I- INTRODUCTION

II- THE PROCESS OF MACHINE LEARNING

III- DATA PREPARATION- PSCAD BATCH SIMULATION

IV- DATA PREPROCESSING-THE FAULT FEATURE EXTRACTION

V- MACHINE LEARNING MODEL

REFERENCES

Abstract

To realize the intelligent of the distribution network, it is necessary to identify the fault type accurately. This paper presents the fault type identification method based on machine learning in active distribution networks. The process of machine learning is divided into four steps: data preparation, data preprocessing, feature extraction and model training. When preparing data, a method of generating fault scenarios in the batch of simulation experiments is presented. The IEEE34 Bus System is built in PSCAD to complete the data preparation for machine learning. Variation multiples of voltage and current are extracted as the features to describe the fault type. Various machine learning models are trained by cross-validation method to get the accuracy of identification. The application of decision tree in fault type identification is presented in the form of a tree diagram. The result of fault type identification is shown by the confusion matrix of the decision tree. All the test results show that the proposed fault identifiers can identify all kinds of fault types in the distribution network.

INTRODUCTION

In modern power grids, fast and accurate fault type identification is an essential operational requirement. Both relay protection and fault location require correct fault type information. There have been a large number of researches on fault classification in the power grid, mainly for the transmission network. Traditional fault type identification is achieved by setting threshold values and relying on logical relationships. The fault location and fault type are inferred based on the logic of the protection and the experience of the operator. This process is difficult to describe by traditional mathematical methods. Artificial intelligence technology has the characteristics of simulating human and has been widely used in this field. The main implementation methods include neural network approach[1], fuzzy neural network[2], expert systems[3], genetic algorithms[4], and Petri net[5]. More than 80% of the faults come from the distribution network in the power system. Fast and accurate fault classification in the distribution network is significant for fault analysis and power restoration, which can effectively improve power supply reliability. In [6], a method for fault type identification using decision trees is proposed. However, the identification of fault types does not involve a specific phase.