Fire-resistant cable for nuclear power and double-layer insulation state monitoring method thereof

Abstract

The application discloses a fire-resistant and flame-retardant cable for nuclear power and a double-layer insulation state monitoring method thereof, and belongs to the field of cable insulation state monitoring.The method comprises the following steps: collecting a broadband dielectric spectrum signal of double-layer insulation of the fire-resistant and flame-retardant cable for nuclear power to obtain double-layer insulation electrical parameters; applying a phased array ultrasonic excitation to the same cable and collecting ultrasonic echo signals and infrared phase-locked thermal imaging time sequence signals; fusing the two types of signals to construct an interface joint feature map; constructing a Weibull distribution double-stress aging model by using the double-layer insulation electrical parameters and the interface joint feature map, and deducing the life of each region of the double-layer insulation to obtain regional residual life; and mapping the double-layer insulation electrical parameters, the interface joint feature map and the regional residual life into a state grade, and outputting a cable maintenance instruction.The application improves the precision and reliability of nuclear power cable state monitoring, and meets the high safety and redundancy operation and maintenance requirements of a nuclear power station.

Description

Technical Field

[0001] This application relates to the field of cable insulation condition monitoring, and more specifically, to a fire-resistant and flame-retardant cable for nuclear power and a method for monitoring the condition of its double-layer insulation. Background Technology

[0002] As a high-safety-level energy facility, the stability of the power system of a nuclear power plant is directly related to the safety of unit operation and the safety of the surrounding environment. Low-voltage power cables, as core power transmission components, undertake key functions such as power supply to equipment. The design life of nuclear power plants is generally 60 years, while the service life of existing low-voltage cables is mostly 30 to 40 years. Replacing them in the middle of the life poses safety hazards and economic costs.

[0003] The harsh environment of nuclear power plants, such as high temperature, radiation, and humidity, as well as fire accidents, place extremely high demands on the fire resistance (e.g., NS grade, 830℃ / 120 minutes), flame retardancy (B2 grade), and anti-aging performance of cables. Nuclear power cables are used in the above-mentioned harsh environment for a long time, and their inner insulation, outer insulation, and the interface between the two layers will age and be damaged at different rates.

[0004] Traditional dielectric spectroscopy cannot separate the resistivity of the inner layer, the dielectric constant of the outer layer, and the interfacial polarization intensity. Existing non-destructive testing is difficult to quantify micron-level defects such as interfacial bonding strength and thermal diffusivity. Aging models can only give the overall lifetime and cannot predict the remaining lifetime of the inner layer, outer layer, and interface. This results in a lack of targeted operation and maintenance decisions and makes it difficult to meet the 60-year high reliability management requirements of nuclear power plants.

[0005] No effective solutions have yet been proposed to address the problems in the relevant technologies. Summary of the Invention

[0006] To overcome the above problems, this application aims to propose a method for monitoring the condition of fire-resistant and flame-retardant cables for nuclear power plants and their double-layer insulation. The purpose is to solve the problems that traditional dielectric spectra cannot separate the inner layer resistivity, outer layer dielectric constant, and interface polarization intensity; existing non-destructive testing cannot quantify micron-level defects such as interface bonding strength and thermal diffusivity; and aging models can only give the overall lifespan and cannot predict the remaining lifespan of the inner layer, outer layer, and interface, resulting in a lack of targeted operation and maintenance decisions and difficulty in meeting the 60-year high reliability management requirements of nuclear power plants.

[0007] Therefore, the specific technical solution adopted in this application is as follows: In a first aspect, the present invention provides a method for monitoring the double-layer insulation condition of fire-resistant and flame-retardant cables for nuclear power plants, the method comprising: Broadband dielectric spectrum signals of double-layer insulation of fire-resistant and flame-retardant cables for nuclear power were collected, and the electrical parameters of double-layer insulation were obtained by inversion through an adaptive hybrid optimization algorithm. Phased array ultrasonic excitation was applied to the same cable and ultrasonic echo signals and infrared phase-locked thermal imaging time signals were acquired. The two types of signals were fused in a feature layer to construct an interface joint feature map for characterizing the interface bonding strength and thermal diffusivity. A Weibull distributed dual-stress aging model was constructed using the electrical parameters of double-layer insulation and the joint feature map of the interface, and the lifetime of each region of double-layer insulation was extrapolated to obtain the remaining lifetime of each region. The electrical parameters of double insulation, interface joint feature map and remaining lifetime of each region are mapped to state levels and multi-source information is fused. After fault classification according to nuclear power safety redundancy rules, cable maintenance instructions are output.

[0008] Alternatively, the method for obtaining the electrical parameters of double-layer insulation is as follows: Broadband excitation was applied to the double-layer insulation of fire-resistant and flame-retardant cables for nuclear power plants, and their broadband dielectric spectrum signals were collected. The dielectric spectrum signal is preprocessed by denoising and normalization to extract characteristic frequency band data; A positive mapping relationship between dielectric spectral response and electrical parameters of double-layer insulation is established based on characteristic frequency band data; Based on the forward mapping relationship, the electrical parameters of the double-layer insulation are obtained by inverse solving through an adaptive hybrid optimization algorithm.

