Temperature-Based Accuracy Estimation on Instrument Transformers via Neural Networks and Microcontrollers

It goes without saying that the presence of instrument transformers (ITs) is crucial for all monitoring activities of the power network. However, in-field operating conditions are far from being like the laboratory's ideal ones. Environmental conditions like temperature and humidity affect the accuracy of ITs. Furthermore, the in-field assessment of the IT's ratio error and phase displacement is a very demanding task yet. To this purpose, this work describes the integration of artificial neural networks (ANNs) and microcontrollers to develop a simple and cheap system to estimate the accuracy of the ITs. The real-time ratio error and phase displacement estimation is based on one of the most impacting influence quantities, the temperature. The paper describes the ANN algorithm, the proposed method, and an example of a distributed measurement system capable of implementing it. The algorithm is then validated with actual measurements obtained from off-the-shelf ITs. The results clearly highlight the benefits of the proposed solution.

I. Introduction

The emergence of the Smart Grid concept can be attributed to the integration of Instrument Transformers (ITs) and Intelligent Electronic Devices (IEDs). IEDs play a crucial role in gathering measurements from ITs and subsequently executing onsite operations or transmitting the data to a central data concentrator. This process constitutes a key aspect of network monitoring and control, essential for System Operators (SOs).