Cable operation environment optimization method based on cable trench acquisition information

Abstract

The invention discloses a cable operation environment optimization method based on cable trench collection information, and relates to the technical field of intelligent operation and maintenance of a power system, and the method comprises the steps: collecting an original parameter data set in a cable trench; a pulse current signal is collected at a cable grounding wire connection point, and a standard pulse discharge capacity sequence is generated through band-pass filtering and amplitude integral conversion; based on the original relative humidity measurement value and the environment temperature value, calculating a compensated relative humidity value through a multi-scale heat conduction-diffusion coupling equation; calculating a cable insulation risk index through a quantum annealing algorithm by using the compensated relative humidity value and the standard pulse discharge capacity sequence; the dynamic measurement deviation of the temperature and humidity sensor is subjected to double compensation of thermal radiation and molecular adsorption by constructing a multi-scale heat conduction-diffusion coupling equation, so that the reliability of environmental state sensing is greatly improved, and the operation safety, stability and intelligent operation and maintenance level of a power cable system are effectively enhanced.

Description

Technical Field

[0001] This invention relates to the field of intelligent operation and maintenance technology for power systems, and in particular to a method for optimizing the cable operating environment based on information collected from cable trenches. Background Technology

[0002] With the continuous expansion of urban power grids and the increasing demands for safety and reliability in power systems, real-time monitoring and intelligent control technologies for cable operating environments have become an important research direction in the field of power equipment operation and maintenance. Underground cable trenches, as key infrastructure of urban power distribution networks, have internal environmental parameters such as humidity, temperature, and partial discharge activity that significantly impact cable insulation conditions. In recent years, cable trench environmental monitoring systems based on Internet of Things (IoT) sensing technology have been widely applied. By deploying temperature and humidity sensors, gas detectors, and video monitoring equipment, continuous acquisition and remote transmission of basic environmental indicators within the cable trench have been achieved. Simultaneously, monitoring technologies for cable partial discharge have also gradually developed. Some systems use high-frequency current transformers (HFCTs) to capture pulse signals on the grounding wire to assess the activity level of insulation defects. This constitutes the mainstream technical approach for current cable trench environmental management.

[0003] Existing technologies still have room for improvement in multi-source information fusion and dynamic environmental response. On the one hand, traditional temperature and humidity measurements are easily affected by local air convection and thermal radiation caused by cable heating, resulting in relative humidity readings deviating from the true microenvironment on the cable surface, thus affecting the accuracy of environmental assessment. On the other hand, pulse current signals are easily affected by electromagnetic interference and loop impedance characteristics during transmission. If only conventional filtering and peak detection methods are used, it is difficult to accurately restore the quantitative characteristics of partial discharge, which limits the linkage between insulation risk assessment and environmental control. Summary of the Invention

[0004] In view of the aforementioned existing problems, the present invention is proposed.

[0005] Therefore, this invention provides a cable operation environment optimization method based on cable trench information to solve the problems of insufficient accuracy in environmental parameter measurement and poor linkage between insulation risk assessment and environmental control.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution:

[0007] This invention provides a method for optimizing the cable operating environment based on information collected from cable trenches. The method includes: collecting raw parameter datasets inside the cable trench; collecting pulse current signals at the cable grounding connection points and generating a standard pulse discharge quantity sequence through bandpass filtering and amplitude integration conversion; calculating a compensated relative humidity value based on the original relative humidity measurement and ambient temperature value using a multi-scale heat conduction-diffusion coupling equation; calculating a cable insulation risk index using a quantum annealing algorithm based on the compensated relative humidity value and the standard pulse discharge quantity sequence; generating a high-risk judgment flag when the cable insulation risk index exceeds a risk threshold; and optimizing the cable trench environment based on the high-risk judgment flag using intelligent louvers driven by shape memory alloys and a ventilation fan.

[0008] As a preferred embodiment of the cable operating environment optimization method based on cable trench information as described in this invention, the original parameter dataset includes original relative humidity measurements, ambient temperature values, ambient temperature change rate, historical fault data, and cable load current.

[0009] As a preferred embodiment of the cable operating environment optimization method based on cable trench information collection described in this invention, the step of collecting pulse current signals at the cable grounding wire connection point and generating a standard pulse discharge quantity sequence through bandpass filtering and amplitude integration conversion is as follows:

[0010] The original pulse current signal is acquired, quantum state bandpass filtering is performed, the effective pulse signal is extracted, and the signal-to-noise ratio of the effective pulse signal is improved to generate an enhanced pulse signal.

[0011] The enhanced pulse signal is processed by topological invariant integration to generate a standard pulse discharge quantity sequence.

[0012] As a preferred embodiment of the cable operating environment optimization method based on cable trench information described in this invention, the step of processing the enhanced pulse signal through topological invariant integration to generate a standard pulse discharge quantity sequence includes the following specific steps.

[0013] The enhanced pulse signal is analyzed by Hilbert transform to extract the instantaneous amplitude envelope and phase gradient;

[0014] The electromagnetic gauge manifold is constructed based on the instantaneous amplitude envelope and phase gradient, and the Chern-Simons invariants are obtained on the electromagnetic gauge manifold.

[0015] The Chern-Simons invariant is corrected, and the calibrated single-pulse discharge quantity is output.

[0016] The calibrated single-pulse discharge quantity is time-domain aligned and normalized to generate a standard pulse discharge quantity sequence.

[0017] As a preferred embodiment of the cable operating environment optimization method based on cable trench information as described in this invention, the step of calculating the compensated relative humidity value based on the original relative humidity measurement value and ambient temperature value using a multi-scale heat conduction-diffusion coupling equation is as follows:

[0018] Calculate the thermal radiation power on the cable surface and, in conjunction with the thermal absorption characteristics of the sensor, obtain the thermal radiation interference.

