Three-dimensional evaluation method for high-cold insulator monitoring triboelectric sensor

Abstract

The application provides a three-dimensional evaluation method for a high-cold insulator monitoring friction electric sensor, and belongs to the technical field of high-cold insulator monitoring. Firstly, a three-dimensional evaluation dimension system is constructed, secondary evaluation indexes under each dimension are set, the dimension weight is calculated by using an analytic hierarchy process, a two-by-two comparison matrix of the dimension weight is constructed based on the priority of high-cold transmission line monitoring business, the dimension weight is calculated by using a geometric mean method and consistency checking is performed; the secondary indexes are quantitatively scored, a Sigmoid nonlinear scoring function is used to realize continuous quantitative scoring of the secondary indexes; a plurality of hard failure conditions are preset; and the maturity level of the equipment is divided according to the grading threshold value. The application has strong universality: as long as the key parameters such as the temperature working interval, annual drift rate and hardware cost of the sensor are obtained, the scoring and evaluation can be completed regardless of the device structure and the research and development unit. In addition to being used as an evaluation tool, the application can also be used as a sensor development and design guide.

Description

Technical Field

[0001] This invention belongs to the field of monitoring technology for insulators in cold regions, specifically, it relates to a three-dimensional evaluation method for triboelectric sensors used in monitoring insulators in cold regions. Background Technology

[0002] Despite the increasing research interest in triboelectric nanogenerators (TENGs) for self-powered monitoring of high-voltage insulators in cold environments, there is currently no standardized evaluation system to assess their comprehensive performance under multiple stress coupling conditions such as extreme cold, icing, freeze-thaw cycles, and ultraviolet radiation.

[0003] As power grid infrastructure continues to extend into high-altitude and cold mountainous areas, insulators in these regions face multiple stresses, including extremely low temperatures, icing, repeated freeze-thaw cycles, and ultraviolet radiation. A 5mm layer of ice can reduce the flashover voltage of insulators by more than 40%; freeze-thaw cycles accelerate the degradation of the hydrophobicity of silicone rubber. Traditional lithium batteries fail below -20℃; photovoltaic equipment cannot generate electricity after snow accumulation; and wired power supply solutions are costly and difficult to construct. Icing flashover and insulation aging are the main causes of line faults, necessitating self-driven sensing solutions.

[0004] Triboelectric nanogenerators (TENGs), first proposed in 2012, can convert environmental mechanical energy into electrical signals, simultaneously achieving energy harvesting and state sensing. Existing research has used TENGs to achieve icing detection (vibration feature migration method, droplet phase transition method, pollution monitoring (charge decay method), and microcrack identification; at the same time, modified materials such as PDMS / ZnO nanocomposites and porous PDMS have been developed to improve the output performance of the devices.

[0005] However, existing research has significant shortcomings: there is a lack of unified evaluation standards adapted to high-altitude, cold, and multi-stress coupled conditions. Numerous prototypes validated in laboratories have failed in field tests (e.g., a 35% false alarm rate, sensor channel blockage after low-temperature icing).

[0006] Current evaluation systems for self-driven sensors focus on single metrics: power density for vibration energy harvesters, photoelectric conversion efficiency for photovoltaic devices, and capacity for batteries. None of these can adequately address the multiple environmental stresses of high-altitude and cold environments. Furthermore, performance evaluation metrics used in different studies are inconsistent, making cross-study comparisons difficult. Existing reviews have summarized progress in TENG environmental monitoring, harsh operating conditions, and extreme scenario adaptation, and there is also a large body of research on TENG power grid monitoring systems. However, CIGRE Technical Report 842 only covers insulator monitoring and does not establish evaluation criteria for self-driven sensors. IEC 60721-2-1 only classifies climate levels, and IEEE Std 528 only specifies inertial sensors; neither standard addresses dedicated TENG equipment for monitoring insulators in high-altitude and cold environments.

[0007] In summary, existing research has significant gaps: there is a lack of a unified system to comprehensively evaluate the performance of TENG sensors under multiple stresses from three dimensions: environmental tolerance, long-term reliability, and field deployability. The fragmented evaluation standards make it difficult to objectively compare sensors developed by different research groups, hindering the transfer of technologies from the laboratory to the power grid, and leaving researchers without clear performance optimization goals. Summary of the Invention

[0008] To address the shortcomings of existing technologies, this invention proposes a three-dimensional evaluation method for triboelectric sensors used in monitoring insulators in cold regions. Based on failure mode analysis, a three-dimensional evaluation framework is systematically built, quantifying sensor maturity from three dimensions: environmental adaptability (EAd), technical reliability (TR), and engineering applicability (EApp). A complete mathematical evaluation model is established, integrating AHP weights, Sigmoid nonlinear scoring, a veto mechanism, and normalization processing. The model is verified through multiple simulations.

[0009] This invention is achieved through the following technical solution: A three-dimensional evaluation method for triboelectric sensors used in monitoring insulators in cold regions: The method specifically includes the following steps: S1: Construct a three-dimensional evaluation system, including environmental adaptability, technical reliability, and engineering practicality. Set secondary evaluation indicators for each dimension, and clarify the definitions of freeze-thaw related indicators in different dimensions to avoid double scoring. S2: The analytic hierarchy process is used to calculate the dimensional weights. Based on the priority of monitoring business for high-altitude and cold-weather transmission lines, a pairwise comparison matrix of dimensional weights is constructed. The geometric mean method is used to calculate the weights of each dimension and perform consistency verification. S3: Secondary indicator quantitative scoring, using the Sigmoid non-linear scoring function, matching the corresponding scoring formula for positive indicators with larger values ​​and negative indicators with smaller values, to achieve continuous quantitative scoring of each secondary indicator from 0 to 100 points. S4: Execute hard failure rejection judgment. Multiple hard failure conditions are preset. If the sensor to be evaluated meets any of the conditions, it is directly judged as unusable in the field and the evaluation process is terminated. S5: Comprehensive maturity calculation and grading. First, calculate the total score of a single dimension by weighting and summing the secondary indicators within the dimension. Then, combine the weights of each dimension and sum them to obtain the comprehensive maturity score of the equipment. Finally, complete the equipment maturity level classification based on the grading threshold.

