A Hybrid Fault Detection and Diagnosis of Grid-Tied PV Systems: Enhanced Random Forest Classifier Using Data Reduction and Interval-Valued Representation

Khaled Dhibi,Hazem Nounou,Majdi Mansouri,Mohamed Nounou,Mohamed Trabelsi,Radhia Fezai,Kais Bouzrara

doi:10.1109/access.2021.3074784

Abstract

This paper proposes a novel fault detection and diagnosis (FDD) technique for grid-tied PV systems. The proposed approach deals with system uncertainties (current/voltage variability, noise, measurement errors,…) by using an interval-valued data representation, and with large-scale systems by using a dataset size-reduction framework. The failures encompassed in this study are the open-circuit/short-circuit, islanding, output current sensor, and partial shading faults. In the proposed FDD approach, named interval reduced kernel PCA (IRKPCA)-based Random Forest (IRKPCA-RF), the feature extraction and selection phase is performed using the IRKPCA models while the fault classification is ensured using the RF algorithm. The main contribution of the proposed approach is to provide a good trade-off between low computation time and high classification metrics. The performance of the proposed IRKPCA-RF approach is assessed using a set of emulated data of a grid-tied PV system operating under healthy and faulty conditions. The presented results show that the proposed IRKPCA-RF approach is characterized by enhanced diagnosis metrics, classification rate, and computation time compared to the classical techniques.

Highlights

Photovoltaic (PV) has become the fastest growing renewable energy technology
AND DISCUSSION the performance of the proposed fault detection and diagnosis (FDD) methods is assessed using a set of emulated PV system data
The diagnosis assessment indicators include: 1) Normalized Classification Accuracy (NCA), which represents the ratio of the number of correct predictions to the total number of input samples, 2) Normalized Recall (NR), which is the percentage of fault measurements that are correctly classified over the total number of measurements in the pertinent fault class, 3) Normalized Precision (NP), which defines the number of samples properly classified divided by the number of classified samples, and 4) Computation Time (CT), which represents the time required to execute the FDD algorithm

Summary

Introduction

Photovoltaic (PV) has become the fastest growing renewable energy technology. the operation of PV systems is generally accompanied by different types of failures due to the harsh environmental conditions or internal malfunctions [1], [2]. The operation of PV systems should be accompanied by the implementation of an accurate fault detection and diagnosis (FDD) algorithm in order to reduce power losses and avoid system collapse [2]–[4]. Many machine learning (ML) techniques were developed to deal with FDD in PV systems [3], [5]–[7]. Among these techniques, artificial neural network (ANN) [8], support vector machine (SVM) [9], [10] and random forest (RF) [11] are the most common

Objectives

Results

Conclusion