Discovering Operational Patterns Using Image-Based Convolutional Clustering and Composite Evaluation: A Case Study in Foundry Melting Processes

2025-11-17

AI Summary

To address the challenge of extracting operational patterns from unlabeled, high-noise, and highly variable time-series data in industrial process monitoring, this paper proposes an unsupervised framework integrating image-based convolutional representation learning and composite evaluation. It innovatively transforms one-dimensional smelting time-series into grayscale images and employs a deep convolutional autoencoder to learn robust, noise-invariant features. A hybrid soft-hard clustering strategy coupled with a two-stage optimization mechanism is designed, alongside a weighted composite clustering validity index—S_eva—integrating Silhouette, Calinski-Harabasz (CH), and Davies-Bouldin Index (DBI). Evaluated on over 3,900 real-world furnace operation records, the method successfully identifies seven interpretable operational modes, clearly distinguishing energy consumption profiles, thermal dynamics, and production cycles. It outperforms conventional and state-of-the-art deep clustering methods in both clustering performance and robustness, while ensuring domain interpretability and engineering deployability.

Technology Category

• Machine Learning: Clustering
• Data Mining & Knowledge Management: Anomaly/Outlier Detection
• Planning, Routing, and Scheduling: Activity and Plan Recognition

Application Category

• Web Mining and Content Analysis: Normalization, clustering, classification, and summarization of Web text
• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs
• Systems and Infrastructure for Web, Mobile and WoT: Data management and stream processing for Web, mobile and wireless applications

Abstract

Industrial process monitoring increasingly relies on sensor-generated time-series data, yet the lack of labels, high variability, and operational noise make it difficult to extract meaningful patterns using conventional methods. Existing clustering techniques either rely on fixed distance metrics or deep models designed for static data, limiting their ability to handle dynamic, unstructured industrial sequences. Addressing this gap, this paper proposes a novel framework for unsupervised discovery of operational modes in univariate time-series data using image-based convolutional clustering with composite internal evaluation. The proposed framework improves upon existing approaches in three ways: (1) raw time-series sequences are transformed into grayscale matrix representations via overlapping sliding windows, allowing effective feature extraction using a deep convolutional autoencoder; (2) the framework integrates both soft and hard clustering outputs and refines the selection through a two-stage strategy; and (3) clustering performance is objectively evaluated by a newly developed composite score, S_eva, which combines normalized Silhouette, Calinski-Harabasz, and Davies-Bouldin indices. Applied to over 3900 furnace melting operations from a Nordic foundry, the method identifies seven explainable operational patterns, revealing significant differences in energy consumption, thermal dynamics, and production duration. Compared to classical and deep clustering baselines, the proposed approach achieves superior overall performance, greater robustness, and domain-aligned explainability. The framework addresses key challenges in unsupervised time-series analysis, such as sequence irregularity, overlapping modes, and metric inconsistency, and provides a generalizable solution for data-driven diagnostics and energy optimization in industrial systems.

Problem

• Unsupervised discovery of operational modes in univariate time-series data
• Handling dynamic unstructured industrial sequences with high variability
• Addressing sequence irregularity overlapping modes and metric inconsistency

Innovation

• Transforms time-series into grayscale images for feature extraction
• Integrates soft and hard clustering with two-stage refinement
• Evaluates clustering using a novel composite score S_eva