Hyperparameter Tuning of OC-SVM for Industrial Gas Turbine Anomaly Detection

Abstract

Gas turbine failure diagnosis is performed in this work based on seven types of tag data consisting of a total of 7976 data. The data consist of about 7000 normal data and less than 500 abnormal data. While normal data are easy to extract, failure data are difficult to extract. So, this study mainly is composed of normal data and a one-class support vector machine (OC-SVM) is used here, which has an advantage in classification accuracy performance. To advance the classification performance, four hyperparameter tuning (manual search, grid search, random search, Bayesian optimization) methods are applied. To analyze the performance of each technique, four evaluation indicators (accuracy, precision, recall, and F-1 score) are used. As a result, about 54.3% of the initial failure diagnosis performance is improved up to 64.88% through the advanced process in terms of accuracy.

Anomaly Detection for Hydraulic Power Units—A Case Study

“…As a method that is easy to implement in Python using the scikit-learn package, we decided to implement OCSVM. It was presented in the article [71] that, using hyperparameters tuning (manual search, grid search, random search, and Bayesian optimization), the quality of the algorithm could be improved. The worst F1 score equal to 0.55521 was obtained by manual search, and the best F1 equal to 0.6488 was obtained for Bayesian optimization hyperparameter tuning.…”

Comprehensive analysis of supervised algorithms for coronary artery heart disease detection

“…The Randomized Search is a non-exhaustive technique hence reducing the chance of overfitting of the model but may not guarantee an optimal combination of hyperparameters. Furthermore, it is observed that the randomized tunned models are more accurate than conventionally tunned models for the models having a less number of critical hyperparameters (Kang et al, 2022;Valarmathi & Sheela, 2021)- (Bischl et al, 2023). Therefore, authors used hyperparameter tunning technique and pseudo code are representing this optimization process are as follows:…”

“…The Randomized Search is a non‐exhaustive technique hence reducing the chance of overfitting of the model but may not guarantee an optimal combination of hyperparameters. Furthermore, it is observed that the randomized tunned models are more accurate than conventionally tunned models for the models having a less number of critical hyperparameters (Kang et al, 2022; Valarmathi & Sheela, 2021)–(Bischl et al, 2023). Therefore, authors used hyperparameter tunning technique and pseudo code are representing this optimization process are as follows: Initialize an empty dictionary variable “space”. Define a search function to save all hyperparameter subspaces. For each hyperparameter subspace stored in space, create a machine‐learning model with a set of hyperparameters obtained by random sampling. Fit each model by using training data samples. A dictionary score to store each model's accuracy or other performance metrics. Return the best model score and its parameters. …”

Anomaly Detection in Gas Turbines Using Outlet Energy Analysis with Cluster-Based Matrix Profile

“…Data-mining techniques have been widely employed across a range of domains to extract relevant knowledge from large datasets, in order to support informed decision-making processes [6]. In particular, increasing attention has been directed towards the application of data-mining techniques for knowledge extraction from time series data-i.e., an ordered set of observations recorded over time pertaining to a particular phenomenon and measured across a defined time span [7,8].…”