[0019] Based on the original relative humidity measurement, ambient temperature value, and adsorption kinetics of water molecules and sensor sensitive materials, the molecular-level humidity measurement deviation is corrected to obtain the molecular adsorption correction amount.

[0020] By using a multi-scale thermal conduction-diffusion coupling equation, the thermal radiation interference and molecular adsorption correction are integrated to generate a compensated relative humidity value.

[0021] As a preferred embodiment of the cable operating environment optimization method based on cable trench information described in this invention, the step of generating a compensated relative humidity value by fusing thermal radiation interference and molecular adsorption correction through a multi-scale heat conduction-diffusion coupling equation is as follows:

[0022] Based on the thermal radiation interference and molecular adsorption correction, a non-Fourier heat conduction-diffusion control equation is constructed, and the non-Fourier heat conduction-diffusion control equation is discretized in three-dimensional space to obtain the dynamic distribution of the three-dimensional temperature field in the cable trench.

[0023] Based on the dynamic distribution of the three-dimensional temperature field within the cable trench, the sensor temperature drift error is calculated, and a compensated relative humidity value is generated by combining the molecular adsorption correction amount.

[0024] As a preferred embodiment of the cable operating environment optimization method based on cable trench information described in this invention, the specific steps for calculating the cable insulation risk index using a quantum annealing algorithm based on the compensated relative humidity value and standard pulse discharge quantity sequence are as follows:

[0025] A humidity gradient field is constructed based on the compensated relative humidity value, and the spatiotemporal distribution characteristics of the standard pulse discharge quantity sequence are extracted at the same time.

[0026] The humidity gradient field and its spatiotemporal distribution characteristics are fused into a four-dimensional spatiotemporal tensor field, and the ground state energy of the humidity-partial discharge coupled Hamiltonian is solved by the quantum annealing algorithm.

[0027] The ground state energy is mapped to a topological risk field and then smoothed using Gaussian kernels to normalize and generate a cable insulation risk index.

[0028] As a preferred embodiment of the cable operating environment optimization method based on cable trench information collection described in this invention, the step of generating a high-risk judgment flag when the cable insulation risk index is greater than the risk threshold is as follows:

[0029] The cable insulation risk index is input into a quantum rotating door-enhanced long short-term memory network, which outputs a risk threshold.

[0030] By comparing the cable insulation risk index with the risk threshold, a high-risk judgment indicator is generated when the cable insulation risk index continuously exceeds the risk threshold.

[0031] As a preferred embodiment of the cable operating environment optimization method based on cable trench information as described in this invention, the specific steps for inputting the cable insulation risk index into a quantum rotating door-enhanced long short-term memory network and outputting a risk threshold are as follows:

[0032] The cable insulation risk index, ambient temperature change rate, cable load current and historical fault sequence are integrated into a multi-dimensional feature vector.

[0033] Multidimensional feature vectors are mapped to quantum state representations through a quantum state embedding layer. The quantum state representations are then processed by forward propagation using a quantum rotation gate-enhanced long short-term memory network to generate a basic risk threshold.

[0034] The risk threshold is obtained by physically correcting the basic risk threshold based on thermodynamic entropy change constraints.

[0035] As a preferred embodiment of the cable operating environment optimization method based on cable trench information collection described in this invention, the method optimizes the cable trench environment based on high-risk assessment indicators, using intelligent louvers driven by shape memory alloys and a ventilation fan. The specific steps are as follows:

[0036] Based on the status of the high-risk assessment indicator, a shape memory alloy driving current control signal is generated;

[0037] The opening and closing angle of the shape memory alloy louver is adjusted by using the driving current control signal of the shape memory alloy.

[0038] Based on the real-time humidity change rate and the opening and closing angle of the shape memory alloy louvers, the target rotation speed of the ventilator is calculated, the optimal turbulence intensity is generated, and the cable operating environment is optimized.

[0039] The beneficial effects of this invention are as follows: By introducing topological invariant integral operations based on Hilbert transform and Chen-Simons invariants to process pulse current signals, high anti-interference and physically interpretable quantification of partial discharge quantity are achieved, significantly improving the accuracy and consistency of discharge feature extraction; by constructing a multi-scale heat conduction-diffusion coupling equation to perform dual compensation for the dynamic measurement deviation of temperature and humidity sensors at the thermal radiation and molecular adsorption levels, the problem of humidity measurement distortion caused by periodic temperature changes in cable trenches is effectively overcome, greatly improving the reliability of environmental condition perception, and forming an integrated closed-loop optimization mechanism of "perception-assessment-decision-execution". This not only improves the accuracy and predictability of cable insulation condition assessment, but also realizes adaptive adjustment of the operating environment, effectively enhancing the operational safety, stability and intelligent operation and maintenance level of power cable systems. Attached Figure Description

[0040] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0041] Figure 1 This is a flowchart of a cable operating environment optimization method based on information collected from cable trenches.

[0042] Figure 2 A flowchart for generating a standard pulse discharge quantity sequence.

[0043] Figure 3 This is a flowchart for calculating the compensated relative humidity value.

[0044] Figure 4 A flowchart for calculating the cable insulation risk index. Detailed Implementation

[0045] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0046] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0047] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.

[0048] Reference Figures 1-4 This is one embodiment of the present invention, which provides a method for optimizing the cable operating environment based on cable trench information, including the following steps:

[0049] S1. Collect raw parameter datasets inside the cable trench.

[0050] Specifically, the raw parameter dataset includes raw relative humidity measurements, ambient temperature values, ambient temperature change rate, historical fault data, and cable load current.

