CN-121347976-B - Sea cable insulation performance on-line monitoring and fault positioning method, medium and equipment

CN121347976BCN 121347976 BCN121347976 BCN 121347976BCN-121347976-B

Abstract

The invention discloses a submarine cable insulation performance on-line monitoring and fault positioning method, medium and equipment, wherein the method comprises the steps of S1, real-time synchronous acquisition of multisource sensing data, S2, collaborative suppression of multi-mode noise and feature extraction, S3, insulation state degradation evaluation of multi-feature fusion, S4, double-end traveling wave accurate fault positioning, and S5, data visualization and alarm linkage. The invention can effectively inhibit noise interference of the marine complex environment, monitor key parameters representing the insulation state of the submarine cable in real time in a multi-dimensional manner, realize early warning of insulation degradation and preliminary judgment of fault modes through an intelligent fusion evaluation model, realize accurate positioning of fault points by utilizing an improved traveling wave accurate capturing and wave velocity dynamic correction technology, remarkably improve the accuracy, sensitivity and fault positioning precision of submarine cable state monitoring and provide strong technical support for safe and economic operation of submarine cables.

Inventors

SU BAOZHONG
CHEN YUGUANG
ZHANG ZHANBIN
ZHOU YU
LIANG QIANG
ZHU ZHENMIN
HU ZHE
WANG SHAOWEI
GUO JUN
LI DONGYI

Assignees

中海油能源发展装备技术有限公司

Dates

Publication Date: 20260512
Application Date: 20251107

Claims (9)

