Multivariate Time-Series Dataset for Industrial Boiler Operation Analysis

This dataset contains raw multivariate time-series data collected from an industrial boiler system operating under real industrial conditions. The data include sensor measurements related to thermal, flow, and operational states and were used for unsupervised clustering using Dynamic Time Warping Self-Organizing Maps (DTW-SOM) and supervised classification using Long Short-Term Memory (LSTM) neural networks. No preprocessing, filtering, normalization, or labeling has been applied to the released files. All data are provided in their original recorded form to enable independent analysis, benchmarking, and method comparison. The dataset supports research in process monitoring, fault detection, time-series clustering, and industrial AI applications.