Fault Detection Method for Dust Removal Equipment Based on Digital Twin and OCSVM Optimized by Clonal Selection

Fault detection for dust removal equipment has always been a common anomaly detection scenario. Currently, dust removal equipment fault detection methods face issues such as data imbalance, low accuracy, and insufficient visualization. To address these issues, a fault detection method based on digital twin and OCSVM optimized by clonal selection is proposed. First, the cubic spline interpolation method is used to fill in missing data. Then, to tackle the issue of data imbalance, the One-Class Support Vector Machine (OCSVM) is introduced for anomaly detection. Meanwhile, leveraging the Clonal Selection Algorithm(CSA)'s ability to search for the optimal solution, this algorithm is applied to optimize OCSVM hyperparameters, improving detection accuracy. Finally, to address visualization issues, digital twin technology is used to visualize the dust removal system, the data, and the detection results. Experimental results show that this method effectively monitors anomalies in dust removal equipment, offering higher monitoring accuracy compared to traditional methods and allowing users to intuitively understand the environmental conditions, facilitating timely ad-justments.