1. The on-line monitoring and fault positioning method for the insulation performance of the submarine cable is characterized by comprising the following steps of: s1, synchronously acquiring multi-source sensing data in real time, namely synchronously deploying a plurality of insulation state detection sensors in base stations or connecting stations at two ends of a submarine cable, and measuring and acquiring multi-source heterogeneous original sensing data in real time; S2, multi-modal noise cooperative suppression and feature extraction, namely preprocessing the acquired multi-source heterogeneous original sensing data stream to obtain data with pure signals; S3, constructing a dynamic insulation state evaluation model based on a deep belief network or a long-short-term memory network, and pre-training the model to learn the multidimensional characteristic parameter spatial distribution and association relation under the normal running state of the submarine cable; s4, double-end traveling wave accurate fault positioning, namely triggering a fault positioning flow when the output of the dynamic insulation state evaluation model triggers high-risk alarm or the mutation fault characteristic is detected; s5, data visualization and alarm linkage are carried out, namely monitoring data and fault locating points are displayed on an interface in real time, and alarm information is pushed in a multistage linkage mode while alarm signals are triggered; in the step S4, the fault locating process further includes: S4.1, capturing and marking traveling wave signals with high precision, namely capturing initial current traveling wave signals generated at the moment of faults by the HFCT sensors at two end stations at an ultrahigh sampling rate, and accurately positioning the time stamp of the initial traveling wave head reaching a local end sensor by utilizing an improved Teager energy operator in combination with wavelet transformation mode maximum point detection; The improved Teager energy operator is used for carrying out time-frequency double-domain self-adaptive parameter adjustment on the original Teager energy operator and is used for improving the sensitivity and the anti-saturation capacity of the traveling wave head with low signal-to-noise ratio; s4.2, dynamically correcting the wave speed in real time, namely dynamically calculating the traveling wave propagation speed under the current condition by utilizing a submarine cable wave speed-temperature-pressure relation model based on the current running condition of the submarine cable; The submarine cable wave speed-temperature-pressure relation model is an experience/semi-experience model which is based on submarine cable structural parameters, actually measured conductor temperature data and buried depth/water pressure data and is fitted by using a physical equation or actually measured data; S4.3, double-end ranging calculation, namely calculating the accurate distance L_fault of the fault point from the end A according to the time difference delta t between the two points, the traveling wave propagation speed v_curr and the known total length L from the end A to the end B of the submarine cable, wherein delta t= |t_A-t_B|, and outputting the accurate positioning result of the fault point by using the following formula: L_fault=[L+(t_A-t_B)*v_curr]/2; Or l_fault= [ L- (t_b-t_a) ×v_curr ]/2; wherein t_A and t_B represent traveling wave head arrival times, and the A-terminal measurement time is earlier than the B-terminal time.
2. The on-line monitoring and fault locating method for submarine cable insulation performance according to claim 1, wherein in the step S1, the plurality of insulation state detection sensors at least comprise high-frequency current transformers HFCT, current sensors, voltage sensors, temperature sensors and acceleration sensors, sampling frequencies are not lower than 50kHz, and the collected multi-source heterogeneous original sensing data at least comprise submarine cable core wire conductor current, submarine cable metal sheath current, submarine cable conductor-to-sheath voltage, submarine cable sheath-to-ground voltage, submarine cable surface temperature, partial discharge signals and submarine cable body vibration acceleration.
3. The submarine cable insulation performance online monitoring and fault locating method according to claim 1, wherein in S2, the preprocessing operation further comprises: S2.1, carrying out broadband collection on current signals and voltage signals, and separating and inhibiting marine environment background noise and power system switching transient noise by utilizing an adaptive filtering algorithm based on empirical mode decomposition or wavelet packet transformation and combining an abnormal mode library trained by marine cable running state historical data; In the process of self-adaptive filtering of broadband signals, an abnormal mode library is established by recording original signal characteristics under different sea conditions and different load conditions for a long time in a good state after sea cable initial operation or overhaul, identifying normal signal clusters by using an unsupervised learning algorithm, marking known type interference mode samples, and carrying out real-time filtering by using the self-adaptive filtering algorithm by utilizing the abnormal mode library to online match interference components; S2.2, enhancing the characteristics of vibration signals, namely filtering low-frequency drift and irrelevant high-frequency noise of vibration signals acquired by an acceleration sensor by utilizing a band-pass filter, and extracting characteristic frequency band energy reflecting the mechanical stress of the submarine cable and the vibration state of an insulating medium; S2.3, multi-dimensional characteristic parameter fusion calculation, namely on-line calculating a multi-dimensional characteristic parameter set FV of key insulation performance based on the pure signals after filtering and noise reduction; The multi-dimensional characteristic parameters of the key insulation performance at least comprise vibration characteristic frequency band energy ratio, sea cable insulation layer dielectric loss tangent value or equivalent value under characteristic frequency; The vibration characteristic frequency band energy ratio is the ratio or the change rate of the frequency band energy which specifically reflects the internal stress concentration or defect characteristic of the insulating medium and the reference stable band energy; The dielectric loss tangent value of the submarine cable insulating layer or the equivalent value under the characteristic frequency thereof is calculated by carrying out real-time vector relation analysis on the conductor-sheath voltage and the capacitive current component flowing on the sheath grounding line, or measuring at the system voltage frequency and the main harmonic frequency point thereof.
4. The on-line monitoring and fault locating method for submarine cable insulation performance according to claim 3, wherein in the step S3, the input of the dynamic insulation state evaluation model is a multidimensional characteristic parameter set FV in a current moment and a historical sliding window obtained by calculation in the step S2.3; Under the online running state of the plurality of insulation state detection sensors, the output of the dynamic insulation state evaluation model is an insulation state health index HI and a potential fault mode confidence vector PF; Wherein: The insulation state health index HI ranges from 0 to 1, wherein 1 represents the optimal state, insulation early warning is triggered when the value of the insulation state health index HI is lower than a threshold value H_th1, high risk warning is triggered when the insulation state health index HI is lower than a threshold value H_th2, and the value of the threshold value H_th1 is larger than the value of the threshold value H_th2; The latent failure mode confidence vector PF points to a specific possible degradation cause.
5. The on-line monitoring and fault locating method for submarine cable insulation performance according to claim 1, wherein in S3, the training method of the dynamic insulation state evaluation model is based on submarine cable full life cycle simulation data, historical operation and maintenance data and normal and known fault case field data obtained through the above steps, and performs end-to-end training.
6. The submarine cable insulation performance online monitoring and fault locating method according to claim 1, wherein in the step S3, the dynamic insulation state assessment model comprises a characteristic self-attention mechanism for highlighting key characteristic weights.
7. The on-line monitoring and fault locating method for submarine cable insulation performance according to claim 1, wherein in the step S5, insulation health index HI trend, multidimensional characteristic parameter FV change trend, potential fault mode probability vector PF and fault locating result are displayed in real time, when insulation early warning, high risk warning or fault locating is triggered, multistage warning information is synchronously sent through an SCADA system, an audible and visual alarm and a mobile terminal APP, and detailed working condition data records are recorded for later retrospective analysis.
8. A computer readable storage medium, characterized in that a computer program is stored, which, when being executed by a processor, causes the processor to perform the steps of the submarine cable insulation performance on-line monitoring and fault localization method according to any one of claims 1-7.
9. A computer device comprising a memory and a processor, the memory storing a computer program which, when executed by the processor, causes the processor to perform the steps of the submarine cable insulation performance on-line monitoring and fault locating method according to any one of claims 1 to 7.