[0010] Furthermore, in S1: The secondary indicators of the environmental adaptability dimension include operating temperature range, freeze-thaw structural durability, and UV / dust erosion resistance. The secondary indicators of the technical reliability dimension include annual performance drift, measurement accuracy under multiple stresses, measurement stability under freeze-thaw conditions, and resistance to false alarms. The secondary indicators of the engineering practicality dimension include live-line installation capability, maintenance-free lifespan, and single-node hardware cost; Among them, the environmental adaptability dimension is a hard failure threshold index, which evaluates the number of freeze-thaw cycles that can be withstood before physical failures such as cracking, delamination, and complete signal loss occur through appearance and mechanical performance testing. The technical reliability dimension is a soft performance indicator, which evaluates the retention rate of sensor performance under freeze-thaw cycles through electrical signal monitoring.

[0011] Furthermore, in S2, the pairwise comparison matrix of the dimension weights sets the importance ratio based on the on-site operational priority. Environmental adaptability takes priority over technical reliability, and technical reliability takes priority over engineering practicality. The weights of each dimension are obtained by normalization using the geometric mean method, and the consistency coefficient is less than the preset threshold, thus the matrix consistency meets the standard.

[0012] Furthermore, in S3, the sigmoid scoring function for all secondary indicators adopts a uniform slope coefficient to ensure that the scoring sensitivity of each indicator is consistent near the corresponding standard threshold; positive indicators adopt a positive sigmoid scoring formula, and negative indicators adopt a negative sigmoid scoring formula, with the scoring covering the complete percentage range.

[0013] Furthermore, in S4, the hard failure conditions include: complete failure when below a preset low temperature threshold, physical damage occurring when below a preset number of freeze-thaw cycles, annual performance drift exceeding a preset upper limit, lack of live installation capability, and single-node hardware cost exceeding a preset upper limit; when any of these conditions is triggered, the overall maturity of the equipment is directly set to zero, and it is determined to be unusable on site.

[0014] Furthermore, in S5, the maturity level classification rule is as follows: The maturity levels, from high to low, include at least four levels: near-commercial, experimental prototype, laboratory prototype, and unusable in the field. Near-commercial level requires that the overall maturity score reach the preset value, and the scores in environmental adaptability, technical reliability, and engineering usability must not be lower than the corresponding minimum score requirements, so as to avoid high scores in one dimension masking the shortcomings in other dimensions.

[0015] A three-dimensional evaluation system for triboelectric sensors used in monitoring insulators in cold regions; The system includes a three-dimensional evaluation module, a dimension weight calculation module, a quantitative scoring module, a failure determination module, and a maturity classification module; The three-dimensional evaluation module is used to construct a three-dimensional evaluation system, including environmental adaptability, technical reliability, and engineering practicality. It sets secondary evaluation indicators for each dimension and clarifies the definitions of freeze-thaw related indicators in different dimensions to avoid double scoring. The dimension weight calculation module uses the analytic hierarchy process to calculate the dimension weights, constructs a pairwise comparison matrix of dimension weights based on the priority of high-altitude power transmission line monitoring services, and calculates the weights of each dimension using the geometric mean method and performs consistency verification. The quantitative scoring module is used for quantitative scoring of secondary indicators. It adopts the Sigmoid non-linear scoring function and matches the corresponding scoring formulas for positive indicators with larger values ​​and negative indicators with smaller values, so as to achieve continuous quantitative scoring of each secondary indicator from 0 to 100 points. The failure determination module is used to perform hard failure rejection determination. It presets multiple hard failure conditions. If the sensor to be evaluated meets any of the conditions, it is directly determined to be unusable on site, and the evaluation process is terminated. The maturity classification module is used for comprehensive maturity calculation and classification. First, it calculates the total score of a single dimension by weighted summation of the secondary indicators within the dimension. Then, it combines the weights of each dimension to obtain the comprehensive maturity score of the equipment. Finally, it completes the classification of equipment maturity levels based on the classification threshold.

[0016] A computer device system includes a memory, a processor, and a computer program stored in the memory, wherein the processor executes the computer program to implement the steps of the above-described method. A computer-readable storage medium having a computer program / instructions stored thereon, which, when executed by a processor, implement the steps of the above-described method.

[0017] A computer program product includes a computer program / instructions that, when executed by a processor, implement the steps of the method described above.

[0018] Compared with the prior art, the beneficial effects of the present invention are as follows: This invention presents a standardized evaluation framework with quantitative mathematical evaluation standards, rather than a completely new sensor prototype. The framework is highly versatile: as long as key parameters such as the sensor's operating temperature range, annual drift rate, and hardware cost are available, scoring and evaluation can be performed regardless of the device structure or the R&D unit. All verifications of this invention rely solely on publicly available literature data and simulation configurations, verifying the framework's discriminative power and monotonic scoring characteristics through literature backtesting and simulation examples.

[0019] In addition to serving as an evaluation tool, this framework can also act as a sensor development and design guide, clearly defining three dimensions of quantitative optimization goals: environmental adaptability (stable operation at -40℃), technical reliability (annual drift <5%), and engineering usability (single-node cost < $120). This helps R&D personnel reduce the risks of prototype deployment in the field and identify performance shortcomings in the early stages of design. It also supports "what-if analysis": R&D personnel can quantitatively measure the maturity improvement brought about by a single performance optimization (such as extending the operating temperature from -30℃ to -40℃), completing a multi-performance trade-off evaluation before manufacturing a prototype.