[0051] S2. Collect pulse current signals at the cable grounding connection point, and generate a standard pulse discharge quantity sequence through bandpass filtering and amplitude integration conversion.

[0052] S2.1. Acquire the original pulse current signal, perform quantum state bandpass filtering, extract the effective pulse signal, and perform signal-to-noise ratio enhancement processing on the effective pulse signal to generate an enhanced pulse signal.

[0053] Specifically, the raw pulse current signal is acquired at the cable grounding connection point using a high-frequency current transformer. Quantum state bandpass filtering is then performed on the raw pulse current signal: the passband is limited to the example range of 1-30MHz to suppress power frequency interference and high-frequency noise, extracting the effective pulse signal. The effective pulse signal generated by the quantum state bandpass filtering is then reconstructed in phase space. Based on the time delay parameter determined by the first zero-crossing point of the autocorrelation function (example value 0.2 milliseconds), a seven-dimensional embedding vector sequence is constructed from the pulse signal time-series data according to the time delay step.

[0054] Perform fractal feature quantization operation: calculate the Hausdorff dimension of the embedded vector distribution in the phase space by logarithmic scanning of the coverage radius, and obtain the box dimension by statistically analyzing the number of non-empty grids through fixed-size grid subdivision, and generate the ratio of the two as a fractal complexity index.

[0055] Nonlinear enhancement operation: Based on the fractal complexity index value, when the fractal complexity index value exceeds the example value of 1.05, the amplitude value of the original pulse signal is amplified by proportional gain (example gain coefficient 0.6). The proportional gain value increases linearly with the fractal complexity index value. When the fractal complexity index value does not reach the example value of 1.05, the signal amplitude remains unchanged, and the enhanced pulse signal with the timestamp perfectly aligned with the original signal is output.

[0056] S2.2. The enhanced pulse signal is processed by topological invariant integration to generate a standard pulse discharge quantity sequence.

[0057] The beneficial effects are as follows: The phase gradient and amplitude envelope of the pulse signal are analytically enhanced through Hilbert transform; Chern-Simons topological invariants are extracted from the constructed electromagnetic gauge field manifold; and electromagnetic interference distortion is automatically corrected using the differential homeomorphism invariance of quantum numbers, outputting a standard pulse discharge quantity sequence. Existing techniques (such as wavelet denoising / peak integration) only analyze the local statistical characteristics of the signal, mainly focusing on the analysis of local statistical properties. This leads to an inability to accurately distinguish between real discharge signals and pseudo signals caused by electromagnetic field distortion under complex background noise or electromagnetic interference, and usually lacks the ability to identify the overall topological structure of the signal, resulting in limitations when processing pulse signals with complex background noise. In contrast, the method based on Hilbert transform and combined with electromagnetic gauge field theory provides a more refined and accurate analytical approach. By extracting Chern-Simons topological invariants, the essential characteristics of the pulse signal can be better captured. Even in the presence of a large amount of electromagnetic interference, it can effectively distinguish between real discharge signals and pseudo signals. The use of the differential homeomorphism invariance of quantum numbers to automatically correct electromagnetic interference distortion further improves the accuracy and reliability of the analysis results.

[0058] S2.2.1. Perform Hilbert transform analysis on the enhanced pulse signal to extract the instantaneous amplitude envelope and phase gradient.

[0059] Specifically, the process of generating an analytic signal by performing a Hilbert transform on the enhanced pulse signal is as follows: the enhanced pulse signal is convolved with the Hilbert transform kernel function to obtain the Hilbert transform result; the enhanced pulse signal is used as the real part and the Hilbert transform result is used as the imaginary part to construct a complex analytic signal; the instantaneous amplitude envelope (the magnitude of the analytic signal in the complex plane) and the instantaneous phase function (the principal argument value) are extracted from the analytic signal; the instantaneous phase function is differentiated in time (for example, using the central difference method with a time step of 1 nanosecond) to obtain the phase gradient.

[0060] S2.2.2. Construct an electromagnetic gauge manifold based on the instantaneous amplitude envelope and phase gradient, and obtain Chern-Simons invariants on the electromagnetic gauge manifold.

[0061] Specifically, using the instantaneous amplitude envelope and phase gradient obtained by Hilbert transform, an electromagnetic gauge field manifold is constructed in three-dimensional space in a parametric form. The instantaneous amplitude envelope serves as the radial coordinate parameter, and the phase gradient serves as the angular coordinate parameter. The origin coordinates of the local coordinate system are defined using the pulse signal timestamp. Through differential geometry operations, a phase-amplitude coupled differential form is constructed based on the gradient vector of the instantaneous amplitude envelope and the curl tensor of the phase gradient. This phase-amplitude coupled differential form is then integrated along a closed path on the electromagnetic gauge field manifold, specifically along an isophase gradient line to obtain the initial values ​​of the Chern-Simons invariants. When a non-steady-state abrupt change in the phase gradient occurs (e.g., the phase gradient change exceeds the example value by 0.2 radians / second), the path integration range is dynamically scaled in the direction of maximum amplitude envelope gradient to cover the range of ±3 standard deviations of the example value. A threshold detector identifies abnormal pulses in the integration result caused by electromagnetic interference (e.g., abrupt changes exceeding 50% of the average integral value). Linear interpolation correction is performed on the interval containing the abnormal points, and the Chern-Simons invariants are output.

[0062] S2.2.3. Perform bias correction on the Chern-Simons invariant and output the calibrated single-pulse discharge quantity.

