Descriptor: Refinery Compressor Sensor Data, One-Year Dataset (RCSD-1YD)

This paper presents a dataset obtained from a centrifugal compressor in an oil refinery located in Greece. This set of measurements, collected over a period of one year (01/01/2022-31/12/2022), was derived from the refinery’s distributed control system (DCS) and was used for monitoring the compressor’s operation, providing a detailed and comprehensive understanding of its performance. Key variables such as turbine axial displacement and oil temperature are monitored, offering invaluable insights into the compressor’s behaviour. These variables are monitored through 25 sensors and the dataset contains approximately 35036 readings for each one of them. The validation of the data is important for the optimal operation of the compressor; thus, the refinery’s industrial control systems validate data to ensure accuracy, consistency, and reliability for control and analysis. Validation methods include flagging anomalies, checking process variable ranges, and identifying noise or signal loss. The sensors’ readings of the examined dataset were compared to recommended operating ranges for the identification of patterns across the variables. The comparison revealed that in general the compressor exhibited stable behavior thus the data are indicative of normal operation with minimal deviations from expected performance. Beyond its on-site practical applications, this dataset serves as a valuable resource for academic research in the field of refinery compressors. Researchers can leverage machine learning techniques to explore new ways for improving compressor efficiency, reliability, and control.