[0009] Alternatively, the method of solving in reverse using an adaptive hybrid optimization algorithm is as follows: Initialize the candidate solution population, and each candidate solution in the candidate solution population includes the double-layer insulation electrical parameters such as inner layer resistivity, outer layer dielectric constant, and interface polarization intensity; Substitute each candidate solution into the forward mapping relationship, calculate the corresponding theoretical dielectric spectrum response, and compare it with the measured dielectric spectrum signal to obtain the fitness error. Based on the fitness error, candidate solutions are cross-referenced, mutated, and updated, and a better combination of parameters is iteratively searched. Repeat the iteration until the fitness error converges to below the preset threshold, and obtain the current optimal solution as the inverted double-layer insulation electrical parameters. The expression for its calculation process is as follows: ; In the formula, Indicates the current candidate solution The corresponding fitness error; This indicates the total number of frequency points measured in the broadband dielectric spectrum. Indicates the frequency point; Indicates the frequency value The dielectric spectrum signal obtained from the actual measurement; Indicates the first Frequency values ​​at each frequency point; Indicates candidate solutions The theoretical dielectric spectrum response obtained after substituting the positive mapping relation; This represents a candidate solution.

[0010] Optionally, a joint interface feature map is constructed to characterize the interfacial bond strength and thermal diffusivity, including: Phased array ultrasonic excitation was applied to the same cable to acquire ultrasonic echo signals; simultaneously, infrared phase-locked thermal imaging excitation was applied to acquire infrared phase-locked thermal imaging timing signals. Time-frequency analysis was performed on the ultrasonic echo signal to extract characteristic parameters related to the interfacial bonding strength. Phase-locked filtering is performed on the infrared phase-locked thermal imaging time-series signal to extract characteristic parameters associated with thermal diffusivity; After normalizing the two types of feature parameters extracted, a fused feature vector is generated by feature layer fusion. By using the adhesive strength feature in the fused feature vector as the abscissa and the thermal diffusivity feature as the ordinate, and drawing them point by point on a two-dimensional plane, a joint feature map of the interface between adhesive strength and thermal diffusivity is formed.

[0011] Optionally, feature parameters associated with thermal diffusivity are extracted, including: A sinusoidal modulated heat source of a single frequency is applied to the cable, and an infrared thermal imager is used to continuously acquire the infrared phase-locked thermal imaging time sequence signal of the cable surface, recording the temperature change sequence of each pixel over time. Generate a reference sine signal and a reference cosine signal with the same modulation frequency, perform correlation demodulation on the temperature time series of each pixel, and calculate the in-phase component and quadrature component; The thermal wave amplitude is calculated using the sum of squares and square roots of the in-phase and quadrature components, and the phase delay is calculated using the arctangent function. This phase delay serves as a characteristic parameter associated with thermal diffusivity. The expression for its calculation process is as follows: ; In the formula, This represents the extracted feature parameters associated with thermal diffusivity; Represents pixels In the Each sampling time Temperature value; This represents the total number of sampling points within one modulation period; Indicates the sampling point; Indicates the reference cosine signal; Indicates a reference sine wave signal; Indicates the first The sampling time corresponding to each sampling point; This indicates the frequency of the applied sinusoidal modulated heat source.

[0012] Optionally, the method for obtaining the remaining lifetime of each region is as follows: In a multi-stress orthogonal aging platform, cable samples were subjected to radiation, damp heat, and thermal cycling. The electrical parameters of the double insulation and the interface joint characteristic diagram were measured periodically at each aging stage, and the aging time and parameter degradation data were recorded. By using aging time and parameter degradation data, a dual-stress lifetime function under a Weibull distribution is fitted to determine shape parameters and scale parameters, and to establish a quantitative relationship between electrical parameters, interface characteristics and failure time. Substitute the measured electrical parameters of the double insulation and the joint feature map of the interface into the lifetime function, and obtain the cumulative failure probability of each region under the current stress condition for the inner insulation, outer insulation and interlayer interface. Based on the cumulative failure probability of each region, the corresponding remaining lifetime is calculated using an inverse function to obtain the regional remaining lifetime. The expression for the calculation process is as follows: ; In the formula, This represents the cumulative failure probability corresponding to the current aging state; This indicates the equivalent aging time corresponding to the currently measured electrical parameters and interface characteristics; The scale parameter of the Weibull distribution is used to represent the scale parameter of the Weibull distribution. Represents the shape parameters of the Weibull distribution; Indicates the current aging time; This indicates the remaining lifespan of each region.

[0013] Alternatively, the method for fault classification based on nuclear power safety redundancy rules is as follows: Using a pre-constructed fuzzy membership function, the measured values ​​of double-layer insulation electrical parameters, interface joint feature map, and regional remaining lifetime are mapped to various state levels. Each state level is used as evidence. Evidence theory is used to determine the initial probability assignment of each piece of evidence. All evidence is then combined using combination rules to obtain a comprehensive confidence distribution. Based on the nuclear power safety redundancy rules, the fused confidence distribution is used to determine the fault level and output the fault level. Match the fault level in the preset maintenance strategy library and output the corresponding cable maintenance command.