Description

Sea cable insulation performance on-line monitoring and fault positioning method, medium and equipment Technical Field The invention relates to the technical field of power cables, in particular to a submarine cable insulation performance on-line monitoring and fault positioning method, medium and equipment. Background Submarine power cables, for short submarine cables, are used as core "arteries" connecting the offshore energy base with the land grid, and the reliability and safety of operation of the submarine power cables are of paramount importance. Because the insulating layer is laid in a complex and severe marine environment (high salt fog, high water pressure, strong corrosion, biological adhesion, ocean current impact and fishery/shipping activity interference), the insulating layer (mainly adopts cross-linked polyethylene XLPE or impregnated paper insulation) bears the coupling effect of multiple stresses such as electricity, heat, machinery, chemistry and the like for a long time, and deterioration such as water tree, electric branch, sheath damage, insulating layer cracking and the like easily occurs, and finally insulation breakdown fault is caused. Once a submarine cable fails, the maintenance difficulty and cost (related to expensive professional ships, long window period and positioning difficulty) are extremely high, and significant economic loss and energy supply interruption can be caused, so that the performance and working state of the submarine cable need to be monitored in real time. The existing submarine cable state monitoring technology mainly has the following defects: 1) The monitoring content is single, and the information fragmentation is that the prior art focuses on the monitoring of one or a few signals, such as the monitoring of grounding current, sheath circulation, temperature or partial discharge, and the like. It is often difficult for a single signal to fully and accurately reflect the overall state of the insulation and its degradation mechanism (e.g., inability to distinguish between water tree and mechanical damage). Multisource heterogeneous data lacks efficient fusion analysis. 2) The noise immunity is weak in a strong interference environment, and the marine environment has abundant background noise (mechanical vibration noise caused by surge impact, electromagnetic noise of a ship engine and a radar, transient noise of power system switch operation and the like), the noise has large amplitude and wide frequency band, and is extremely easy to submerge weak insulation degradation characteristic signals (especially early partial discharge signals and slight leakage current changes), so that effective signal extraction is difficult, and the false alarm rate and the missing alarm rate are high. 3) The insulation state evaluation model is simple, low in precision and early warning lagging, and most of the current common evaluation methods are based on simple threshold comparison (such as insulation resistance is lower than a certain value and partial discharge exceeds a certain value) or a simple regression model. The models are difficult to capture complex nonlinear relations among multiple characteristic parameters and the deteriorated dynamic evolution process, have low sensitivity to early and slow-changing insulation defects, have high misjudgment rate and have short effective early warning time. 4) The fault positioning precision is low, the reliability is poor, and the existing positioning mainly depends on the traditional double-end traveling wave method or impedance method. The traveling wave method is easily affected by complicated marine environment noise (especially secondary reflected wave interference) and line parameter fluctuation (traveling wave speed change caused by temperature and water depth change), so that the wave head is inaccurate in calibration, the wave speed setting error is large, the positioning accuracy is usually in the range of hundreds of meters or even a few kilometers, and the accurate repair requirement of the submarine cable is difficult to meet (the positioning error is required to be better than 1% or less than 100 meters). The impedance method is sensitive to the transition resistance, and the error is larger in submarine cable faults. 5) The intelligent diagnosis and linkage are lacking, the existing monitoring system often lacks intelligent diagnosis capability, the monitoring data is difficult to organically link with the health state of equipment, the potential risk and the accurate position of the final fault point, and the operation and maintenance decision support is insufficient. Disclosure of Invention The sea cable insulation performance on-line monitoring and fault locating method, medium and equipment integrating multisource synchronous high-precision sensing, strong noise suppression, multi-feature fusion intelligent evaluation and high-precision fault locating into a whole can at least solve one of the technical