Process Phase Monitoring in Industrial Manufacturing Processes with a Hybrid Unsupervised Learning Strategy

Monitoring production processes, such as those in the chemical industry, involves several key aspects, including identifying unknown process phases, tracking their sequence and duration, and detecting anomalies that may occur within phases. As demonstrated in several industrial applications, the ability of Self-Organizing Maps (SOMs) to detect anomalies, identify and visualize process phases is highly beneficial for comprehensive monitoring and understanding of technical processes. This paper presents a hybrid unsupervised learning strategy (HULS) for monitoring complex industrial processes. Addressing the limitations of SOMs, especially in scenarios with unbalanced datasets and highly correlated process variables, HULS combines existing unsupervised learning techniques to address these challenges. Based on a laboratory batch process, the capabilities of the HULS concept for the detection of unknown process phases are evaluated in comparison to the standard SOM model.