A long-tailed distribution time-series dataset in boiler equipment

Industrial boilers are critical in various industries, transforming fuel's chemical energy into heat energy in the form of steam. These equipment primarily serve sectors such as power generation, chemicals, metallurgy, and paper production, providing essential heat and steam. Time-series analysis of boiler parameters offers valuable insights into the interactions of various operational factors, enabling optimization of settings and extending the equipment's service life. However, due to the challenging operational environments and data transmission issues, sensors often record incomplete or erroneous data. This study focuses on collecting operational data from a coal-fired boiler at a chemical plant in Zhejiang. By preprocessing this flawed data, we generate a high-dimensional time-series dataset that includes key operational parameters like pressure, temperature, flow rate, and oxygen levels. Using the boiler outlet steam temperature as a key indicator of equipment condition, the dataset also reveals long-tailed distribution, offering a foundation for addressing long-tailed issues in industrial equipment analytics.