Conditional-TimeGAN for Realistic and High-Quality Appliance Trajectories Generation and Data Augmentation in Nonintrusive Load Monitoring

• Published in: IEEE transactions on instrumentation and measurement 2024, Vol.73, p.1-15
• Main Authors: Liu, Z. G., Ji, T. Y., Chen, J. W., Zhang, L. J., Zhang, L. L., Wu, Q. H.
• Format: Article
• Language: English
• Subjects: Adaptation models; Conditional-time-series generative adversarial network (C-TimeGAN); Data analysis; Data augmentation; data generation; Data models; Energy consumption; energy disaggregation; Generative adversarial networks; Hidden Markov models; Load modeling; Machine learning; Model accuracy; Monitoring; nonintrusive load monitoring (NILM); one-class support vector machine (OCSVM); Outliers (statistics); Support vector machines; Synthetic data; Training
• ISSN: 0018-9456; 1557-9662
• DOI: 10.1109/TIM.2024.3381263

Nonintrusive load monitoring (NILM) strives to achieve real-time monitoring of individual appliance energy consumption and usage by leveraging aggregate power readings. Most of the existing NILM models are based on machine learning. To improve the generalization and accuracy of NILM models, it is crucial to expand the dataset. However, collecting a large amount of power data is a challenging and common task. To address this, we propose a novel model called conditional-time-series generative adversarial network (C-TimeGAN), which extends upon TimeGAN by incorporating constraint conditions to generate high-quality and realistic appliance trajectories. This model preserves the benefits of unsupervised GANs' flexibility and supervised training's stepwise control in TimeGAN while addressing challenges in appliance data generation, such as rapid power changes and transients. In addition, we utilize a one-class support vector machine (OCSVM) for postprocessing to further enhance data quality by detecting anomalies and removing outliers. Experiments show that the synthetic data from our C-TimeGAN+ model (C-TimeGAN integrated with OCSVM) outperform the baseline in terms of both quality and quantity, exhibiting higher fidelity, diversity, and novelty. Furthermore, C-TimeGAN+ serves as a practical data augmentation tool, enhancing disaggregation accuracy and generalization of other NILM models. We also discuss the optimal proportion of synthetic data for augmentation, improving the practical applicability of the data.