[0020] This invention proposes the industry's first standardized evaluation tool specifically for TENG sensors used in monitoring insulators in cold regions, filling the gap in evaluation standards proposed by the CIGRE 2030 strategy; This invention constructs a three-tiered scoring architecture of "environment-technology-engineering" to fully cover the performance trade-off space, which is different from the drawback of single-index evaluation that covers up fatal engineering defects; This invention incorporates a hard failure rejection mechanism: when a key indicator in a single dimension fails to meet the standard, the equipment is directly judged as unqualified, thus avoiding the situation where a high score in one dimension compensates for a fatal defect in another dimension. Attached Figure Description

[0021] Figure 1 This is the 3D evaluation framework of the present invention; Figure 2 A multi-dimensional performance comparison of five sensor technologies; Figure 3 It is a multimodal fusion architecture with autonomous energy integration nodes. Detailed Implementation

[0022] 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.

[0023] Unless otherwise specified, the experimental methods used in the following examples are conventional methods. Unless otherwise specified, the materials, reagents, methods, and instruments used are all conventional materials, reagents, methods, and instruments in the art, and can be obtained commercially by those skilled in the art.

[0024] Combination Figures 1 to 3 This invention proposes a three-dimensional evaluation method for triboelectric sensors used in monitoring insulators in cold regions. I. Derivation of Evaluation Dimensions: The goal of this framework is to establish a unified and quantifiable evaluation tool to objectively compare all TENG sensors used for monitoring insulators in high-altitude and cold-weather environments, regardless of device structure or R&D institution. The framework's theoretical foundation comprises two core elements: 1. Analytic Hierarchy Process (AHP) Multi-Criterion Decision Making: Based on on-site operational priorities, reasonable weights are assigned to multiple performance dimensions; 2. Sigmoid nonlinear scoring function: achieves smooth and continuous scoring, with the highest scoring sensitivity near key thresholds, and covers the entire range of 0 to 100 points.

[0025] Extreme cold environment stress includes temperatures below -40°C, ice adhesion, ultraviolet radiation, and dust erosion. Below -40°C, the polymer dielectric constant changes abruptly, and charge transport efficiency decreases by more than 30% relative to the 20°C baseline; freeze-thaw packaging structures cause sensor sensitivity attenuation of approximately 30%. Therefore, the evaluation system must cover three core aspects: equipment operational capability, environmental durability, and field feasibility. Combining power grid field requirements and industry standards, three evaluation dimensions are defined: 1. Environmental adaptability (EAd): Extreme climate tolerance – operating temperature range (compliant with IEC 60721-2-1), freeze-thaw structural durability, UV / dust erosion resistance; 2. Technical Reliability (TR): Long-term operational stability – annual performance drift (benchmarked against IEEE Std 528, measurement accuracy under multiple stresses, measurement stability under freeze-thaw conditions, and resistance to false alarms); 3. Engineering Applicability (EApp): On-site Feasibility – Live-line Installation Capability, Maintenance-free Lifespan (>5 years), Single-node Hardware Cost (<120 USD).

[0026] To address the issue of duplicate evaluation of freeze-thaw indicators in the EAD and TR dimensions, freeze-thaw related indicators are clearly defined and distinguished under the two dimensions (Table 1) to avoid double scoring: Table 1. Definitions of Freeze-Thaw Indicators for Each Dimension

[0027] The two types of indicators are tested using independent samples: structural durability is tested through appearance and mechanical performance; measurement stability is monitored through electrical signals, and the same type of attenuation phenomenon will not be counted repeatedly.

[0028] II. Quantitative Scoring Model (1) Calculation of AHP Dimension Weights Based on the State Grid's priority for monitoring high-altitude and cold-weather transmission lines, a pairwise comparison matrix was constructed (1 = equally important, 3 = slightly important, 5 = significantly important, 7 = extremely important), as shown in Table 2: Table 2. Pairwise Comparison Matrix of AHP Dimension Weights

[0029] Weight allocation logic: Environmental adaptability is slightly more important than technical reliability (3:1) – if a sensor cannot function properly at -40°C, it has no value for field use regardless of its accuracy. Environmental adaptability is significantly more important than engineering practicality (5:1) – in remote and frigid areas, equipment survival is a priority over cost; Technical reliability is slightly more important than engineering practicality (3:1) – frequent false alarms of icing conditions can lead to unnecessary de-icing operations on the lines, resulting in huge economic losses.

[0030] The weights are calculated using the geometric mean method: EAd geometric mean:

[0031] TR Geometric Mean:

[0032] EApp geometric mean:

[0033] Sum of geometric mean:

[0034] Normalization yields the weights for each dimension:

[0035] Consistency check: Maximum eigenvalue

[0036] Consistency Indicators

[0037] Random Consistency Index

[0038] Consistency coefficient The matrix consistency meets the standard.

[0039] (2) Scoring function for secondary indicators All secondary indicators have equal weight under the same dimension. Currently, there is no authoritative literature that distinguishes the priority of secondary indicators for this application scenario.

[0040] A higher index value indicates better performance (e.g., freeze-thaw cycle count). A positive sigmoid scoring formula is used. (1) Smaller metric values ​​generally indicate better performance (e.g., annual drift, cost, false alarm rate), and an inverse Sigmoid formula is used. (2) In the formula: These are the actual measured values ​​of the indicators; To correspond to the standard threshold; a uniform slope coefficient is used. This ensures that all indicators have consistent scoring sensitivity around the threshold.

[0041] Taking cost metrics as an example (threshold) (USD): $50 score ≈ 74, $120 score = 50, $350 score ≈ 97, $500 score ≈ 99.7; the function covers the entire range of 0 to 100, and the score changes most smoothly and continuously near the threshold.

[0042] (3) Hard failure veto mechanism If any of the following hard failure conditions are met, the overall maturity level of the equipment will be set to 0, and it will be judged as unusable in the field: 1. Completely ineffective in the temperature range of -40℃ and above; 2. Damage occurs before 100 freeze-thaw cycles are completed; 3. Annual performance drift > 15%; 4. Live installation is not possible; 5. Single-node hardware cost > $500.

