Partial discharge of electrical equipment is not only the main factor of insulation deterioration, but also an important parameter to effectively characterize insulation defects. The accurate detection of partial discharge can detect the potential faults that endanger the safety of equipment in time. In order to solve the problems that rigid structure, small size structure and wide bandwidth performance of the existing microstrip antenna sensors are difficult to balance, the UHF detection method based on antenna is studied in this paper, and a new flexible microstrip antenna sensor is developed considering antenna size and bandwidth. Using polyimide as base material, the narrow-band characteristics are improved by using partial flooring technology, and the beveled meandering technology is introduced to expand the working bandwidth while keeping the antenna area unchanged. Aiming at the problem of unstable antenna performance caused by single size parameter adjustment in the process of size optimization, the nonlinear relationship between antenna size parameters and working bandwidth is studied, and the mathematical model between them is established using radial basis function (RBF) neural network. The improved beluga whale optimization (IBWO) algorithm is used to optimize the size of antenna with the maximum working bandwidth as the target. The simulation results show that the size of the new flexible microstrip antenna is reduced by 59.59%. The operating bandwidth is increased from 0.598GHz~0.6GHz to 0.3GHz~3GHz, which fully covers the design requirements of partial discharge detection. By simulating partial discharge test and comparing with Archimedean spiral antenna and three-dimensional spiral antenna, the results show that the new flexible microstrip antenna has more reliable detection performance.