 0.70 to 0.95. Preferred options are 1 to 4; Redundancy penalty : 0.1~2.0; Gating threshold : 0.6~0.9; Number of dominant factor outputs :3~10.

[0122] It should be noted that, without altering the essential features of the invention, the following optional modifications are also permitted: 1) In standardized evaluation, trend indicators can be replaced by piecewise slope, cumulative increment or CUSUM statistic; outlier indicators can be replaced by MAD outlier detection or quantile-based outlier intensity. 2) The risk probability mapping can be replaced by Logistic with piecewise exponential, Weibull reliability mapping or fuzzy membership function, but it must satisfy monotonicity and limited output range; 3) Intra-set fusion can be replaced by weighted logit fusion with weighted Noisy-OR, DS evidence theory, or Bayesian fusion, but data reliability and related redundancy must be explicitly considered; 4) The cross-set gating coefficient can be replaced by a combination of "completeness rate + consistency + drift" driven by completeness rate.

[0123] In some embodiments, the present invention is applied to scenarios involving dense cable channels, including cable tunnels, integrated utility tunnel power compartments, cable trenches, shafts, and cable mezzanines within stations. These scenarios are characterized by enclosed spaces, limited heat dissipation, dense circuits, numerous accessories (intermediate joints / terminals), complex environmental factors (high temperature and humidity, water accumulation, corrosive gases, flammable gases and dust), and high risks of external disturbances (damage from third-party construction, illegal intrusion, fire spread). Because risk factors differ significantly across time scales, dimensional scales, and mechanistic pathways, single-threshold alarms or experience-based scoring are insufficient to achieve stable and verifiable comprehensive assessments.

[0124] In some embodiments, the present invention is implemented based on a "channel integrated online monitoring platform + edge acquisition gateway + risk assessment module" architecture. The edge acquisition gateway is used to access online monitoring data such as temperature, partial discharge, sheath current, water level, gas, smoke, and intrusion, and performs data synchronization, missing data repair, outlier removal, and quality labeling. The risk assessment module is deployed on a station-side server or cloud platform to perform standardized indicator construction, risk probability mapping, and hierarchical fusion calculation. The results are output to the operation and maintenance system to form a closed loop for inspection strategies and handling.

[0125] Compared with the prior art, the present invention has at least the following beneficial effects: By dividing risk factors into a set of continuously monitored factors and a set of alternative factors and then merging them in a hierarchical manner, the fusion bias caused by inconsistent monitoring frequencies is avoided, thereby improving the stability of the assessment. By constructing cross-dimensional unified and standardized risk indicators, we can achieve comparable quantification and unified discrimination of heterogeneous risk data; By using a derivable mapping from standardized indicators to risk probabilities, risk outcomes can have clear mathematical meaning and are easy to calibrate and verify. By using a log-probability weighted fusion method within the set and a reliability-discrimination-relevance penalty weighting mechanism, the ability to withstand noise, missing tests, and redundancy is improved. By using a gating coefficient to achieve cross-set adaptive fusion, a stable and reliable overall risk can still be output while continuously monitoring changes in data quality. Output the contribution of dominant factors and the chain of evidence to provide quantitative basis and interpretable support for operation and maintenance actions such as inspection intensification, defect location, load reduction or shutdown for maintenance.

[0126] In this embodiment, the present invention provides a method for risk level assessment of cable channels. Its advantages are as follows: The present invention can detect real-time changes in cable channels to obtain a continuous monitoring factor dataset and detect periodic changes in cable channels to obtain a substitution factor dataset; it calculates standardized risk indicators corresponding to the continuous monitoring factor dataset and the substitution factor dataset respectively, obtaining continuous risk indicator values ​​and substitution risk indicator values; it maps the continuous risk indicator values ​​and substitution risk indicator values ​​to corresponding risk probability values, obtaining continuous risk probability values ​​and substitution risk probability values; it performs a two-layer fusion of the continuous risk indicator values ​​and substitution risk indicator values ​​to obtain a fused risk probability value; and it determines the risk level of the cable channel based on the comparison result between the fused risk probability value and a preset risk threshold. The present invention can simultaneously acquire datasets with different characteristics of the cable channel, calculate and fuse risk probability values ​​using datasets with different characteristics, and then perform risk assessment based on the fused risk probability value. This can better reflect the actual condition of the cable channel, reduce assessment bias, and improve assessment accuracy.