[0043] (4) Calculation of overall maturity score Single-dimensional total score (weighted sum of average scores for secondary indicators within the dimension):

[0044] Overall Equipment Maturity (M):

[0045] Maturity rating standards (aligned with ISO 16290:2013 Technology Readiness Level (TRL)): 1. Near commercial grade: ,and This corresponds to TRL 6~7 (relevant environment prototype verification, actual working condition prototype verification); 2. Prototype level: This corresponds to TRL 4~5; 3. Laboratory prototype level: This corresponds to TRL 2~3; 4. Not usable on-site: Or it may trigger a hard veto condition.

[0046] Setting minimum score thresholds for each dimension prevents excessively high scores in one dimension from masking weaknesses in others, ensuring balanced performance across all three dimensions for near-commercial-grade equipment. Currently, there is no dedicated TRL standard for TENG equipment in the power industry; this invention uses ISO 16290 as a classification reference, and will further optimize the classification thresholds based on actual feedback from the power grid.

[0047] III. Basis for Setting Indicator Thresholds All indicator thresholds are determined by a combination of three types of information: ① Failure mode analysis of cold-weather insulators (extraction of key stresses and quantification of their impact); ② IEC and IEEE international standards (environmental classification, sensor performance requirements); ③ Field operation data of State Grid's cold-weather transmission lines (actual needs of the power industry).

[0048] Explanation of core thresholds: 1. Operating temperature range: -50℃ to +50℃, covering the area's historical lowest temperature and summer sunshine warming; 2. Freeze-thaw durability: Performance degradation <10% after 1000 cycles; matching insulators have a design life of 20 years. 3. Annual drift limit: <5%, derived from the IEEE Std 528 inertial sensor standard; 4. Project Requirements: Supports live installation, 5-year maintenance-free operation, and complies with power grid operation and maintenance standards; 5. Cost target: Single-node BOM cost < $120, achievable for large-scale mass production.

[0049] Detailed scoring rules for some secondary indicators: 1. UV / Dust Resistance: Performance retention rate is scored according to ISO 4892-3 accelerated aging (1000h UV irradiation) and GB / T 2423.37 dust test (100h): ≥95% gets 100 points, ≥90% gets 80 points, ≥85% gets 60 points, <85% gets 40 points; 2. False alarm resistance: Statistical false alarm rate under the coupled conditions of 10kV / m corona interference + 15m / s wind speed: <1% gets 100 points, <3% gets 80 points, <5% gets 60 points, ≥5% gets 40 points.

[0050] Table 3 Breakdown of Hardware Costs for a Single Sensor Node (for Mass Production of 3000 Units Per Year)

[0051] Note: This cost only covers the material cost of the sensor hardware and does not include on-site installation labor, logistics, or long-term calibration and maintenance costs. These costs vary significantly depending on the deployment scenario and need to be calculated separately by the power grid.

[0052] Cost sensitivity analysis: The target price of $120 is benchmarked against the existing market price of industrial IoT sensors ($150 / $300), which aligns with the power grid's 2023 / 2024 smart sensor bidding and procurement standards. With an annual production scale of 3,000 units, a 40% / 60% discount on electronic components in bulk and a 50% / 70% reduction in labor costs through automated assembly make the $120 material cost target entirely achievable; further cost reduction can be achieved through large-scale manufacturing technology for triboelectric components.

[0053] Small-scale pilot projects (10 / 100 units) have no bulk discounts, resulting in a 50% / 100% increase in unit cost, reaching $180 / $240. The engineering usability score drops from 72 to 55 / 62, and the overall equipment maturity score drops from 84.2 to 78 / 81, downgrading the level to prototype, but this will not trigger the $500 cost veto threshold. Regular grid procurement in batches of 100 / 500 units maintains a cost range of $160-$220, and the cost reduction trend resulting from mass production expansion can be tracked through the EApp score within this framework.

[0054] Simulation calculations show that when material costs fluctuate between $100 and $200, the overall maturity level (M) fluctuates by ≤ ±3.2 points; for equipment to reach near-commercial level (M≥85), material costs are ≤ $160. Pilot prototypes priced at $180 to $240 are classified as experimental prototypes, and the framework scoring can intuitively quantify the trade-off between cost and performance.

[0055] All thresholds are uniformly formulated in accordance with international standards and power grid operation data, and have neutral and universal benchmarking value.

[0056] IV. Literature Data Normalization Processing Scheme To achieve unified backtracking scoring across different document sensors, a three-step normalization formula is designed:

[0057] In the formula The temperature coefficient is extracted from the temperature-output fitting curves of each sensor in the literature; the reference temperature is uniformly set to -25℃.

[0058] Example 1 Sensor (Based on fitting of three sets of data: -20℃, -25℃, and -30℃) Example 2: Sensor Example 3: Sensor .

[0059] For temperature-response nonlinear devices, the slope of the fitted straight line within the operating range is used; the sources of uncertainty must be labeled for the extrapolated range. The test duration is uniformly extrapolated to the standard 1000 freeze-thaw cycles and 1-year long-term operating benchmark using a decay model.

[0060] Limitations and uncertainties of the normalization scheme: 1. Linear Approximations cannot fully capture the strong nonlinear temperature response caused by phase transitions such as the glass transition of polymers, and the normalization results at extreme low temperatures have biases. 2. Inconsistent testing conditions (humidity, wind speed, substrate material) can introduce additional measurement errors. 3. The temperature coefficient is based solely on a single set of data from a single article, without considering batch-to-batch variations in materials; 4. When the test temperature range is narrow, extrapolating to the target temperature will amplify the uncertainty.

[0061] Monte Carlo uncertainty quantification: The above error is quantified using 1000 random propagation simulations, assuming temperature readings, The scoring thresholds all follow a normal distribution, and the standard deviation is taken as 10% of the measurement error or the mean of the index in the literature.

