Abstract: In order to reduce the SNR of signal detection, a novel chaotic attractor was used as a detection tool. Lyapunov exponent method was used to analyze the dynamic characteristics. The number of chaotic oscillators used in covering the initial phase difference was reduced by increasing the accuracy of the critical chaotic state threshold. A novel phase state identification method was proposed and designed in circuit simulation. It can realize the automatically identify phase states by using the period characteristics of the system. The period of intermittent chaotic state was recorded. Finally, the detection and simulation of the partial discharge signal with exponential attenuation oscillation were carried out. The simulation results show that the novel chaotic attractor has rich dynamic characteristics and superior anti-noise performance. It can achieve lower SNR detection. Unaffected by the transition process, the novel phase state identification method is applicable to any chaotic oscillator and has the feasibility of circuit implementation. A combination of the novel attractor and new method can realize detection of ultra-high frequency millivolt-level partial discharge signals of transformer.