[0127] This invention also provides a risk level assessment device for cable channels, see [link to relevant documentation]. Figure 3 The diagram shows a structural schematic of a risk level assessment device for a cable channel provided in an embodiment of the present invention.

[0128] As an example, the risk level assessment device for the cable channel may include: The acquisition module 201 is used to acquire a continuous monitoring factor dataset and a replacement factor dataset of the cable channel, wherein the continuous monitoring factor dataset is a dataset obtained by detecting the real-time change characteristics of the cable channel, and the replacement factor dataset is a dataset obtained by detecting the periodic change characteristics of the cable channel. Indicator calculation module 202 is used to calculate the standardized risk indicators corresponding to the continuous monitoring factor dataset and the alternative factor dataset respectively, and obtain the continuous risk indicator value and the alternative risk indicator value. The probability calculation module 203 is used to map the continuous risk index value and the alternative risk index value to the corresponding risk probability value, respectively, to obtain the continuous risk probability value and the alternative risk probability value. The risk assessment module 204 is used to perform a two-layer fusion of the continuous risk index value and the alternative risk index value to obtain a fused risk probability value, and to determine the risk level of the cable channel based on the comparison result of the fused risk probability value and the preset risk threshold.

[0129] Optionally, the step of obtaining a fused risk probability value through a two-layer fusion of the continuous risk index value and the alternative risk index value, and determining the risk level of the cable channel based on the comparison result of the fused risk probability value and a preset risk threshold, includes: Based on adaptive weights, the continuous risk index value and the alternative risk index value are fused at the first layer to obtain the continuous fused index value and the alternative fused index value. The continuous fusion index value and the alternative fusion index value are fused in a second layer according to the gating coefficient to obtain the fusion risk probability value. The gating coefficient is determined based on the data completeness rate of the continuous monitoring factor dataset. The risk level of the cable channel is determined based on the comparison result between the fusion risk probability value and the preset risk threshold.

[0130] Optionally, the first-level fusion of the continuous risk indicator value and the alternative risk indicator value based on adaptive weights to obtain the continuous fusion indicator value and the alternative fusion indicator value includes: The weight index parameters corresponding to the continuous monitoring factor dataset and the alternative factor dataset are determined respectively, wherein the weight index parameters include: reliability index value and discrimination index value; The continuous adaptive weight value corresponding to the continuous risk indicator value is determined based on the weight indicator parameters, and the alternative adaptive weight value corresponding to the alternative risk indicator value is determined based on the weight indicator parameters. The continuous risk index value is obtained by performing a first-layer fusion using the continuous adaptive weight value, and the alternative risk index value is obtained by performing a first-layer fusion using the alternative adaptive weight value.

[0131] Optionally, the operation of calculating the standardized risk index corresponding to the continuous monitoring factor dataset to obtain the continuous risk index value includes: Calculate the horizontal deviation index value, trend deterioration index value, and abnormal change index value of the continuous monitoring factor dataset, respectively. The continuous risk index value is obtained by weighted fusion of the horizontal deviation index value, the trend deterioration index value, and the abnormal mutation index value.

[0132] Optionally, the operation of calculating the standardized risk index corresponding to the alternative factor dataset to obtain the alternative risk index value includes: If the alternative factor dataset is a dataset obtained based on ledgers or low-frequency inspections, then a hierarchical function is used to calculate the standardized risk index corresponding to the alternative factor dataset to obtain the alternative risk index value. If the substitution factor dataset is a dataset related to the external environment and vulnerability characteristics of cable channels, then the product of the exposure index value and the vulnerability index value corresponding to the substitution factor dataset is calculated to obtain the substitution risk index value.

[0133] Optionally, mapping the continuous risk index value and the alternative risk index value to corresponding risk probability values ​​to obtain the continuous risk probability value and the alternative risk probability value includes: Determine the calibration parameters, which include benchmark calibration parameters and risk point calibration parameters; Based on the calibration parameters, the Logistic mapping function is called to map the continuous risk index value and the alternative risk index value to the corresponding risk probability value, thereby obtaining the continuous risk probability value and the alternative risk probability value.