[0062] Simulation results: Example 1 sensor maturity confidence interval of 95% ±2.1 points, Example 2 ±3.4 points, Example 3 ±1.8 points; the self-validation result after 1000 resampling cycles differs from the Monte Carlo error by only 0.3 points, proving the reliability of the uncertainty analysis. None of the sensor error intervals cause a jump in maturity level, demonstrating the strong resistance of the framework to normalization error.

[0063] Note: Monte Carlo error only characterizes the fluctuations introduced by measurement and normalization in a single document and does not include the bias of cross-laboratory repeated testing.

[0064] Normalization is applicable under the following conditions: ① At least two sets of temperature test data must be provided; ② The test duration and attenuation trend must be disclosed; ③ The material system must be able to achieve linear fitting.

[0065] Excluded scenarios: ① Single-temperature point testing only; ② Phase change occurs and segmented fitting data is lacking; ③ Operating conditions such as temperature, humidity, and wind speed deviate from the normal range; ④ Description of testing methods and calibration procedures is missing; ⑤ Irreversible aging of materials occurs before testing.

[0066] All five sensor literatures selected in the examples meet all normalization conditions.

[0067] V. Preliminary Validation of the Framework Five sensor literature samples were selected, covering the complete maturity gradient (laboratory prototype, experimental prototype, and near-commercial prototype), and all of them fully disclosed data on temperature, freeze-thaw cycles, and long-term stability under high-altitude and cold conditions.

[0068] 1. Backtracking verification of sensors from publicly available literature: The scoring model of this invention was used to score five sensors, and the results are shown in Table 4: Table 4. Verification Scoring Results of Three Representative TENG Sensors

[0069] Failure mode analysis of each sensor: (1) The vibration sensor has the lowest environmental adaptability score (23 points), and it completely stops working at -35℃, which does not meet the hard standard of -40℃ high cold. Even if the sensitivity index is the best, it cannot make up for the fatal defect of low temperature failure. (2) The droplet phase change sensor has moderate environmental adaptability (68 points). It can work at low temperatures, but ice bridges accumulate between the electrodes after prolonged icing, causing blockage of the sensing interface. The cause of the failure is the duration of icing rather than the temperature. (3) The environmental adaptability of the charge decay sensor is limited (55 points), and the shortcoming is ultraviolet aging decay; the shortcoming of the microcrack sensor is the narrow operating temperature range. (4) The multimodal fusion system scores leading in all three dimensions (EAd=85, TR=88, EApp=72), and can work stably down to -40℃ after packaging, with excellent long-term reliability; the shortcoming is the cost of a single node of about US$350, and cost optimization is the only bottleneck to achieve commercial-grade performance.

[0070] Compared with traditional single-index evaluation: the vibration sensor has the highest sensitivity among the five devices, and the single-index method would rate it as the best; however, the multi-dimensional framework comprehensively judges its maturity as the lowest (M=34.5), and the defect of failure in extreme cold cannot be compensated. On the other hand, the multi-modal sensor with balanced three-dimensional performance has the highest comprehensive score (M=84.2).

[0071] Comparative analysis shows that a single performance indicator can mask fatal engineering shortcomings, and high sensitivity without environmental tolerance is of no value for field use; the three-dimensional framework of this invention fully covers the performance trade-off space, and the evaluation results are in line with the actual implementation needs of the power grid.

[0072] The multimodal fusion system achieved an overall score of 84.2, just 0.8 points short of the commercial threshold of M≥85, with hardware cost being the bottleneck. Based on the Sigmoid cost function, after reducing the cost per node from $350 to $120, the cost component score increased from 97 to 50, and the overall engineering usability score rose from 72 to 80. Substituting these values ​​into the maturity formula...

[0073] The overall score exceeded the commercial-grade threshold, proving that cost optimization alone can upgrade the equipment to near-commercial grade. Sensitivity analysis showed that when the cost fluctuated between $100 and $200, the maturity score fluctuated by ≤±3.2 points; the cost to stably meet commercial-grade requirements was ≤$160.

[0074] Table 5. Horizontal comparison between the framework of this invention and existing evaluation methods

[0075] 2. Simulation configuration to verify the monotonicity of the framework: Multiple sets of simulated sensor configurations were generated to verify the monotonically increasing scoring characteristic of the framework across the entire performance range.

[0076] Simulation generation logic: The secondary indicators of EAD, TR, and EApp in each dimension are sampled independently and uniformly in the range of 0 to 100, and the dimension scores are aggregated according to the weighting rules of this invention; after randomly generating 1000 sets of configurations, 10 sets of typical results are selected to cover the entire maturity range.

[0077] Simultaneously, by employing relevant sampling (dimensional correlation coefficients of 0.3 and 0.7) and repeated validation using a non-uniform Beta distribution, the law that maturity M monotonically increases with the score of any dimension always holds true, with no score reversal or grade jump phenomenon.

[0078] Table 6 Simulation Sensor Configuration and Framework Output (Demonstration Only)

[0079] Figure 1: Three-dimensional evaluation coordinate system, with thresholds and commercial maturity ranges (EAd > 80, TR > 80, EApp > 70) marked for each dimension; each coordinate axis corresponds to a secondary indicator: EAd axis: operating temperature range, freeze-thaw structural durability, UV / dust resistance; TR axis: annual drift, freeze-thaw measurement stability, and resistance to false alarms; EApp axis: electrically mounted, maintenance-free lifespan, low hardware cost; The scoring follows the Sigmoid function formulas (1)(2).

[0080] Figure 2: Comparison of 3D radar images of five sensors. The error bars represent the 95% confidence interval of Monte Carlo simulation (normalized error ±5%~±7%). The area of ​​the radar image does not represent the overall maturity. Due to the unequal weights of the three dimensions in AHP, it is only used to intuitively compare the performance differences of a single dimension.

