Abstract: Power cables have excellent insulation performance and are widely used in rail transit systems and power systems, but they may suffer from insulation faults under the long-term action of external insulation degradation factors, which affect the stability of the system power supply. In order to accurately determine the status of cable insulation and ensure the safe and stable operation of the system, a simulation study is conducted on the detection of insulation defects in power cables based on electrical capacitance tomography. A model of power cable with eight-electrode capacitive sensor is constructed. The geometry of the capacitive sensor is optimized in terms of signal output strength, and the sensitivity field distribution is optimized using the median filtering algorithm. By integrating the optimized capacitive sensor and sensitivity field, employing Landweber image reconstruction algorithm, the reconstruction of images representing four typical cable insulation defects: air gap, water tree, wedge-shaped scratch, and composite defects, is facilitated, thus enabling the detection of power cable insulation defects.