[0134] Optionally, the acquisition of the continuous monitoring factor dataset and the alternative factor dataset of the cable channel includes: Obtain a set of risk factors, which includes a set of continuous monitoring factors and a set of alternative factors. The set of continuous monitoring factors includes multiple feature factors used to characterize the real-time state changes of the cable channel, and the set of alternative factors includes multiple feature factors used to characterize the periodic changes of the cable channel. Based on the data collected from the factors in the continuously monitored factor set, a continuously monitored factor dataset is obtained. Based on the data collected from the factors in the alternative factor set, an alternative factor dataset is obtained.

[0135] Those skilled in the art will understand that, for ease of description and brevity, the specific working process of the device described above can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0136] Furthermore, this application also provides an electronic device, including: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the risk level assessment method for cable channels as described in the above embodiments.

[0137] Furthermore, embodiments of this application also provide a computer-readable storage medium storing a computer-executable program for causing a computer to perform the risk level assessment method for cable channels as described in the above embodiments.

[0138] In the description of the embodiments of the present invention, it should be noted that the terms "above," "below," etc., indicating the orientation or positional relationship are based on the orientation or positional relationship shown in the accompanying drawings, and are only for the convenience of describing the embodiments of the present invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the present invention. When an element such as a layer, region, or substrate is referred to as being "above" or "on top of" another element, it may be directly on the other element, or there may be an intermediate element. Conversely, when an element is referred to as being "directly on" or "above" another element, there is no intermediate element. It should also be understood that when an element is referred to as being "below" or "under" another element, it may be directly below or under the other element, or there may be an intermediate element. Conversely, when an element is referred to as being "directly below" or "under" another element, there is no intermediate element. Unless otherwise explicitly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to fixed connections, detachable connections, or integral connections; they can refer to mechanical connections or electrical connections; they can refer to direct connections or indirect connections through an intermediate medium; and they can refer to the internal connection between two components. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0139] Those skilled in the art will understand that embodiments of this application may also include computer program products. Therefore, this application may take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application may take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0140] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), devices, and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0141] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0142] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0143] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the technical principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A method for assessing the risk level of a cable channel, characterized in that, The method includes: Obtain a continuous monitoring factor dataset and a replacement factor dataset for the cable channel, wherein the continuous monitoring factor dataset is a dataset obtained by detecting the real-time change characteristics of the cable channel, and the replacement factor dataset is a dataset obtained by detecting the periodic change characteristics of the cable channel; Calculate the standardized risk indicators corresponding to the continuous monitoring factor dataset and the alternative factor dataset respectively to obtain the continuous risk indicator value and the alternative risk indicator value; The continuous risk index value and the alternative risk index value are mapped to the corresponding risk probability values ​​respectively to obtain the continuous risk probability value and the alternative risk probability value; The risk level of the cable channel is determined by performing a two-layer fusion of the continuous risk index value and the alternative risk index value, and then comparing the fusion risk probability value with a preset risk threshold.

2. The risk level assessment method for cable ducts according to claim 1, characterized in that, The step of obtaining a fused risk probability value through a two-layer fusion of the continuous risk index value and the alternative risk index value, and determining the risk level of the cable channel based on the comparison result of the fused risk probability value and a preset risk threshold, includes: Based on adaptive weights, the continuous risk index value and the alternative risk index value are fused at the first layer to obtain the continuous fused index value and the alternative fused index value. The continuous fusion index value and the alternative fusion index value are fused in a second layer according to the gating coefficient to obtain the fusion risk probability value. The gating coefficient is determined based on the data completeness rate of the continuous monitoring factor dataset. The risk level of the cable channel is determined based on the comparison result between the fusion risk probability value and the preset risk threshold.

