Simulation Research On Cable Local Aging Diagnosis Based On Frequency Domain Reflection

During operation,the cable is inevitably affected by factors such as heat,electric field,mechanical action,and radiation,which gradually ages the insulation material and reduces the reliability of the cable operation.Most cable faults are often developed from latent defects such as local insulation aging,while the traditional cable detection methods are mostly limited to the overall aging state diagnosis or the location after a fault occurs.The cable can be regarded as a transmission line at high frequencies,and the parameters of each section of cable are determined by the properties of the cable's conductors and dielectric materials,physical dimensions and shape,etc.When the signal propagates on the cable,it will refract and reflect where the parameters change.The reflection coefficient spectrum of cable contains the information of local aging state,which can realize the diagnosis of cable local aging.The diagnosis research of local cable aging is based on the principle of frequency domain reflection,the main studies of this paper is as follows:1.It analyzed the phenomenon and reason of cross linked polyethylene(XLPE)cable insulation aging,and studied the aging mechanism of thermal aging and electrical aging.According to the dielectric properties of the dielectric,it analyzed the trend of the dielectric constant of XLPE material with the aging state.2.According to the characteristics of cable structure and the principle of electromagnetic field,considering the skin effect of conductor and the dielectric response of semiconducting material at high frequency,the frequency domain calculation model of cable distribution parameters was established.Based on the transmission line theory,a local aging model of cable was proposed,and it calculated the reflection coefficient spectrum of cable.The reflection coefficient spectrum of a 10kV cable shows that the reflection coefficient spectrum of a cable with local aging defects decays exponentially with increasing frequency,and periodically oscillates around the reflection coefficient spectrum of an intact cable.3.The reflection coefficient spectrum of cable is windowed and transformed into spatial domain function by inverse fast Fourier transform(IFFT).The simulation calculation positioning error of 10kV cable can reach 0.01%.The accuracy and sensitivity of the diagnosis results are related to the test frequency and the number of test points.The diagnosis results at 0.1?60 MHz frequency and 600 points show that the maximum value corresponding to the aging point is linearly related to the real part of the complex dielectric constant of the insulating material,it is proportional to the aging length,and exponentially decays with the distance from the aging point to the measuring end.In addition,In addition,the positioning results between multiple points do not affect each other.4.Based on the correlation between the diagnosis results and the aging characteristic quantities,it constructed a local aging condition evaluation model of the cable,and the aging characteristic quantity is evaluated with genetic algorithm.The error is less than 1%compared with the actual aging characteristic quantities.The research shows that the method based on frequency domain reflection principle has good accuracy and sensitivity,and can quickly detect the latent aging defects of cables,which is of great significance to the stable operation of power system.