Edge Computing in Smart Hydropower Station: Noise Suppression for Condition Monitoring

Smart hydropower station (SHS) possesses a key role in smart energy system, thus the health management of electrical equipment in SHS becomes extremely important. Condition monitoring is widely used to assess the condition of electromechanical equipment in SHS. However, the explosive growth of monitoring data has brought great challenges to the centralized condition monitoring method, while the condition monitoring is always accompanied by noise from variety sources. Wavelet de-noising technique is one of the most prevalent methods for purifying monitoring signals from white noise, but the threshold selection and the threshold function is still a critical challenge. To this end, under the cloud-edge collaborative framework in SHS, this paper proposes a novel wavelet de-noising method, the method includes a short time 3o based threshold decision method and a slide energy windows based threshold processing rule. The simulation results verify the effectiveness and superiority of the proposed method over other available methods.