3. The risk level assessment method for cable ducts according to claim 2, characterized in that, The first-level fusion of the continuous risk index value and the alternative risk index value based on adaptive weights, to obtain the continuous fusion index value and the alternative fusion index value, includes: The weight index parameters corresponding to the continuous monitoring factor dataset and the alternative factor dataset are determined respectively, wherein the weight index parameters include: reliability index value and discrimination index value; The continuous adaptive weight value corresponding to the continuous risk indicator value is determined based on the weight indicator parameters, and the alternative adaptive weight value corresponding to the alternative risk indicator value is determined based on the weight indicator parameters. The continuous risk index value is obtained by performing a first-layer fusion using the continuous adaptive weight value, and the alternative risk index value is obtained by performing a first-layer fusion using the alternative adaptive weight value.

4. The risk level assessment method for cable ducts according to claim 1, characterized in that, The operation of calculating the standardized risk index corresponding to the continuously monitored factor dataset and obtaining the continuous risk index value includes: Calculate the horizontal deviation index value, trend deterioration index value, and abnormal change index value of the continuous monitoring factor dataset, respectively. The continuous risk index value is obtained by weighted fusion of the horizontal deviation index value, the trend deterioration index value, and the abnormal mutation index value.

5. The risk level assessment method for cable ducts according to claim 1, characterized in that, The operation of calculating the standardized risk index corresponding to the alternative factor dataset and obtaining the alternative risk index value includes: If the alternative factor dataset is a dataset obtained based on ledgers or low-frequency inspections, then a hierarchical function is used to calculate the standardized risk index corresponding to the alternative factor dataset to obtain the alternative risk index value. The standardized risk index includes the index value of the hierarchical mapping of any factor in the alternative factor dataset or the index value of the vulnerability combination mapping. If the substitution factor dataset is a dataset related to the external environment and vulnerability characteristics of cable channels, then the product of the exposure index value and the vulnerability index value corresponding to the substitution factor dataset is calculated to obtain the substitution risk index value.

6. The risk level assessment method for cable ducts according to claim 1, characterized in that, The step of mapping the continuous risk index value and the alternative risk index value to corresponding risk probability values ​​to obtain the continuous risk probability value and the alternative risk probability value includes: Determine the calibration parameters, which include benchmark calibration parameters and risk point calibration parameters; Based on the calibration parameters, the Logistic mapping function is called to map the continuous risk index value and the alternative risk index value to the corresponding risk probability value, thereby obtaining the continuous risk probability value and the alternative risk probability value.

7. The risk level assessment method for cable ducts according to any one of claims 1-6, characterized in that, The acquisition of the continuous monitoring factor dataset and alternative factor dataset for the cable channel includes: Obtain a set of risk factors, which includes a set of continuous monitoring factors and a set of alternative factors. The set of continuous monitoring factors includes multiple feature factors used to characterize the real-time state changes of the cable channel, and the set of alternative factors includes multiple feature factors used to characterize the periodic changes of the cable channel. Based on the data collected from the factors in the continuously monitored factor set, a continuously monitored factor dataset is obtained. Based on the data collected from the factors in the alternative factor set, an alternative factor dataset is obtained.

8. A risk level assessment device for cable channels, characterized in that, The device includes: The acquisition module is used to acquire a continuous monitoring factor dataset and a replacement factor dataset for the cable channel, wherein the continuous monitoring factor dataset is a dataset obtained by detecting the real-time change characteristics of the cable channel, and the replacement factor dataset is a dataset obtained by detecting the periodic change characteristics of the cable channel. The indicator calculation module is used to calculate the standardized risk indicators corresponding to the continuous monitoring factor dataset and the alternative factor dataset respectively, and obtain the continuous risk indicator value and the alternative risk indicator value. The probability calculation module is used to map the continuous risk index value and the alternative risk index value to the corresponding risk probability value, respectively, to obtain the continuous risk probability value and the alternative risk probability value. The risk assessment module is used to perform a two-layer fusion of the continuous risk index value and the alternative risk index value to obtain a fused risk probability value, and to determine the risk level of the cable channel based on the comparison result of the fused risk probability value and the preset risk threshold.

9. An electronic device, comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, when the processor executes the program, it implements the risk level assessment method for cable channels as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions for causing a computer to perform the risk level assessment method for cable channels as described in any one of claims 1-7.

Images