[0063] Specifically, the process of correcting the Chern-Simons invariant and outputting the calibrated single-pulse discharge quantity is as follows: After obtaining the path integral result of the Chern-Simons invariant (this result comes from the initial value generated by integrating the electromagnetic gauge field manifold based on the instantaneous amplitude envelope and phase gradient along a closed path), the Chern-Simons invariant integral sequence is analyzed using a threshold detector. By obtaining the moving average of the integral result and setting a relative change threshold (e.g., the relative change threshold is set to a deviation exceeding 50% of the example value of the moving average), abnormal pulse points caused by electromagnetic interference are identified. When an abnormal point is detected (e.g., a specific abnormal point is a point where the integral value suddenly exceeds 50% of the moving average), the integral value of the neighboring point in the time zone of the abnormal point is extracted (the neighboring point is taken from the point within a 1-millisecond interval of the example values ​​before and after the abnormal point that does not exceed the threshold). Linear interpolation is performed based on the integral value of the neighboring point to construct a univariate linear interpolation function in the time zone of the abnormal point. The univariate linear interpolation function corrects the integral result of the Chern-Simons invariant at the abnormal point. The corrected integral value is used as the calibrated single-pulse discharge quantity.

[0064] S2.2.4. Perform time-domain alignment and normalization on the calibrated single-pulse discharge quantity to generate a standard pulse discharge quantity sequence.

[0065] Specifically, the calibrated single-pulse discharge quantity is obtained, and the local maximum point in the waveform of each calibrated single-pulse discharge quantity is located using a peak detection algorithm. Using the time of the local maximum point as a reference, the origin of the time coordinate of each calibrated single-pulse discharge quantity is aligned to the maximum point (time scale zero relocation). After completing the time-domain alignment, the local maximum values ​​of all calibrated single-pulse discharge quantities are extracted to obtain the global maximum value (for example, the highest value of all pulse peaks in a certain measurement, taken as 8.72 pC). Normalization is performed based on the global maximum value. The quotient of all data points of each calibrated single-pulse discharge quantity and the global maximum value (generating a dimensionless proportional value with a value range of 0 to 1 in the example) is rearranged according to the original time order of pulse occurrence to form a standard pulse discharge quantity sequence.

[0066] S3. Based on the original relative humidity measurement and ambient temperature value, the compensated relative humidity value is calculated through the multi-scale heat conduction-diffusion coupling equation.

[0067] S3.1. Calculate the thermal radiation power on the cable surface, and combine it with the thermal absorption characteristics of the sensor to obtain the thermal radiation interference.

[0068] Specifically, the thermal radiation power on the cable surface is calculated according to the Stefan-Boltzmann law; based on the thermal absorption characteristics of the sensor, the specific heat capacity and mass value of the sensor's sensitive unit are extracted; the temperature rise caused by the sensor absorbing thermal radiation per unit time is calculated; and the product of the temperature rise and the sensor's temperature-humidity cross-sensitivity coefficient is used as the thermal radiation interference.

[0069] It should be noted that the expression for calculating the surface thermal radiation power of the cable is:

[0070]

[0071] Where P represents the thermal radiation power of the cable surface, ∈ represents the emissivity of the cable surface, σ represents the Stefan-Boltzmann constant, A represents the unit surface area of ​​the cable, and T cab T represents the measured surface temperature of the cable. env This indicates the ambient temperature of the cable trench.

[0072] The expression for calculating the temperature rise caused by the sensor absorbing thermal radiation per unit time is:

[0073]

[0074] Where ΔT represents the temperature rise caused by the sensor absorbing thermal radiation per unit time, α represents the sensor's ability to absorb thermal radiation (dimensionless), Δt represents the time interval, m represents the mass of the sensor, and C represents the specific heat capacity of the sensor material.

[0075] S3.2. Based on the original relative humidity measurement value, ambient temperature value, and the adsorption kinetics of water molecules and sensor sensitive materials, the molecular-level humidity measurement deviation is corrected to obtain the molecular adsorption correction amount.

[0076] Specifically, the mass change of the sensor's sensitive material surface is measured in real time using a quartz crystal microbalance. Combined with the molar mass of water molecules and Avogadro's constant, the number of water molecules adsorbed is obtained. Based on the characteristic parameters of the sensitive material, the adsorption activation energy and molecular collision frequency factor are extracted to obtain the dynamic adsorption coverage. Based on the difference between the dynamic adsorption coverage and the equilibrium adsorption coverage, the molecular-level adsorption hysteresis is obtained. Based on the product of the molecular-level adsorption hysteresis and the humidity response coefficient per unit coverage of the sensitive material, the molecular adsorption correction is generated.

[0077] S3.3. By using the multi-scale thermal conduction-diffusion coupling equation, the thermal radiation interference and molecular adsorption correction are integrated to generate the compensated relative humidity value.

[0078] S3.3.1. Based on the thermal radiation interference and molecular adsorption correction, a non-Fourier heat conduction-diffusion control equation is constructed, and the non-Fourier heat conduction-diffusion control equation is discretized in three-dimensional space to obtain the dynamic distribution of the three-dimensional temperature field in the cable trench.

[0079] Specifically, the steps for constructing the non-Fourier heat conduction-diffusion control equation based on the numerical values ​​of thermal radiation interference and molecular adsorption correction are as follows: the thermal radiation interference is introduced into the heat conduction term as an additional heat source term, and the molecular adsorption correction is added to the mass diffusion term as a diffusion coefficient correction term to form the non-Fourier heat conduction-diffusion control equation.

