Research On XLPE Cable Insulation Defect Detection Technology Based On Tomography

Crosslinked polyethylene(XLPE)cables are widely used in power systems for their low cost,eminent insulation performance.If there are latent defects in cable insulation,it may cause partial discharges which leads to insulation breakdown in long-term operation,or even cause explosion or fire.This is a huge challenge to the safe operation and power supply reliability of the power system.Therefore,it is necessary to control the quality of cable access power system and improve its operation and maintenance quality.Due to the lack of effective methods to detect latent insulation defects,a new method for cable insulation detection based on Electrical capacitance-tomography(ECT)is proposed in this paper.Firstly,the basic structure of ECT system and the basic principle of the ECT technology are introduced.The finite element simulation model of the YJV1*185XLPE power cable and the eight-electrode ECT sensor is established,and the sensitivity field distribution is solved.The electric field,electric potential distribution and capacitance value are calculated for the image reconstruction of the insulation.The differences between the structure of single core cable and the current mainstream application scenarios of ECT technology are compared.The copper core structure in the cable will have a certain influence on the imaging process.Secondly,the capacitance value and uniformity of sensitivity field are selected as the optimization objectives,and simulation models are established to solve the corresponding sensitivity fields.Three geometrical parameters of the sensor-the electrode thickness,the electrode angle and the ratio between the radius of the shielding and the cable are optimized.The optimum parameter combination of the sensor is obtained by single factor optimization and orthogonal design.Five different cable insulation defects-single air gap,metal tip,semiconductor particle,double air gaps and local aging are simulated and detected by the optimized sensor.The potential,electric field distribution and capacitance are compared.For air gap defect,the capacitance decreases between adjacent electrodes,while for metal,semiconductor and local aging defects the opposite effect is found.With the Linear Back Projection(LBP)algorithm,images of the interior of the faulty cables are constructed.The results show that the air gap defects can be clearly distinguished from other defects and the number of defects can also be shown in the image.Finally,three kinds of ECT sensors are made,and the experimental platform for cable insulation detection is built.Two kinds of cable defects-air gap and metal tip are simulated,and the ECT sensors are used to conduct the cable insulation detection.The capacitance value detected by the four-electrode sensor is the largest,but the measurement data is too less to obtain high quality image.The eight-electrode sensor has balanced performance and good imaging effect.The change of the capacitance detected by the 12-electrode sensor is very small and there are many artifacts in the image.The experimental results show that ECT can successfully detect and distinguish two different types of defects-air gap and metal tip,and it is feasible to use it in cable insulation detection.