Abstract:Distributed energy generation increases the need for smart grid monitoring, protection, and control. Localization, classification, and fault detection are essential for addressing any problems immediately and resuming the smart grid as soon as possible. Simultaneously, the capacity to swiftly identify smart grid issues utilizing sensor data and easily accessible frequency and voltage data from PMU devices is a prerequisite of this task. Therefore, this paper proposes new methods using fuzzy logic and adaptive fuzzy neural networks as well as machine learning and meta-heuristic algorithms. First, line voltage is used by a fuzzy thresholding method to estimate when a transmission line defect would develop in less than 1.2 clock cycles. Next, features taken from frequency signals in the real-time interval are utilized to classify the type of error using machine learning systems (decision tree algorithm and random forest algorithm) optimized with wild horse meta-heuristic algorithm. To locate the precise problem location, we finally use a neural fuzzy inference system that is capable of adapting to new data. We employ a simulated power transmission system in MATLAB to test our proposed solutions. Mean square error (MSE) and confusion matrix are used to assess the efficiency of a classifier or detector. For the decision tree algorithm method, the detector attained an acceptable MSE of 2.34e−4 and accuracy of 98.1%, and for the random forest algorithm method, an acceptable MSE of 3.54e−6 and accuracy of 100%. Furthermore, the placement error was less than 153.6 m in any direction along the line.

What problem does this paper attempt to address?

This paper attempts to address the problem of fault detection, classification, and location in power transmission lines. Specifically, with the increase in distributed energy resources, the need for monitoring, protection, and control in smart grids is also increasing. Quickly and accurately detecting, classifying, and locating faults on power transmission lines is crucial for immediately addressing issues and restoring the smart grid as soon as possible. However, existing methods have problems with slow speed and low accuracy in real-time applications. Therefore, this paper proposes a new method that utilizes fuzzy logic, adaptive fuzzy neural networks, machine learning, and metaheuristic algorithms to improve the efficiency and accuracy of fault detection, classification, and location. ### Main Issues 1. **Fault Detection**: How to detect transmission line faults in less than 1.2 clock cycles. 2. **Fault Classification**: How to classify fault types using real-time frequency signal features. 3. **Fault Location**: How to accurately locate the fault position. ### Solutions 1. **Fault Detection**: Use a fuzzy threshold method to estimate the occurrence of faults by measuring line voltage. 2. **Fault Classification**: Use optimized decision tree algorithms and random forest algorithms (optimized by the Wild Horse Optimization Algorithm) to classify fault types. 3. **Fault Location**: Use an Adaptive Neuro-Fuzzy Inference System (ANFIS) to accurately locate the fault position. ### Experimental Validation - A power transmission system was simulated in MATLAB to test the proposed method. - Mean Squared Error (MSE) and confusion matrix were used to evaluate the efficiency of the classifier or detector. - The detector using the decision tree algorithm achieved an MSE of 2.34e-4 and an accuracy of 98.1%, while the detector using the random forest algorithm achieved an MSE of 3.54e-6 and an accuracy of 100%. - The fault location error was less than 153.6 meters in any direction. ### Main Contributions - Proposed an intelligent strategy for various fault types and their location methods in transmission systems. - Addressed fault location methods and fault type issues affecting power systems, smart grids, and microgrids using machine learning systems and metaheuristic algorithms. - Provided advanced methods to quickly and accurately address issues in smart grids and microgrids by considering fault-tolerant control, automatic fault location, and self-healing communication structures, and using intelligent algorithms to identify and minimize risks. - Improved the flexibility of these intelligent systems by considering different parameters and minimized distance errors. ### Research Background - The introduction of smart grids has improved the reliability and adaptability of power systems. - Microgrids require fast and adaptive fault classification procedures due to their kinetic energy storage limitations and dynamic response of electronic power converters. - Fault detection and classification in transmission lines are crucial for root cause analysis and rapid restoration of power systems. - The application of deep learning in fault analysis has become very effective due to its ability to automatically extract representative features from large datasets. ### Limitations of Existing Methods - Current deep learning-based fault detection algorithms require large amounts of data, and defining and collecting this data for all possible fault scenarios is very difficult. - Existing fault location methods have problems with slow speed and low accuracy in practical applications. ### Conclusion The proposed method performs excellently in fault detection, classification, and location, with high accuracy and low error, providing an effective solution for fault management in smart grids.

Fault Detection, Classification and Localization Along the Power Grid Line Using Optimized Machine Learning Algorithms

Application of machine learning methods in fault detection and classification of power transmission lines: a survey

Performance Evaluation of Machine Learning for Fault Selection in Power Transmission Lines

Fault detection and classification using deep learning method and neuro‐fuzzy algorithm in a smart distribution grid

Fault Detection in Power Transmission Lines Using AI Model

Assessment of Envelope- and Machine Learning-Based Electrical Fault Type Detection Algorithms for Electrical Distribution Grids

A Novel Machine Learning-Based Approach for Fault Detection and Location in Low-Voltage DC Microgrids

Fault Detection and Classification in Ring Power System With DG Penetration Using Hybrid CNN-LSTM

Single-Ended Fault Location for Hybrid Transmission Line Using Embedded Artificial Intelligence

Integrating fault detection and classification in microgrids using supervised machine learning considering fault resistance uncertainty

Fault Prediction and Awareness for Power Distribution in Grid Connected RES Using Hybrid Machine Learning

Transmission lines Fault Detection and Classification Using Deep Learning Neural Network

Grid Distribution Fault Occurrence and Remedial Measures Prediction/Forecasting through Different Deep Learning Neural Networks by Using Real Time Data from Tabuk City Power Grid

Enhancing Power Grid Resilience through Real-Time Fault Detection and Remediation Using Advanced Hybrid Machine Learning Models

Fault Prognosis in Smart Distribution System with Distributed Generation Using Fast Fourier Transform Assisted Machine Learning

Fault classification and localization in microgrids: Leveraging discrete wavelet transform and multi-machine learning techniques considering single point measurements

Optimal PMU Placement for Fault Classification and Localization Using Enhanced Feature Selection in Machine Learning Algorithms

Electromagnetic field and artificial intelligence based fault detection and classification system for the transmission lines in smart grid

Enhancing Fault Detection and Classification in MMC-HVDC Systems: Integrating Harris Hawks Optimization Algorithm with Machine Learning Methods

Fast and Accurate Fault Detection and Classification in Transmission Lines using Extreme Learning Machine

Fault classification and detection for photovoltaic plants using machine learning algorithms