[0014] Optionally, the process of obtaining the comprehensive confidence distribution includes: The state level of each indicator is used as a result in the identification framework of the corresponding evidence, and an initial confidence value is assigned to each state level of the corresponding evidence according to the pre-constructed membership function. A global uncertainty mass function is introduced to assign the unassigned residual confidence level in each piece of evidence to the unknown result, and the confidence level of all results is normalized. The normalized result is used as the initial probability assignment for the evidence. The initial probability assignments of all evidence are iteratively fused to obtain the comprehensive confidence distribution.

[0015] In a second aspect, the present invention provides a fire-resistant and flame-retardant cable for nuclear power plants. The fire-resistant and flame-retardant cable for nuclear power plants includes a tin-plated copper conductor, a calcined mica tape, an inner insulation layer, an outer insulation layer, a ceramicized polyolefin filling layer, a wrapping layer, a copper tape wrapping shielding layer, and an outer sheath. The tin-plated copper conductor, the inner insulation layer, and the outer insulation layer are respectively connected to the broadband dielectric spectrum signal acquisition terminal, the phased array ultrasonic excitation and receiving terminal, and the infrared thermal imaging acquisition terminal of the monitoring host, and cooperate with the monitoring host to perform a double-layer insulation status monitoring method.

[0016] Compared with existing technologies, this application has the following advantages: This application obtains layered electrical parameters through broadband dielectric spectrum inversion, constructs a joint interface feature map by integrating ultrasonic and infrared phase-locked thermal imaging features, and achieves regional remaining lifetime prediction by combining a multi-stress aging platform, Weibull distribution and particle filtering. Furthermore, it completes fault classification decision-making by utilizing fuzzy membership degree and evidence theory, forming a closed-loop process from signal acquisition to maintenance instructions. This enables complementary quantitative detection of double-layer insulation electrical performance and interface defects, as well as independent lifetime extrapolation of the inner layer, outer layer, and interface, significantly improving the accuracy and reliability of nuclear power cable condition monitoring and meeting the high-safety redundancy operation and maintenance requirements of nuclear power plants. Attached Figure Description

[0017] The above-mentioned features, characteristics, and advantages of this application, as well as their implementation methods, will become clearer and more understandable in conjunction with the following description of the embodiments, which are illustrated in detail with reference to the accompanying drawings. Schematic diagrams are shown here: Figure 1 This is a flowchart of the double-layer insulation condition monitoring method for fire-resistant and flame-retardant cables for nuclear power plants in this application; Figure 2 This is a schematic diagram of the structure of the fire-resistant and flame-retardant cable for nuclear power plants in this application.

[0018] In the picture: 1. Tin-plated copper conductor; 2. Calcined mica tape; 3. Inner insulation layer; 4. Outer insulation layer; 5. Ceramicized polyolefin filler layer; 6. Wrapping tape layer; 7. Copper tape wrapped shielding layer; 8. Outer sheath. Detailed Implementation

[0019] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present application.

[0020] This embodiment provides a method for monitoring the double-layer insulation condition of fire-resistant and flame-retardant cables used in nuclear power plants. Layered electrical parameters are obtained through broadband dielectric spectrum inversion. An interface joint feature map is constructed by fusing ultrasonic and infrared phase-locked-array thermal imaging features. Regional remaining life prediction is achieved by combining a multi-stress aging platform, Weibull distribution, and particle filtering. Fault classification decisions are completed using fuzzy membership and evidence theory, forming a closed-loop process from signal acquisition to maintenance instructions. Figure 1 As shown, the method includes: Broadband dielectric spectrum signals of double-layer insulation for fire-resistant and flame-retardant cables used in nuclear power plants were collected, and the electrical parameters of the double-layer insulation were obtained by inversion using an adaptive hybrid optimization algorithm.

[0021] Preferably, the method for obtaining the electrical parameters of double-layer insulation is as follows: Broadband excitation was applied to the double-layer insulation of fire-resistant and flame-retardant cables for nuclear power plants, and their broadband dielectric spectrum signals were collected. The dielectric spectrum signal is preprocessed by denoising and normalization to extract characteristic frequency band data; A positive mapping relationship between dielectric spectral response and electrical parameters of double-layer insulation is established based on characteristic frequency band data; Based on the forward mapping relationship, the electrical parameters of the double-layer insulation are obtained by inverse solving through an adaptive hybrid optimization algorithm.

[0022] Preferably, the method of solving the problem in reverse using an adaptive hybrid optimization algorithm is as follows: Initialize the candidate solution population, and each candidate solution in the candidate solution population includes the double-layer insulation electrical parameters such as inner layer resistivity, outer layer dielectric constant, and interface polarization intensity; Substitute each candidate solution into the forward mapping relationship, calculate the corresponding theoretical dielectric spectrum response, and compare it with the measured dielectric spectrum signal to obtain the fitness error. Based on the fitness error, candidate solutions are cross-referenced, mutated, and updated, and a better combination of parameters is iteratively searched. Repeat the iteration until the fitness error converges to below the preset threshold, and obtain the current optimal solution as the inverted double-layer insulation electrical parameters. The expression for its calculation process is as follows: ; In the formula, Indicates the current candidate solution The corresponding fitness error; This indicates the total number of frequency points measured in the broadband dielectric spectrum. Indicates the frequency point; Indicates the frequency value The dielectric spectrum signal obtained from the actual measurement; Indicates the first Frequency values ​​at each frequency point; Indicates candidate solutions The theoretical dielectric spectrum response obtained after substituting the positive mapping relation; This represents a candidate solution.

