Hybrid Support System for Decision Making Based on MLP-ANN, IED and SCADA for Disturbances Analysis of Electrical Power Distribution Transformers

The operation and maintenance of the power system require attention, precise diagnostics on failure and agility on system recovery. In addition, each operation needs to be carefully planned and executed, once errors can be fatal. To improve the operation and maintenance tasks, in this article is presented the proposal of a support system for decision making units based on artificial neural network (ANN), intelligent electronic devices (IED), supervisory control and data acquisition (SCADA) system for disturbances analysis of electrical power distribution transformers. The intelligent system is hybrid in the sense that it performs on-line tasks in real time for data acquisition systems via IED and off-line tasks are performed for analysis of disturbances in electrical power distribution transformers. The hybrid decision making support system (HDMSS) has built a MLP-ANN engine for classifying patterns and providing support for decisions. The MLP-ANN engine is evaluated for fault detection in distribution transformer of electrical power substation. The proposed method was evaluated using real data collected directly from IED, such as: digital relays. The on-line simulations results show the effectiveness and the feasibility of the proposed system based on artificial neural network.