Multi-fault diagnosis method for insulation condition of power transformer based upon cloud model

Power transformer constitutes the core component of power system for energy transmission and distribution. The operation status of power transformer directly affects the security and stability of the power grid operation. Any faults of power transformer can result in significant economic losses. Monitoring and diagnosis provide an effective way to detect the potential fault. However, the traditional approach is often limited for the concurrent potential multi-fault, and is also prone to make serious diagnostic deviation due to its definite thresholds. Predicated on this background, this paper is inspired by the innovative engineering model and then proposes an improved model to diagnose the multi-fault of power transformer accurately. Through this way, it aims to present a more efficient approach than the previous ones. Specifically, the new multi-fault diagnosis method can realize the transition from the precise quantitative numerical value to the qualitative evaluation concept, which is described as digital characteristics. Consequently, the collected data can be analyzed by the reasoning rules in the perspective of qualitative cloud concept. Once the relationship between the fault type and the cloud concept is determined, the diagnosis results can be deduced from the characteristic gas with different contents which has already been converted to the qualitative concept. A case study proves the feasibility and validity of the proposed model, and offers a new way for the multi-fault diagnosis of power transformer.