[0023] It should be explained that the adaptive hybrid optimization algorithm is a hybrid algorithm formed by the fusion of differential evolution and particle swarm optimization. The adaptive part refers to the algorithm dynamically adjusting the mutation factor, crossover probability or inertia weight during the iteration process to balance the global search and local convergence capabilities. Differential evolution is responsible for exploring the parameter space over a large range, while particle swarm optimization uses the optimal information of individuals and the group to accelerate convergence. The two are executed alternately or in concert to improve the inversion accuracy and stability. The forward mapping relationship refers to the mathematical relationship that calculates the theoretical dielectric spectrum response from given electrical parameters (inner layer resistivity, outer layer dielectric constant, interface polarization intensity); it is usually based on an equivalent circuit model, expressing the dielectric response of layered insulation as a function of frequency. The preset threshold is usually determined based on the measurement noise level and engineering accuracy requirements, for example, 10. −4 The relative error can be less than 1%; a maximum number of iterations (e.g., 500 generations) can also be set as an auxiliary stopping condition to prevent the calculation from going indefinitely; in actual implementation, the threshold needs to be calibrated by simulation. Phased array ultrasonic excitation was applied to the same cable, and ultrasonic echo signals and infrared phase-locked thermal imaging timing signals were acquired. The two types of signals were fused in a feature layer to construct a joint interface feature map for characterizing the interface bonding strength and thermal diffusivity.

[0024] Preferably, constructing a joint interface feature map to characterize the interfacial bond strength and thermal diffusivity includes: Phased array ultrasonic excitation was applied to the same cable to acquire ultrasonic echo signals; simultaneously, infrared phase-locked thermal imaging excitation was applied to acquire infrared phase-locked thermal imaging timing signals. Time-frequency analysis was performed on the ultrasonic echo signal to extract characteristic parameters related to the interfacial bonding strength. Phase-locked filtering is performed on the infrared phase-locked thermal imaging time-series signal to extract characteristic parameters associated with thermal diffusivity; After normalizing the two types of feature parameters extracted, a fused feature vector is generated by feature layer fusion. By using the adhesive strength feature in the fused feature vector as the abscissa and the thermal diffusivity feature as the ordinate, and drawing them point by point on a two-dimensional plane, a joint feature map of the interface between adhesive strength and thermal diffusivity is formed.

[0025] Preferably, the characteristic parameters associated with thermal diffusivity are extracted, including: A sinusoidal modulated heat source of a single frequency is applied to the cable, and an infrared thermal imager is used to continuously acquire the infrared phase-locked thermal imaging time sequence signal of the cable surface, recording the temperature change sequence of each pixel over time. Generate a reference sine signal and a reference cosine signal with the same modulation frequency, perform correlation demodulation on the temperature time series of each pixel, and calculate the in-phase component and quadrature component; The thermal wave amplitude is calculated using the sum of squares and square roots of the in-phase and quadrature components, and the phase delay is calculated using the arctangent function. This phase delay serves as a characteristic parameter associated with thermal diffusivity. The expression for its calculation process is as follows: ; In the formula, This represents the extracted feature parameters associated with thermal diffusivity; Represents pixels In the Each sampling time Temperature value; This represents the total number of sampling points within one modulation period; Indicates the sampling point; Indicates the reference cosine signal; Indicates a reference sine wave signal; Indicates the first The sampling time corresponding to each sampling point; This indicates the frequency of the applied sinusoidal modulated heat source.

[0026] It should be explained that time-frequency analysis commonly uses short-time Fourier transform or wavelet transform; changes in interfacial bonding strength affect the reflection coefficient and energy attenuation of ultrasound between layers; characteristics include: echo peak amplitude (reflected energy), attenuation rate of time-domain waveform envelope, or energy ratio of a specific frequency band; in practice, a mapping curve between bonding strength and ultrasound characteristics can be established first through standard sample experiments, and then the corresponding characteristic values ​​can be extracted from the measured signal; In phase-locked-in thermal imaging, the amplitude is significantly affected by factors such as emissivity, surface reflection, and heating inhomogeneity, while the phase delay mainly depends on the propagation time of the heat wave in the material, is more sensitive to changes in thermal diffusivity, and can effectively suppress surface interference; therefore, the phase delay is preferentially used as a characteristic parameter; in the formula... That is, the phase delay for each pixel; The specific implementation of drawing the interface joint feature map of adhesive strength and thermal diffusivity point by point on a two-dimensional plane is as follows: An ultrasonic phased array can focus on multiple spatial points, each point yielding an adhesion strength value; each pixel of an infrared thermal imager corresponds to a spatial position; coordinate calibration is required beforehand to ensure a one-to-one correspondence between ultrasonic scanning points and thermal imaging pixels; then, points are plotted on a plane with the adhesion strength of each spatial point as the abscissa and the thermal diffusivity as the ordinate, and all points constitute a joint feature map; this map can visually display interface defects (such as debonding points falling in a specific area of ​​the joint feature map).