[0081] 3. Frame robustness sensitivity analysis: The threshold ranges for each indicator were adjusted, and the fluctuation range of maturity scores was calculated. The results are shown in Table 7. None of the threshold changes caused a jump in the equipment maturity level, and the framework grading remained stable. Taking the maximum fluctuation as an example, after adjusting the freeze-thaw threshold of sensor 2, the maturity fluctuation was 8.2 points, still falling within the 60-85 point range for the test prototype.

[0082] Table 7. Range of Overall Maturity Fluctuation under Threshold Changes

[0083] Currently, the verification relies solely on literature and simulation data; the complete results of this invention include: a three-dimensional evaluation framework, backtesting scoring of five literature-based sensors, simulation configuration verification, robustness and sensitivity analysis, Monte Carlo uncertainty quantification, and SPICE energy self-sufficiency simulation.

[0084] VI. Multimodal Fusion Sensing Architecture: The multimodal fusion architecture of this invention relies on standard device parameters and SPICE circuit simulation, and has two main functions: 1. Demonstrate how to design a sensor system with balanced scores across the three dimensions based on the three-dimensional evaluation framework of this invention; 2. Based on publicly available TENG performance data and industrial device parameters, conservatively estimate the energy self-sufficiency capability; physical processing and experimental verification are listed as subsequent work.

[0085] Based on the framework optimization approach, a multi-channel TENG unit and an intelligent signal decoupling circuit are integrated (Figure 3): Front-end sensing unit: Vibration, droplet phase change, surface charge, and strain sensor arrays, paired with a wide-temperature MEMS temperature and humidity sensor BME280 (-40°C to +85°C), switched in a time-division manner by an 8-channel analog multiplexer ADG708 (-40°C to +125°C); a time-division multiplexing circuit realizes the switching between sensing and energy harvesting modes. Under light wind vibration conditions, the TENG output power is 10~100μW / cm²; a conservative value of 10μW / cm² is taken for small-amplitude insulator vibration.

[0086] Low-power edge processing: Employing a wide-temperature, ultra-low-power STM32L4 MCU (-40~+85℃) with a 1D CNN neural network (parameters < 5k, power consumption < 50mW), it distinguishes three types of fault characteristics: 1. Low-frequency vibration (<50Hz) + sudden temperature change: icing failure; 2. Long-term charge decay: contamination of the insulator surface; 3. High-frequency strain pulse (>100Hz): Microcracks in composite materials during aging.

[0087] Energy storage solution: Hybrid energy storage of 1F supercapacitor + 50mAh wide-temperature lithium battery (-40~+85℃), wireless transmission is only triggered in fault state, which greatly reduces power consumption.

[0088] Fault identification simulation (proof of concept only) The simulation was completed using the 2021-2023 insulator fault public dataset. The dataset synchronously collected vibration, surface charge, and strain signals, covering three types of faults: icing, contamination, and aging.

[0089] 1D CNN network structure: 2 convolutional layers (32 / 64 kernels, kernel size 3), 2 max pooling layers, 2 fully connected layers (64 / 3 output); dataset is divided into training / validation / test sets in a 7:1.5:1.5 ratio, with a total of 12,000 samples, 4,000 samples for each of the three fault classes; Adam optimizer, learning rate 0.001, 200 iterations, early stopping strategy (stops after 20 consecutive iterations with no improvement in accuracy);

[0090] Simulation results: The average identification accuracy of the three types of faults is 93.2%, and the standard deviation of 10 independent runs is ±1.1%; the misclassification rate among faults is less than 2% (1.7% misclassification of dirt due to icing, 1.2% misclassification of aging due to dirt, and 0.9% misclassification of icing due to aging), and the network can effectively separate the coupled signals of multiple faults.

[0091] Important Note: The 93.2% simulation accuracy is only used to demonstrate the concept of multi-fault decoupling and does not represent the performance of the actual equipment; the simulation dataset is a standardized clean signal and is not superimposed with field drift, environmental noise, or electromagnetic interference; the CNN fault diagnosis accuracy under ideal working conditions is consistent with the level of existing literature and is only used as a demonstration of the lower limit of the feasibility of the solution.

[0092] Energy budget calculation: Under small-amplitude vibration conditions of 0.1g and 5Hz, a 2cm² TENG can collect approximately 35μW of energy; the sensor's active sampling power consumption is 85μW, but it is only activated for 3 seconds per day, with an average power consumption of only 3μW.

[0093] The total capacity of the hybrid energy storage is 15mWh, the energy storage efficiency is 80%, and it can work continuously for more than 90 days under extreme conditions of no wind and no vibration, far exceeding the minimum design requirement of 30 days without wind; SPICE simulation verification shows that the minimum remaining state of charge after 30 days is 23%.

[0094] Energy management strategies: Three-level state machine control based on hysteresis threshold: 1. Normal high-sensitivity sensing mode; 2. When the SoC (System on Charge) of the energy storage is less than 40%, switch to the priority energy harvesting mode with a 5% hysteresis range (when the SoC is greater than 45%, switch back to sensing). 3. If SoC < 15%, switch to ultra-low power emergency reporting mode (transmit fault alarm once per hour), 3% hysteresis interval (if SoC > 18%, switch back to energy harvesting). If the energy storage recovers to 25%, it will automatically exit the energy harvesting mode and resume normal sensing. The 5% and 3% hysteresis bands prevent frequent mode oscillations under critical energy storage conditions; under extreme conditions such as heavy snow cover or prolonged windless periods, the equipment still retains basic fault alarm capabilities.

[0095] Source of SPICE simulation parameters: All simulation parameters are traceable: 1. Supercapacitor leakage current: According to the official manual of CAP-XX HS series, the leakage current is 1μA at 25℃ and rises to 15μA at -40℃; 2. Intermittent Vibration Distribution: Meteorological data from the experimental station from 2018 to 2023 were fitted with a Weibull distribution (shape factor k=1.2, scale λ=4.5 hours), and the goodness of fit was... It can accurately reproduce the duration of windless intervals; 3. Annual performance degradation of TENG: Literature data on accelerated aging of PTFE triboelectric materials shows an annual degradation of 5%.