[0080] The non-Fourier heat conduction-diffusion control equations are discretized in three dimensions: a three-dimensional grid coordinate system for the cable trench is established (the grid spacing is 0.1m in the example), the spatial differential terms of the control equations are discretized using the finite volume method, the transient terms are solved iteratively using the second-order Runge-Kutta method, and the time step is advanced by combining the boundary conditions (measured values ​​of cable surface temperature and trench wall insulation conditions), and the temperature value of each grid point is output to generate the dynamic distribution of the three-dimensional temperature field in the cable trench.

[0081] It should be noted that the expression for the non-Fourier heat conduction-diffusion control equation is:

[0082]

[0083] Where ρ represents air density, τ q The thermal relaxation time is represented by k, the thermal conductivity by D, and the diffusion coefficient by δ. ads The value represents the molecular adsorption correction amount, B represents the water vapor concentration, and Q represents the molecular adsorption correction amount. radThe value represents the amount of thermal radiation interference, where E represents the temperature field and e represents time. This represents the first-order partial derivative of temperature with respect to time. This represents the second partial derivative of temperature with respect to time. This represents the vector differential operator.

[0084] S3.3.2. Based on the dynamic distribution of the three-dimensional temperature field in the cable trench, calculate the sensor temperature drift error, and generate a compensated relative humidity value by combining the molecular adsorption correction amount.

[0085] Specifically, the spatial coordinates of the humidity sensor in the three-dimensional grid coordinate system of the cable trench are located, and the real-time temperature value of the spatial coordinates is extracted; the temperature drift error is calculated; and the compensated relative humidity value is generated by combining the molecular adsorption correction amount with the original relative humidity measurement value.

[0086] It should be noted that the expression for calculating the sensor temperature drift error is as follows:

[0087] ΔF = η·(Gg);

[0088] Where ΔF represents the temperature drift error in humidity measurement caused by temperature, η represents the temperature drift coefficient of humidity sensor, G represents the current real-time ambient temperature, and g represents the reference temperature when the sensor was calibrated at the factory.

[0089] The expression for generating the compensated relative humidity value is:

[0090] H = H raw +ΔF+δ ads ;

[0091] Where H represents the compensated relative humidity value, H raw This represents the original relative humidity measurement value.

[0092] S4. Using the compensated relative humidity value and standard pulse discharge quantity sequence, the cable insulation risk index is calculated by quantum annealing algorithm.

[0093] The beneficial effects are as follows: By fusing the compensated humidity gradient field with the spatiotemporal distribution of pulsed discharge into a four-dimensional spatiotemporal tensor field, a humidity-partial discharge coupled Hamiltonian is constructed. The ground state energy is solved using the quantum annealing algorithm and mapped to a topological risk field, generating a cable insulation risk index that essentially characterizes the evolution law of insulation defects. Existing technologies (such as fuzzy logic / support vector machines) only linearly weight isolated parameters (temperature, humidity, partial discharge amplitude), which cannot describe the spatiotemporal coupling mechanism of humidity diffusion and partial discharge, ignore the nonlinear guiding effect of humidity gradient on discharge path, and lack spatiotemporal correlation modeling of partial discharge, leading to misjudgment of intermittent discharge risk. In comparison, by modeling humidity and partial discharge in more detail and using more advanced algorithms (such as the quantum annealing algorithm), the risk of cable insulation can be predicted more accurately. Especially for problems caused by the interaction of humidity diffusion and partial discharge, it not only improves the accuracy of cable insulation status assessment, but also helps to identify potential faults in advance, thereby taking preventive measures to avoid accidents.

[0094] S4.1. Construct a humidity gradient field based on the compensated relative humidity value, and extract the spatiotemporal distribution characteristics from the standard pulse discharge quantity sequence.

[0095] Specifically, a humidity gradient field is constructed based on the compensated relative humidity values: a three-dimensional grid coordinate system is constructed for the cable trench. For the compensated relative humidity value of each grid point, the rate of change in the three spatial directions is obtained. For internal grid points, the gradient value in the X direction is obtained by taking the difference between the compensated relative humidity values ​​of the grid points to the left and right of the internal grid point and the quotient of twice the grid spacing. The gradient value in the Y direction is obtained by taking the difference between the compensated relative humidity values ​​of the grid points in front of and behind the internal grid point and the quotient of twice the grid spacing. The gradient value in the Z direction is obtained by taking the difference between the compensated relative humidity values ​​of the grid points below and above the internal grid point and the quotient of twice the grid spacing. For boundary grid points, the gradient value in the boundary direction is generated by taking the difference between the compensated relative humidity values ​​of the boundary grid point and the adjacent single-sided grid point and the quotient of the grid spacing. The gradient values ​​in the X, Y, and Z directions of each grid point are combined into a three-dimensional vector to form humidity gradient field data reflecting the spatial humidity variation characteristics, thus generating the humidity gradient field.

[0096] Extracting the spatiotemporal distribution characteristics of the standard pulse discharge quantity sequence: By analyzing the characteristics of the pulse amplitude value changing with time and the distribution characteristics of the pulse amplitude value in the cable trench spatial coordinates in the standard pulse discharge quantity sequence, the cluster density and trend characteristic values ​​of the event timestamps, spatial location coordinates and amplitude distribution of the standard pulse discharge quantity sequence are statistically analyzed to obtain the spatiotemporal distribution characteristics.

[0097] S4.2. The humidity gradient field and spatiotemporal distribution characteristics are fused into a four-dimensional spatiotemporal tensor field, and the ground state energy of the humidity-partial discharge coupled Hamiltonian is solved by the quantum annealing algorithm.

