Abstract:Anomaly detection is a critical task that involves the identification of data points that deviate from a predefined pattern, useful for fraud detection and related activities. Various techniques are employed for anomaly detection, but recent research indicates that deep learning methods, with their ability to discern intricate data patterns, are well-suited for this task. This study explores the use of Generative Adversarial Networks (GANs) for anomaly detection in power generation plants. The dataset used in this investigation comprises fuel consumption records obtained from power generation plants operated by a telecommunications company. The data was initially collected in response to observed irregularities in the fuel consumption patterns of the generating sets situated at the company's base stations. The dataset was divided into anomalous and normal data points based on specific variables, with 64.88% classified as normal and 35.12% as anomalous. An analysis of feature importance, employing the random forest classifier, revealed that Running Time Per Day exhibited the highest relative importance. A GANs model was trained and fine-tuned both with and without data augmentation, with the goal of increasing the dataset size to enhance performance. The generator model consisted of five dense layers using the tanh activation function, while the discriminator comprised six dense layers, each integrated with a dropout layer to prevent overfitting. Following data augmentation, the model achieved an accuracy rate of 98.99%, compared to 66.45% before augmentation. This demonstrates that the model nearly perfectly classified data points into normal and anomalous categories, with the augmented data significantly enhancing the GANs' performance in anomaly detection. Consequently, this study recommends the use of GANs, particularly when using large datasets, for effective anomaly detection.

Data Enhancement for Data-Driven Modeling in Power Plants Based on a Conditional Variational-Adversarial Generative Network

Variational Autoencoder Generative Adversarial Network for Synthetic Data Generation in Smart Home

Ensemble Data Augmentation for Imbalanced Fault Diagnosis.

Synthetic Active Distribution System Generation Via Unbalanced Graph Generative Adversarial Network.

Addressing Imbalance of Sample Datasets in Dissolved Gas Analysis by Data Augmentation: Generative Adversarial Networks

Anomaly Detection in Power Generation Plants with Generative Adversarial Networks

Power Plant Model Parameter Calibration Using Conditional Variational Autoencoder

A Method for Measuring Carbon Emissions from Power Plants Using a Cnn-Lstm-Attention Model with Bayesian Optimization

A novel deep generative modeling-based data augmentation strategy for improving short-term building energy predictions

Deep Generative Modeling-based Data Augmentation with Demonstration using the BFBT Benchmark Void Fraction Datasets

Deep learning GAN-based data generation and fault diagnosis in the data center HVAC system

Data Augmentation for Data-driven Methods in Power System Operation: A Novel Framework Using Improved GAN and Transfer Learning

A hybrid NOx emission prediction model based on CEEMDAN and AM-LSTM

Partial Discharge Data Augmentation Based on Improved Wasserstein Generative Adversarial Network With Gradient Penalty

Predicting Critical Heat Flux with Uncertainty Quantification and Domain Generalization Using Conditional Variational Autoencoders and Deep Neural Networks

A Novel Cross-Domain Data Augmentation and Bearing Fault Diagnosis Method Based on an Enhanced Generative Model

Demand Side Data Generating Based on Conditional Generative Adversarial Networks

Prediction of NOx emission concentration from coal-fired power plant based on joint knowledge and data driven

A Generative Adversarial Neural Network Model for Industrial Boiler Data Repair.

Hybrid Network Model Based on Data Enhancement for Short-term Power Prediction of New PV Plants

System-level Anomaly Detection for Nuclear Power Plants Using Variational Graph Auto-encoders