Anomaly Identification Method for the Detection Process of Determining Formaldehyde Emission From Wood-Based Panels by the Chamber Method Using GCN-BiLSTM Model

Abstract

The chamber method is a standardized testing methods for determining formaldehyde emission from wood-based panels. Due to the complexity and multiple stages involved in the overall detection process, abnormalities are often difficult to perceive when they occur, which may compromise the accuracy of formaldehyde concentration measurements. Therefore, this paper proposes an anomaly identification method for the formaldehyde detection process of wood-based panels, combining Graph Convolutional Neural Network and Bidirectional Long and Short-Term Memory network (GCN-BiLSTM) models. The proposed method first establishes a graph relationship model between the detection time and the operational information of the detection equipment based on the detection standard requirements. Subsequently, the GCN aggregates neighborhood information from the detection equipment nodes to extract graph structural features. The BiLSTM is then used to capture the temporal dependencies of the equipment’s operational features. Finally, the Softmax function is used to obtain the classification results of anomalies. Experimental results demonstrate that the GCN-BiLSTM model effectively identifies anomalies in the formaldehyde detection process of wood-based panels. It exhibits higher detection accuracy compared to other methods. The research presented in this paper provides significant theoretical support and practical implications for the intelligent management of formaldehyde detection in wood-based panels.

Keywords

• Chamber method
• formaldehyde emission
• wood-based panels
• detection process
• anomaly identification
• GCN-BiLSTM