[0098] Specifically, the humidity gradient field includes spatial gradient change data of the compensated relative humidity value at three-dimensional coordinate points in the cable trench, and the spatiotemporal distribution features include pulse event timestamps, spatial location coordinates, and cluster density and trend feature data of the corresponding amplitude distribution recorded in the standard pulse discharge quantity sequence. The spatial gradient change data in the humidity gradient field and the timestamp coordinate data in the spatiotemporal distribution features are spatiotemporally grid aligned and matched to generate a structured data set containing humidity gradient and pulse discharge amplitude distribution density in a four-dimensional coordinate system, forming a four-dimensional spatiotemporal tensor field.

[0099] The humidity-partial discharge coupled Hamiltonian is composed of a Schrödinger-type operator that reflects the diffusion effect of the humidity gradient field and a topological invariant operator that describes the spatiotemporal distribution characteristics of pulsed discharge. The quantum annealing algorithm obtains the ground state eigenvalue function and outputs the ground state energy by gradually adjusting the transverse magnetic field strength during the adiabatic evolution process.

[0100] S4.3. Map the ground state energy to a topological risk field, perform Gaussian kernel smoothing, and normalize to generate the cable insulation risk index.

[0101] Specifically, the ground-state energy, as a scalar field data, is distributed on the three-dimensional spatial coordinates of the cable trench. The ground-state energy is converted into a topological invariant density distribution value through an exponential mapping function, forming a topological risk field. The topological risk field is then smoothed using Gaussian kernel convolution: a three-dimensional Gaussian function with a standard deviation parameter of 0.5 meters is used to calculate a weighted average of each spatial point in the topological risk field, eliminating local noise fluctuations. The smoothed topological risk field data is then processed using a linear normalization function, compressing the entire field data to the [0,1] interval. Historical operating data is extracted from the topological risk field, and the maximum value is used as the upper limit benchmark value for normalization, while the minimum value is used as the lower limit benchmark value. The values ​​of each point in the current topological risk field are scaled proportionally to output the cable insulation risk index.

[0102] S5. When the cable insulation risk index is greater than the risk threshold, a high-risk judgment mark is generated.

[0103] S5.1. Input the cable insulation risk index into the quantum rotating door-enhanced long short-term memory network and output the risk threshold.

[0104] S5.1.1. The cable insulation risk index, ambient temperature change rate, cable load current and historical fault sequence are integrated into a multi-dimensional feature vector.

[0105] Specifically, the cable insulation risk index is generated by extracting the arithmetic mean and standard deviation coefficient of the values ​​at each spatial point to generate a cable insulation risk index sub-vector; the ambient temperature change rate is directly used as the temperature change feature scalar; the cable load current is directly used as the current load feature scalar; the historical fault sequence is generated by counting the frequency of fault events in the last thirty days to generate a historical fault feature scalar; the four types of data are combined according to fixed dimensions: the first dimension is the cable insulation risk index sub-vector, the second dimension is the temperature change feature scalar, the third dimension is the current load feature scalar, and the fourth dimension is the historical fault feature scalar to generate a four-dimensional feature vector.

[0106] S5.1.2. Multidimensional feature vectors are mapped to quantum state representations through a quantum state embedding layer. A quantum rotation gate-enhanced long short-term memory network is used to process the quantum state representations through forward propagation, generating a basic risk threshold.

[0107] Specifically, the process of mapping a four-dimensional feature vector to a quantum state representation through a quantum state embedding layer is as follows: the four-dimensional feature vector is input into a qubit register for amplitude encoding, and a four-dimensional phase space distribution is generated by using quantum phase angle parameterization operations to obtain the quantum state representation;

[0108] The process of forward propagation processing of quantum state representation using a quantum rotation gate-enhanced long short-term memory network is as follows: The quantum state representation is input into the quantum rotation gate-enhanced long short-term memory network, and the quantum state is controlled by parameterized unitary transformation through the quantum rotation gate unit. The quantum state is then processed by the classical unit of the quantum rotation gate-enhanced long short-term memory network, and the time-series state feature vector after quantum control is generated through the linear output layer.

[0109] It should be noted that the training process of the quantum rotating door-enhanced long short-term memory network is as follows: a four-dimensional feature vector is used as the input data for training samples (including historical cable insulation risk index sub-vectors, historical ambient temperature change rate, historical cable load current, and historical fault feature scalars), and corresponding real risk threshold values ​​(historical fault record judgment) are labeled; the quantum rotating door angle parameters are optimized by the quantum gradient descent algorithm, and the weight parameters of the quantum rotating door-enhanced long short-term memory network are optimized by the stochastic gradient descent algorithm, resulting in the trained quantum rotating door-enhanced long short-term memory network.

[0110] S5.1.3. Based on the thermodynamic entropy change constraint, the basic risk threshold is physically corrected to obtain the risk threshold.

[0111] Specifically, the process of physically correcting the basic risk threshold based on thermodynamic entropy change constraints to obtain the risk threshold is as follows: obtain the cable heat generation power through Joule's law, obtain the real-time thermodynamic entropy generation rate through the ratio of the average temperature of the cable trench to the cable heat generation power; at the same time, obtain the ideal entropy production rate reference value based on Carnot cycle theory; generate the entropy change constraint correction term through the conversion coefficient calibrated by the constant entropy change control experiment; and use the sum of the entropy change constraint correction term and the basic risk threshold as the risk threshold.

[0112] S5.2. Compare the cable insulation risk index with the risk threshold. When the cable insulation risk index continuously exceeds the risk threshold, a high-risk judgment mark is generated.

[0113] Specifically, the process of comparing the cable insulation risk index with the risk threshold and generating a high-risk judgment mark is as follows: On the timestamp sequence, the cable insulation risk index and the risk threshold are read synchronously at each moment. When the cable insulation risk index is greater than the risk threshold within 3 consecutive time windows (with a window length of 3 seconds in the example), the current area is determined to be a high-risk area, and a high-risk judgment mark is generated. If the duration of the cable insulation risk index is less than 3 seconds of the example value, or the time window is not continuous, a high-risk judgment mark is not generated.