[0027] A Weibull distributed dual-stress aging model was constructed using the electrical parameters of double-layer insulation and the joint feature map of the interface, and the lifetime of each region of double-layer insulation was extrapolated to obtain the remaining lifetime of each region.

[0028] Preferably, the method for obtaining the remaining lifetime of each region is as follows: In a multi-stress orthogonal aging platform, cable samples were subjected to radiation, damp heat, and thermal cycling. The electrical parameters of the double insulation and the interface joint characteristic diagram were measured periodically at each aging stage, and the aging time and parameter degradation data were recorded. By using aging time and parameter degradation data, a dual-stress lifetime function under a Weibull distribution is fitted to determine shape parameters and scale parameters, and to establish a quantitative relationship between electrical parameters, interface characteristics and failure time. Substitute the measured electrical parameters of the double insulation and the joint feature map of the interface into the lifetime function, and obtain the cumulative failure probability of each region under the current stress condition for the inner insulation, outer insulation and interlayer interface. Based on the cumulative failure probability of each region, the corresponding remaining lifetime is calculated using an inverse function to obtain the regional remaining lifetime. The expression for the calculation process is as follows: ; In the formula, This represents the cumulative failure probability corresponding to the current aging state; This indicates the equivalent aging time corresponding to the currently measured electrical parameters and interface characteristics; The scale parameter of the Weibull distribution is used to represent the scale parameter of the Weibull distribution. Represents the shape parameters of the Weibull distribution; Indicates the current aging time; This indicates the remaining lifespan of each region.

[0029] Online status monitoring of the control circuit cable of the emergency diesel generator inside the containment of Unit 1, Phase II of a certain group; Environmental conditions: The containment is subjected to a high-irradiation (dose rate 0.1 Gy / h), high-humidity and heat (temperature 85℃, humidity 90%RH), alternating temperature (-20℃~120℃ thermal cycling) multi-stress coupled environment for a long time; Significance of monitoring: Emergency diesel generators are the core backup power source in the event of a nuclear power plant accident. Insulation failure of their control circuit cables will directly lead to a nuclear safety level accident. Early warning and preventive replacement plans are required. Monitoring object: Fire-resistant and flame-retardant double-insulated cables for nuclear power plants with a 60-year lifespan; Inner insulation: cross-linked polyethylene, 1.2mm thick, rated temperature 90℃, fire resistance rating 950℃ / 3h; Outer insulation: Low-smoke halogen-free flame-retardant polyolefin, 0.8mm thick, oxygen index ≥32; Service status: It has been in continuous service for 12 years with no historical insulation failure records; Set the following specific parameters, as shown in Table 1: Table 1 Signal Acquisition and Preprocessing Table

[0030] Establish a positive mapping between electrical parameters and dielectric spectral response: The parameter vector to be inverted: ; Inner XLPE insulation resistivity (Ω·cm); : Relative permittivity of the outer LSZH insulation layer; : Interlayer interface polarization intensity (S / m); Forward mapping function: ; The adaptive hybrid optimization algorithm inversion is as follows: Population size: 50; Maximum number of iterations: 500 generations; Preset convergence threshold: E(p) ≤ 10⁻ 4 (or relative error <1%) In differential evolution, the mutation factor F adapts to a range of [0.5, 1.0], and the crossover probability CR adapts to a range of [0.7, 0.9]. Inertia weights in particle swarm optimization w The adaptive range is [0.4, 0.9], and the learning factors are c1=c2=2.0; Substitute the above parameters into the current candidate solution. Corresponding fitness error In the iterative process: the average error of the first generation was 0.235 → the average error of the 100th generation was 0.0087 → the error converged to 8.9 × 10⁻⁶ in the 287th generation. -5 (The threshold is met), and the inversion results are shown in Table 2. Table 2 Inversion Results

[0031] The cable interface status signal was simultaneously acquired using both phased array ultrasonic and infrared phase-locked thermal imaging technologies. Specific parameter settings are as follows: Phased array ultrasonic section: A phased array ultrasonic probe with a probe frequency of 5MHz and an array element number of 64 was used. The focusing depth was set to 2.0mm to accurately align with the interlayer interface of the double insulation. The scanning step size was set to 1mm to scan the target cable monitoring section with a length of 10m point by point.

