Resumen:
In this paper, we describe the procedure of implementing a reinforcement learning algorithm, TD3, to learn the performance of a cooling water pump and how this type of learning can be used to detect degradations and evaluate its health condition. These types of machine learning algorithms have not been used extensively in the scientific literature to monitor the degradation of industrial components, so this study attempts to fill this gap, presenting the main characteristics of these algorithms’ application in a real case. The method presented consists of several models for predicting the expected evolution of significant behavior variables when no anomalies exist, showing the performance of different aspects of the pump. Examples of these variables are bearing temperatures or vibrations in different pump locations. All of the data used in this paper come from the SCADA system of the power plant where the cooling water pump is located.
Resumen divulgativo:
Este artículo describe un procedimiento para implementar un algoritmo de aprendizaje por refuerzo, TD3, para conocer el comportamiento de una bomba de agua de refrigeración y cómo este tipo de aprendizaje puede utilizarse para detectar degradaciones y evaluar su estado. Se incluyen ejemplos con casos reales.
Palabras Clave: TD3; reinforcement learning; cooling water pump; performance monitoring; health condition; failure mode risk
Índice de impacto JCR-JIF y cuartil WoS: 5,200 - Q2 (2025)
Referencia DOI:
https://doi.org/10.3390/computers14120540
Publicado en papel: Diciembre 2025.
Publicado on-line: Diciembre 2025.
Cita:
M.A. Sanz-Bobi, I. Rodríguez-Muñoz-de-Baena, F.J. Bellido-López, A. Muñoz, J. Anguera, D. González-Calvo, T. Álvarez Tejedor, "TD3 Reinforcement Learning Algorithm Used for Health Condition Monitoring of a Cooling Water Pump", Computers, Vol. 14, nº. 12, pp. 540, Diciembre 2025. [Online: Diciembre 2025] doi: 10.3390/computers14120540
Descargar el artículo