[0114] S6. Based on high-risk assessment criteria, the cable trench environment is optimized by using intelligent louvers and ventilators driven by shape memory alloys.

[0115] S6.1. Generate a shape memory alloy drive current control signal based on the status of the high-risk judgment mark.

[0116] Specifically, the following steps are performed based on the high-risk determination flag status: read the current compensated relative humidity value and ambient temperature value, and map the compensated relative humidity value to the equivalent temperature value; collect the inherent parameters of the shape memory alloy, including the initial target temperature rise of the martensitic phase transformation, the resistivity of the alloy wire, the specific heat capacity of the alloy wire, the mass density of the alloy wire, and the cross-sectional area of ​​the alloy wire; calculate the driving current intensity value; collect the cumulative time value of the high-risk determination flag being continuously true; and use it as the driving current duration value; generate a shape memory alloy driving current control signal containing the current intensity value and the duration value.

[0117] It should be noted that the expression for calculating the driving current intensity is as follows:

[0118]

[0119] Where I represents the driving current intensity, O represents the cross-sectional area of ​​the alloy wire, Q represents the mass density of the alloy wire, X represents the specific heat capacity of the alloy wire, and ΔTZ eff The target temperature rise for the initiation of the martensitic phase transformation is represented by Y, which represents the resistivity of the alloy wire, and z represents the target temperature rise for the initiation of the martensitic phase transformation. reqThis indicates the duration of the drive current.

[0120] S6.2. The opening and closing angle of the shape memory alloy louvers is adjusted by using the shape memory alloy driving current control signal.

[0121] Specifically, the process of adjusting the opening and closing angle of a shape memory alloy (SMI) louver using a shape memory alloy driving current control signal is as follows: The current intensity value in the shape memory alloy driving current control signal is applied to the driving alloy wire of the shape memory alloy louver, heating the shape memory alloy to the austenitic phase transformation temperature through the Joule heating effect, triggering a change in the length of the shape memory alloy; based on the geometric transmission ratio between the shape memory alloy length change rate and the louver linkage mechanism, the change in the opening and closing angle of the shape memory alloy louver is obtained; the opening and closing angle value of the shape memory alloy louver is updated in real time until the duration value in the shape memory alloy driving current control signal ends, completing the adjustment operation.

[0122] S6.3. Based on the real-time humidity change rate and the opening and closing angle of the shape memory alloy louvers, calculate the target speed of the ventilator, generate the optimal turbulence intensity, and optimize the cable operating environment.

[0123] Specifically, based on the real-time humidity change rate and the current shape memory alloy louver opening angle, a target speed value is generated according to the target speed calculation formula of the ventilator. The target speed value is converted into a ventilator drive voltage signal through a proportional-integral controller, so that the ventilator operates at the target speed value. When the speed exceeds the example value of 1000 rpm, the optimal turbulence intensity with a Reynolds number greater than the example value of 4000 is automatically generated to complete the environmental optimization.

[0124] It should be noted that the numerical expression for calculating the target rotational speed is:

[0125]

[0126] Where, N target This indicates the target rotational speed of the fan, where M represents the humidity change rate minus the rotational speed gain coefficient. θ represents the relative humidity change rate, θ represents the angle between the fan outlet direction and the main monitoring area of ​​the sensor, and N0 represents the base maintenance speed.

[0127] This embodiment also provides a computer device applicable to the cable operating environment optimization method based on cable trench information collection, including: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to implement the cable operating environment optimization method based on cable trench information collection proposed in the above embodiment.

[0128] The computer device can be a terminal, comprising a processor, memory, communication interface, display screen, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, carrier networks, NFC (Near Field Communication), or other technologies. The display screen can be an LCD screen or an e-ink screen. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad on the computer device's casing, or an external keyboard, touchpad, or mouse.

[0129] This embodiment also provides a storage medium storing a computer program. When executed by a processor, the program implements the cable operating environment optimization method based on cable trench information collection proposed in the above embodiments. The storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read Only Memory (EPROM), Programmable Red-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0130] In summary, this invention achieves high anti-interference capability and physically interpretable quantification of partial discharge quantity by introducing topological invariant integration operations based on Hilbert transform and Chern-Simons invariants to process pulse current signals, significantly improving the accuracy and consistency of discharge feature extraction. Furthermore, by constructing a multi-scale heat conduction-diffusion coupling equation to compensate for the dynamic measurement deviation of temperature and humidity sensors at both the thermal radiation and molecular adsorption levels, it effectively overcomes the humidity measurement distortion caused by periodic temperature changes within cable trenches, greatly improving the reliability of environmental condition perception. This forms an integrated closed-loop optimization mechanism of "perception-assessment-decision-execution," not only improving the accuracy and predictability of cable insulation condition assessment but also enabling adaptive adjustment of the operating environment, effectively enhancing the operational safety, stability, and intelligent operation and maintenance level of power cable systems.

[0131] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A method for optimizing the cable operating environment based on information collected from cable trenches, characterized in that: include, Collect raw parameter datasets inside the cable trench; Pulse current signals are collected at the cable grounding connection point, and standard pulse discharge quantity sequences are generated through bandpass filtering and amplitude integration conversion. Based on the original relative humidity measurement and ambient temperature value, the compensated relative humidity value is calculated through the multi-scale heat conduction-diffusion coupling equation. The cable insulation risk index is calculated using the compensated relative humidity value and standard pulse discharge quantity sequence through the quantum annealing algorithm. When the cable insulation risk index is greater than the risk threshold, a high-risk judgment mark is generated; Based on high-risk assessment criteria, intelligent louvers and ventilators driven by shape memory alloys are used to optimize the environment of cable trenches.