[0096] Figure 3: Self-driven integrated sensor node architecture, labeled ADG708 multiplexer, STM32L4 main controller, and BME280 environmental sensor; power flow, signal flow, and fault output are labeled independently, with energy harvesting power of 35μW, single sampling power consumption of 85μW, and daily average power consumption of 3μW; the dashed box represents the complete self-driven sensor node.

[0097] Three-stage complete verification route: Phase 1 (Q4 2024 – Q1 2025): Laboratory bench validation. Commercial standard TENG modules were placed in a -50℃ to +50℃ high and low temperature chamber to complete freeze-thaw cycle testing. Output voltage, current, and charge density were collected every 5℃. The measured temperature coefficient was compared with that in the literature. In comparison, if the deviation is less than ±15%, the normalization scheme is deemed effective, and the hardware calibration of the scoring function is completed.

[0098] Phase Two (Q2-Q3 2025): The prototype at the 2800m high-altitude cold station will be connected to the grid. The complete winter operating conditions will be compared in parallel with traditional monitoring methods (leakage current, infrared temperature measurement), and the first set of frame calibration test data will be collected.

[0099] Phase 3 (Q4 2025 to Q1 2026): Large-scale pilot project of State Grid transmission lines, deploying 50 / 100 sensors to cover 10 / 20 towers; operating temperature range -40℃ / +30℃, wind speed 0 / 20m / s, snow accumulation 0~50cm, to verify the predictive capabilities of the framework in real power grid operation and maintenance scenarios.

[0100] In addition to serving as a benchmarking and evaluation tool, this framework can also guide the forward development of sensors, clearly defining quantitative optimization goals in three dimensions: stable operation at -40℃ (EAd), annual drift <5% (TR), and single-node material cost <120 USD (EApp). This significantly reduces the risk of on-site deployment of laboratory prototypes, which is the core engineering value of this invention.

[0101] Framework Extensibility and Adaptability: The current framework is designed for extremely cold environments, but the three-layer AHP weighted architecture has general extensibility. Only the secondary indicators need to be replaced and the AHP weights recalibrated. The scoring function and rejection mechanism can be fully reused and adapted to other extreme working conditions. Coastal salt spray environment: Replace UV / dust resistance with salt spray corrosion resistance (ISO 9227) and high humidity resistance (IEC 60068-2-78), and increase the weight of salt spray indicators; adjust the weight of the three dimensions as follows: ; Tropical rainforest environment: New indicators include sustained high temperature (≥40℃), high humidity (>95% RH), and mold bio-erosion. The technical reliability dimension focuses on the stability of temperature and humidity cycle drift; weighting. ; Desert environment: Strengthening indicators of sandstorm erosion and drastic diurnal temperature shock; weighting .

[0102] Only the AHP weights and secondary indicator definitions are adjusted in different regions. The Sigmoid scoring and hard failure rejection logic remain unchanged, ensuring the uniformity of evaluation rules across scenarios. This is a universal evaluation methodology for all types of self-driven sensors in extreme climates.

[0103] The three-dimensional evaluation framework proposed in this invention quantifies sensor maturity from three dimensions: environmental adaptability (EAd), technical reliability (TR), and engineering applicability (EApp).

[0104] This framework uses the Analytic Hierarchy Process (AHP) to determine dimensional weights: environmental adaptability Technical reliability Engineering practicality Consistency test coefficient To meet consistency requirements, the Sigmoid function is used to score each secondary indicator, and a hard failure rejection mechanism is set up—if the sensor fails to meet key indicators (such as not working at -40℃, annual performance drift exceeding 15%, or single-node cost exceeding $500), it is directly judged as unqualified.

[0105] Retrospective scoring was conducted on five existing literature papers on triboelectric sensors and multiple simulation configurations. The results show that the framework can clearly classify device maturity levels: laboratory prototype maturity score. Near-commercial-grade equipment The hierarchical differentiation effect is significant; 1000 Monte Carlo uncertainty propagation simulations validated that no equipment level misjudgment occurred under normal data fluctuations, with a 95% confidence interval fluctuation range of ±1.8 to ±3.4 points. The scoring criteria are aligned with International Electrotechnical Commission (IEC) and Institute of Electrical and Electronics Engineers (IEEE) standards, as well as State Grid's field operation data.

[0106] Simultaneously, a multimodal fusion sensing architecture was proposed, and SPICE circuit simulation verified that the equipment had a minimum remaining charge of 23% after 30 days of energy storage, achieving long-term energy self-sufficiency.

[0107] The research results of this invention can provide a general benchmarking method for new self-powered sensors under extreme climates, fill the technical evaluation gap between laboratory prototypes and on-site grid deployment, and respond to the standardization construction requirements for improving the resilience of power grids in extreme environments in the International Council on Large Electric Systems (CIGRE) 2030 Strategy.

[0108] A computer device system includes a memory, a processor, and a computer program stored in the memory, wherein the processor executes the computer program to implement the steps of the above-described method. A computer-readable storage medium having a computer program / instructions stored thereon, which, when executed by a processor, implement the steps of the above-described method.

[0109] A computer program product includes a computer program / instructions that, when executed by a processor, implement the steps of the method described above.

[0110] The above provides a detailed description of the three-dimensional evaluation method for triboelectric sensors used in monitoring insulators in cold regions, as proposed in this invention. The principles and implementation methods of this invention have been explained. The descriptions of the above embodiments are only for the purpose of helping to understand the method and core ideas of this invention. At the same time, those skilled in the art will recognize that there will be changes in the specific implementation methods and application scope based on the ideas of this invention. Therefore, the content of this specification should not be construed as a limitation of this invention.