[0032] Infrared phase-locked thermal imaging section: A 500W modulation heat source is used, with the modulation frequency f0 set to 0.1Hz, which is most sensitive to changes in the thermal diffusivity of a 2mm thick insulating layer; an infrared thermal imager with a resolution of 640×512 pixels is used, with a sampling frequency set to 10Hz, a single-cycle sampling point number K of 100, corresponding to a modulation cycle T = 1 / f0 = 10s, and a sampling time t. k The values ​​are 0.1s, 0.2s, ... up to 10.0s (k = 1 ~ 100). Ultrasonic interface adhesion strength feature extraction, as detailed below: (1) Time-frequency analysis method: Short-time Fourier Transform (STFT), window length 256, overlap rate 50%; (2) Extracted features: echo peak amplitude, time-domain waveform envelope attenuation rate, and energy ratio in the 0.5MHz~2MHz frequency band; (3) Calibration mapping curve (obtained through standard sample experiments): τ b =0.85×A+0.2 (Unit: MPa, A is the peak echo amplitude); (4) Measured results: The peak amplitude of the echo at a certain warning point is A=0.72, corresponding to the interfacial bonding strength. τ b =0.812MPa (new cable reference value 1.5MPa, degradation rate 45.9%). Substituting the parameters: K=100, f0=0.1Hz, the temperature sequence of a certain pixel is T=[25.1, 25.3, 25.6, ..., 24.9]℃; The calculation yielded: =0.32; =0.45; =35.3°; Thermal diffusivity conversion: α =1.2×10 -7 m² / s (new cable reference value 2.1×10) -7 m² / s, degradation rate 42.9%). The 10,000 ultrasonic scanning points are matched one-to-one with the pixels of the thermal imager, with a calibration error of <0.5mm; x-axis = interfacial bonding strength τ b (0~1.5MPa), ordinate = thermal diffusivity α (0~2.5×10) -7 m 2 / s); The regional divisions are as follows: Normal range: (0.8~1.5MPa, 1.5~2.5×10) -7 m 2 / s); Warning zone: (0.3~0.8MPa, 0.8~1.5×10) -7 m 2 / s); Debonding defect area: (0~0.3MPa, 0~0.8×10) -7 m 2 / s); The monitoring results showed that 98% of the area was in the normal aging range, 2% of the area was in the warning range, and there were no debonding defects. Stress combination test: Irradiation (0.05 / 0.1 / 0.2 Gy / h) × Temperature (60 / 85 / 110℃) × Damp heat (70 / 90%RH); Test cycle: 10,000 hours, with electrical parameters and interface characteristics measured every 1,000 hours; The fitting results of the Weibull dual-stress lifetime function are as follows: The shape parameter is 3.2 (typical value for cable insulation aging; a shape parameter >1 indicates an increasing probability of failure over time); the scale parameter is 28.5a (characteristic lifetime corresponding to a cumulative failure probability F = 63.2%); and the failure warning threshold is 10%. (1) Normal aging area (98%); Measured cumulative failure probability F current =2.5%; Calculation process: t current =8.8 years; R 失效 =19.7 years; R 预警 =4.9 years; (2) Early warning aging areas (2%); Measured cumulative failure probability F current =8.5%; The calculation process is as follows: t curren =13.2 years; R 失效 =15.3 years; R 预警 =0.5 years; Therefore, it can be concluded that the overall insulation condition of the cable is good, and the remaining warning life of 98% of the area is about 4.9 years, which can continue to be used; the remaining warning life of 2% of the warning area is only 0.5 years, which needs to be closely monitored within 6 months. If the parameters deteriorate further, partial replacement should be implemented. This application enables accurate assessment of the electrical performance of double-layer insulation and the state of interlayer interfaces in nuclear power units, as well as prediction of remaining life in different regions, meeting the high reliability and high safety requirements for operation and maintenance of nuclear power plants.

[0033] Preferably, the method for fault classification based on nuclear power safety redundancy rules is as follows: Using a pre-constructed fuzzy membership function, the measured values ​​of double-layer insulation electrical parameters, interface joint feature map, and regional remaining lifetime are mapped to various state levels. Each state level is used as evidence. Evidence theory is used to determine the initial probability assignment of each piece of evidence. All evidence is then combined using combination rules to obtain a comprehensive confidence distribution. Based on the nuclear power safety redundancy rules, the fused confidence distribution is used to determine the fault level and output the fault level. Match the fault level in the preset maintenance strategy library and output the corresponding cable maintenance command.

[0034] Preferably, the process of obtaining the comprehensive confidence distribution includes: The state level of each indicator is used as a result in the identification framework of the corresponding evidence, and an initial confidence value is assigned to each state level of the corresponding evidence according to the pre-constructed membership function. A global uncertainty mass function is introduced to assign the unassigned residual confidence level in each piece of evidence to the unknown result, and the confidence level of all results is normalized. The normalized result is used as the initial probability assignment for the evidence. The initial probability assignments of all evidence are iteratively fused to obtain the comprehensive confidence distribution.

[0035] It should be explained that fuzzy membership functions are used to convert continuous measured values ​​into discrete state levels (such as normal, attention, warning, and severe). Trapezoidal or triangular functions are typically used, with each state level corresponding to a membership function. For example, for remaining lifespan, the following can be set: when lifespan > 10 years, the normal membership is 1; from 5 to 10 years, the attention membership increases linearly; from 1 to 5 years, the warning membership is dominant; and from < 1 year, the severe membership is 1, with the sum of the memberships for each level being 1. In actual measurements, the sum of the confidence scores of each level output by the membership function is often less than 1 (because of the presence of fuzzy boundaries or noise). In this case, all the differences (i.e., the unassigned confidence scores) are assigned to the unknown results (represented as the entire set), which represents the degree to which the evidence cannot determine the specific state. This approach avoids misjudgment caused by forced assignment. Nuclear power safety redundancy rules typically include: higher weighting for critical indicators (such as insulation resistance and interface debonding area); if the severity level confidence of any evidence exceeds a threshold (such as 0.7), it is directly determined as the highest fault level; at least two independent pieces of evidence must point to the same level to confirm a medium fault; when there is an unknown with high confidence, it is determined that further investigation is required, etc. Specific rules need to be pre-set and solidified in the monitoring system according to nuclear safety guidelines.