2. The cable operating environment optimization method based on cable trench information as described in claim 1, characterized in that: The raw parameter dataset includes raw relative humidity measurements, ambient temperature values, ambient temperature change rate, historical fault data, and cable load current.

3. The cable operating environment optimization method based on cable trench information as described in claim 2, characterized in that: The process involves acquiring pulse current signals at the cable grounding connection point, and generating a standard pulse discharge quantity sequence through bandpass filtering and amplitude integration conversion. The specific steps are as follows: The original pulse current signal is acquired, quantum state bandpass filtering is performed, the effective pulse signal is extracted, and the signal-to-noise ratio of the effective pulse signal is improved to generate an enhanced pulse signal. The enhanced pulse signal is processed by topological invariant integration to generate a standard pulse discharge quantity sequence.

4. The cable operating environment optimization method based on cable trench information as described in claim 3, characterized in that: The process of enhancing the pulse signal through topological invariant integration to generate a standard pulse discharge quantity sequence involves the following steps: The enhanced pulse signal is analyzed by Hilbert transform to extract the instantaneous amplitude envelope and phase gradient; The electromagnetic gauge manifold is constructed based on the instantaneous amplitude envelope and phase gradient, and the Chern-Simons invariants are obtained on the electromagnetic gauge manifold. The Chern-Simons invariant is corrected, and the calibrated single-pulse discharge quantity is output. The calibrated single-pulse discharge quantity is time-domain aligned and normalized to generate a standard pulse discharge quantity sequence.

5. The cable operating environment optimization method based on cable trench information as described in claim 4, characterized in that: The method involves calculating the compensated relative humidity value based on the original relative humidity measurement and ambient temperature value using a multi-scale heat conduction-diffusion coupling equation. The specific steps are as follows: Calculate the thermal radiation power on the cable surface and, in conjunction with the thermal absorption characteristics of the sensor, obtain the thermal radiation interference. Based on the original relative humidity measurement, ambient temperature value, and adsorption kinetics of water molecules and sensor sensitive materials, the molecular-level humidity measurement deviation is corrected to obtain the molecular adsorption correction amount. By using a multi-scale thermal conduction-diffusion coupling equation, the thermal radiation interference and molecular adsorption correction are integrated to generate a compensated relative humidity value.

6. The cable operating environment optimization method based on cable trench information as described in claim 5, characterized in that: The method involves using a multi-scale thermal conduction-diffusion coupling equation to fuse thermal radiation interference and molecular adsorption correction to generate a compensated relative humidity value. The specific steps are as follows: Based on the thermal radiation interference and molecular adsorption correction, a non-Fourier heat conduction-diffusion control equation is constructed, and the non-Fourier heat conduction-diffusion control equation is discretized in three-dimensional space to obtain the dynamic distribution of the three-dimensional temperature field in the cable trench. Based on the dynamic distribution of the three-dimensional temperature field within the cable trench, the sensor temperature drift error is calculated, and a compensated relative humidity value is generated by combining the molecular adsorption correction amount.

7. The cable operating environment optimization method based on cable trench information as described in claim 6, characterized in that: The method of calculating the cable insulation risk index using the compensated relative humidity value and standard pulse discharge quantity sequence via quantum annealing algorithm is as follows: A humidity gradient field is constructed based on the compensated relative humidity value, and the spatiotemporal distribution characteristics of the standard pulse discharge quantity sequence are extracted at the same time. The humidity gradient field and its spatiotemporal distribution characteristics are fused into a four-dimensional spatiotemporal tensor field, and the ground state energy of the humidity-partial discharge coupled Hamiltonian is solved by the quantum annealing algorithm. The ground state energy is mapped to a topological risk field and then smoothed using Gaussian kernels to normalize and generate a cable insulation risk index.

8. The cable operating environment optimization method based on cable trench information as described in claim 7, characterized in that: When the cable insulation risk index exceeds the risk threshold, a high-risk judgment flag is generated. The specific steps are as follows: The cable insulation risk index is input into a quantum rotating door-enhanced long short-term memory network, which outputs a risk threshold. By comparing the cable insulation risk index with the risk threshold, a high-risk judgment indicator is generated when the cable insulation risk index continuously exceeds the risk threshold.

9. The cable operating environment optimization method based on cable trench information as described in claim 8, characterized in that: The specific steps for inputting the cable insulation risk index into a quantum rotating gate-enhanced long short-term memory network and outputting a risk threshold are as follows. The cable insulation risk index, ambient temperature change rate, cable load current and historical fault sequence are integrated into a multi-dimensional feature vector. Multidimensional feature vectors are mapped to quantum state representations through a quantum state embedding layer. The quantum state representations are then processed by forward propagation using a quantum rotation gate-enhanced long short-term memory network to generate a basic risk threshold. The risk threshold is obtained by physically correcting the basic risk threshold based on thermodynamic entropy change constraints.

10. The cable operating environment optimization method based on cable trench information as described in claim 8, characterized in that: The method for optimizing the cable trench environment based on high-risk assessment criteria, using intelligent louvers driven by shape memory alloys and a ventilation fan, involves the following specific steps. Based on the status of the high-risk assessment indicator, a shape memory alloy driving current control signal is generated; The opening and closing angle of the shape memory alloy louver is adjusted by using the driving current control signal of the shape memory alloy. Based on the real-time humidity change rate and the opening and closing angle of the shape memory alloy louvers, the target speed of the ventilator is calculated, the optimal turbulence intensity is generated, and the cable operating environment is optimized.

Images