Claims

1. A three-dimensional evaluation method for triboelectric sensors used in monitoring insulators in cold regions, characterized in that: The method specifically includes the following steps: S1: Construct a three-dimensional evaluation system, including environmental adaptability, technical reliability, and engineering practicality. Set secondary evaluation indicators for each dimension, and clarify the definitions of freeze-thaw related indicators in different dimensions to avoid double scoring. S2: The analytic hierarchy process is used to calculate the dimensional weights. Based on the priority of monitoring business for high-altitude and cold-weather transmission lines, a pairwise comparison matrix of dimensional weights is constructed. The geometric mean method is used to calculate the weights of each dimension and perform consistency verification. S3: Secondary indicator quantitative scoring, using the Sigmoid non-linear scoring function, matching the corresponding scoring formula for positive indicators with larger values ​​and negative indicators with smaller values, to achieve continuous quantitative scoring of each secondary indicator from 0 to 100 points. S4: Execute hard failure rejection judgment. Multiple hard failure conditions are preset. If the sensor to be evaluated meets any of the conditions, it is directly judged as unusable in the field and the evaluation process is terminated. S5: Comprehensive maturity calculation and grading. First, calculate the total score of a single dimension by weighting and summing the secondary indicators within the dimension. Then, combine the weights of each dimension and sum them to obtain the comprehensive maturity score of the equipment. Finally, complete the equipment maturity level classification based on the grading threshold.

2. The method according to claim 1, characterized in that: In S1: The secondary indicators of the environmental adaptability dimension include operating temperature range, freeze-thaw structural durability, and UV / dust erosion resistance. The secondary indicators of the technical reliability dimension include annual performance drift, measurement accuracy under multiple stresses, measurement stability under freeze-thaw conditions, and resistance to false alarms. The secondary indicators of the engineering practicality dimension include live-line installation capability, maintenance-free lifespan, and single-node hardware cost; Among them, the environmental adaptability dimension is a hard failure threshold index, which evaluates the number of freeze-thaw cycles that can be withstood before physical failures such as cracking, delamination, and complete signal loss occur through appearance and mechanical performance testing. The technical reliability dimension is a soft performance indicator, which evaluates the retention rate of sensor performance under freeze-thaw cycles through electrical signal monitoring.

3. The method according to claim 2, characterized in that: In S2, the pairwise comparison matrix of the dimension weights sets the importance ratio based on the on-site operational priority. Environmental adaptability takes priority over technical reliability, and technical reliability takes priority over engineering practicality. The weights of each dimension are obtained by normalization using the geometric mean method, and the consistency coefficient is less than the preset threshold, thus the matrix consistency meets the standard.

4. The method according to claim 3, characterized in that: In S3, all secondary indicators use a uniform slope coefficient for their Sigmoid scoring functions to ensure consistent scoring sensitivity for each indicator near its corresponding standard threshold. Positive indicators use a positive Sigmoid scoring formula, while negative indicators use a negative Sigmoid scoring formula, covering the entire percentage range.

5. The method according to claim 4, characterized in that: In S4, the hard failure conditions include: complete failure when the temperature is below a preset low temperature threshold, physical damage occurring when the number of freeze-thaw cycles is below a preset limit, annual performance drift exceeding a preset upper limit, lack of live installation capability, and single-node hardware cost exceeding a preset upper limit; when any of these conditions is triggered, the overall maturity of the equipment is directly set to zero, and it is determined to be unusable on site.

6. The method according to claim 5, characterized in that: In S5, the maturity level classification rule is as follows: The maturity levels, from highest to lowest, include at least four levels: near-commercial, experimental prototype, laboratory prototype, and field-unavailable. For near-commercial applications, the overall maturity level must reach a preset score, and the scores for environmental adaptability, technical reliability, and engineering practicality must not be lower than the corresponding minimum score requirements, in order to avoid high scores in one dimension masking shortcomings in other dimensions.

7. A three-dimensional evaluation system for triboelectric sensors used in monitoring insulators in cold regions, characterized in that: The system is used to execute the three-dimensional evaluation method for triboelectric sensors for monitoring insulators in cold regions as described in any one of claims 1 to 6; The system includes a three-dimensional evaluation module, a dimension weight calculation module, a quantitative scoring module, a failure determination module, and a maturity classification module; The three-dimensional evaluation module is used to construct a three-dimensional evaluation system, including environmental adaptability, technical reliability, and engineering practicality. It sets secondary evaluation indicators for each dimension and clarifies the definitions of freeze-thaw related indicators in different dimensions to avoid double scoring. The dimension weight calculation module uses the analytic hierarchy process to calculate the dimension weights, constructs a pairwise comparison matrix of dimension weights based on the priority of high-altitude power transmission line monitoring services, and calculates the weights of each dimension using the geometric mean method and performs consistency verification. The quantitative scoring module is used for quantitative scoring of secondary indicators. It adopts the Sigmoid non-linear scoring function and matches the corresponding scoring formulas for positive indicators with larger values ​​and negative indicators with smaller values, so as to achieve continuous quantitative scoring of each secondary indicator from 0 to 100 points. The failure determination module is used to perform hard failure rejection determination. It presets multiple hard failure conditions. If the sensor to be evaluated meets any of the conditions, it is directly determined to be unusable on site, and the evaluation process is terminated. The maturity classification module is used for comprehensive maturity calculation and classification. First, it calculates the total score of a single dimension by weighted summation of the secondary indicators within the dimension. Then, it combines the weights of each dimension to obtain the comprehensive maturity score of the equipment. Finally, it completes the classification of equipment maturity levels based on the classification threshold.

8. A computer device system, comprising a memory, a processor, and a computer program stored in the memory, characterized in that, The processor executes the computer program to implement the steps of the method according to any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program / instructions stored thereon, characterized in that, When the computer program / instructions are executed by the processor, they implement the steps of the method according to any one of claims 1 to 6.

10. A computer program product, comprising a computer program / instructions, characterized in that, When executed by a processor, the computer program instructions implement the steps of the method according to any one of claims 1 to 6.

Images