[0036] This embodiment also provides a fire-resistant and flame-retardant cable for nuclear power plants, such as... Figure 2 As shown, the fire-resistant and flame-retardant cable for nuclear power includes a tin-plated copper conductor 1, a calcined mica tape 2, an inner insulation layer 3, an outer insulation layer 4, a ceramicized polyolefin filling layer 5, a wrapping layer 6, a copper tape wrapped shielding layer 7, and an outer sheath 8. The tin-plated copper conductor 1, the inner insulation layer 3, and the outer insulation layer 4 are respectively connected to the broadband dielectric spectrum signal acquisition terminal, the phased array ultrasonic excitation and receiving terminal, and the infrared thermal imaging acquisition terminal of the monitoring host, and cooperate with the monitoring host to perform a double-layer insulation status monitoring method.

[0037] It should be noted that the calculation formulas and all parameters involved in the calculations in this application have been dimensionless beforehand. The process of dimensionless processing is well known in the industry and will not be described here.

[0038] Although the present application has disclosed the preferred embodiments above, the embodiments are merely examples for the purpose of illustration and are not intended to limit the present application. Those skilled in the art can make some modifications and refinements without departing from the spirit and scope of the present application. The scope of protection claimed by the present application should be determined by the claims.

Claims

1. A method for monitoring the double-layer insulation condition of fire-resistant and flame-retardant cables used in nuclear power plants, characterized in that, The method includes: Broadband dielectric spectrum signals of double-layer insulation of fire-resistant and flame-retardant cables for nuclear power were collected, and the electrical parameters of double-layer insulation were obtained by inversion through an adaptive hybrid optimization algorithm. Phased array ultrasonic excitation was applied to the same cable and ultrasonic echo signals and infrared phase-locked thermal imaging time signals were acquired. The two types of signals were fused in a feature layer to construct an interface joint feature map for characterizing the interface bonding strength and thermal diffusivity. A Weibull distributed dual-stress aging model was constructed using the electrical parameters of double-layer insulation and the joint feature map of the interface, and the lifetime of each region of double-layer insulation was extrapolated to obtain the remaining lifetime of each region. The electrical parameters of double insulation, interface joint feature map and remaining lifetime of each region are mapped to state levels and multi-source information is fused. After fault classification according to nuclear power safety redundancy rules, cable maintenance instructions are output.

2. The method for monitoring the condition of double-layer insulation according to claim 1, characterized in that, The method for obtaining the electrical parameters of double-layer insulation is as follows: Broadband excitation was applied to the double-layer insulation of fire-resistant and flame-retardant cables for nuclear power plants, and their broadband dielectric spectrum signals were collected. The dielectric spectrum signal is preprocessed by denoising and normalization to extract characteristic frequency band data; A positive mapping relationship between dielectric spectral response and electrical parameters of double-layer insulation is established based on characteristic frequency band data; Based on the forward mapping relationship, the electrical parameters of the double-layer insulation are obtained by inverse solving through an adaptive hybrid optimization algorithm.

3. The method for monitoring the condition of double-layer insulation according to claim 2, characterized in that, The method of inverse solution using the adaptive hybrid optimization algorithm is as follows: Initialize the candidate solution population, and each candidate solution in the candidate solution population includes the double-layer insulation electrical parameters such as inner layer resistivity, outer layer dielectric constant, and interface polarization intensity; Substitute each candidate solution into the forward mapping relationship, calculate the corresponding theoretical dielectric spectrum response, and compare it with the measured dielectric spectrum signal to obtain the fitness error. Based on the fitness error, candidate solutions are cross-referenced, mutated, and updated, and a better combination of parameters is iteratively searched. Repeat the iteration until the fitness error converges to below the preset threshold, and obtain the current optimal solution as the inverted double-layer insulation electrical parameters. The expression for its calculation process is as follows: ； In the formula, Indicates the current candidate solution The corresponding fitness error; This indicates the total number of frequency points measured in the broadband dielectric spectrum. Indicates the frequency point; Indicates the frequency value The dielectric spectrum signal obtained from the actual measurement; Indicates the first Frequency values ​​at each frequency point; Indicates candidate solutions The theoretical dielectric spectrum response obtained after substituting the positive mapping relationship; This represents a candidate solution.

4. The method for monitoring the condition of double-layer insulation according to claim 1, characterized in that, The construction of the joint interface feature map for characterizing interfacial adhesion strength and thermal diffusivity includes: Phased array ultrasonic excitation was applied to the same cable, and ultrasonic echo signals were acquired; simultaneously, infrared phase-locked thermal imaging excitation was applied, and infrared phase-locked thermal imaging timing signals were acquired. Time-frequency analysis was performed on the ultrasonic echo signal to extract characteristic parameters related to the interfacial bonding strength; Phase-locked filtering is performed on the infrared phase-locked thermal imaging time-series signal to extract characteristic parameters associated with thermal diffusivity; After normalizing the two types of feature parameters extracted, a fused feature vector is generated by feature layer fusion. By using the adhesive strength feature in the fused feature vector as the abscissa and the thermal diffusivity feature as the ordinate, and drawing them point by point on a two-dimensional plane, a joint feature map of the interface between adhesive strength and thermal diffusivity is formed.

5. The method for monitoring the condition of double-layer insulation according to claim 1, characterized in that, The extraction of feature parameters associated with thermal diffusivity includes: A single-frequency sinusoidal modulated heat source is applied to the cable, and an infrared thermal imager is used to continuously acquire the infrared phase-locked thermal imaging time sequence signal of the cable surface, recording the temperature change sequence of each pixel over time. Generate a reference sine signal and a reference cosine signal with the same modulation frequency, perform correlation demodulation on the temperature time series of each pixel, and calculate the in-phase component and quadrature component. The thermal wave amplitude is calculated using the sum of squares and square roots of the in-phase and quadrature components, and the phase delay is calculated using the arctangent function. This phase delay serves as a characteristic parameter associated with thermal diffusivity. The expression for its calculation process is as follows: ； In the formula, This represents the extracted feature parameters associated with thermal diffusivity; Represents pixels In the Each sampling time Temperature value; This represents the total number of sampling points within one modulation period; Indicates the sampling point; Indicates the reference cosine signal; Indicates a reference sine wave signal; Indicates the first The sampling time corresponding to each sampling point; This indicates the frequency of the applied sinusoidal modulated heat source.

6. The method for monitoring the condition of double-layer insulation according to claim 1, characterized in that, The method for obtaining the remaining lifetime of each region is as follows: In a multi-stress orthogonal aging platform, cable samples were subjected to radiation, damp heat, and thermal cycling. The electrical parameters of the double insulation and the interface joint characteristic diagram were measured periodically at each aging stage, and the aging time and parameter degradation data were recorded. By using aging time and parameter degradation data, a dual-stress lifetime function under a Weibull distribution is fitted to determine shape parameters and scale parameters, and to establish a quantitative relationship between electrical parameters, interface characteristics and failure time. Substitute the measured electrical parameters of the double insulation and the joint feature map of the interface into the lifetime function, and obtain the cumulative failure probability of each region under the current stress condition for the inner insulation, outer insulation and interlayer interface. Based on the cumulative failure probability of each region, the corresponding remaining lifetime is calculated using an inverse function to obtain the regional remaining lifetime. The expression for the calculation process is as follows: ； In the formula, This represents the cumulative failure probability corresponding to the current aging state; This indicates the equivalent aging time corresponding to the currently measured electrical parameters and interface characteristics; The scale parameter of the Weibull distribution is used to represent the scale parameter of the Weibull distribution. Represents the shape parameters of the Weibull distribution; Indicates the current aging time; This indicates the remaining lifespan of each region.

7. The method for monitoring the condition of double-layer insulation according to claim 1, characterized in that, The method for fault classification based on nuclear power safety redundancy rules is as follows: Using a pre-constructed fuzzy membership function, the measured values ​​of double-layer insulation electrical parameters, interface joint feature map, and regional remaining lifetime are mapped to various state levels. Each state level is used as evidence. Evidence theory is used to determine the initial probability assignment of each piece of evidence. All evidence is then combined using combination rules to obtain a comprehensive confidence distribution. Based on the nuclear power safety redundancy rules, the fused confidence distribution is used to determine the fault level and output the fault level. Match the fault level in the preset maintenance strategy library and output the corresponding cable maintenance command.

8. The method for monitoring the condition of double-layer insulation according to claim 7, characterized in that, The process of obtaining the comprehensive confidence distribution includes: The state level of each indicator is used as a result in the identification framework of the corresponding evidence, and an initial confidence value is assigned to each state level of the corresponding evidence according to the pre-constructed membership function. A global uncertainty mass function is introduced to assign the unassigned residual confidence level in each piece of evidence to the unknown result, and the confidence level of all results is normalized. The normalized result is used as the initial probability assignment for the evidence. The initial probability assignments of all evidence are iteratively fused to obtain the comprehensive confidence distribution.

9. A fire-resistant and flame-retardant cable for nuclear power, the fire-resistant and flame-retardant cable for nuclear power comprising a tin-plated copper conductor (1), a calcined mica tape (2), an inner insulation layer (3), an outer insulation layer (4), a ceramicized polyolefin filler layer (5), a wrapping layer (6), a copper tape wrapped shielding layer (7), and an outer sheath (8), characterized in that, The tin-plated copper conductor (1), inner insulation layer (3), and outer insulation layer (4) are respectively connected to the broadband dielectric spectrum signal acquisition terminal, phased array ultrasonic excitation and receiving terminal, and infrared thermal imaging acquisition terminal of the monitoring host, and cooperate with the monitoring host to perform the double-layer insulation status monitoring method according to any one of claims 1 to 8.

Images