Data Fusion-based Structural Damage Identification Model with PCNN Network

During operation, the durability and safety of civil structures are impacted by a number of problems, including structural aging, negative environmental impacts, and other problems. Although civil structures are huge and have complicated types, identifying local damage directly during regular operations is difficult. In addition to possibly presenting certain security problems, the initial damage could totally damage the component's functionality. In order to detect issues early on and evaluate the system's overall safety and dependability, it is crucial to do regular inspections and condition evaluations of engineering structures. This PCNN model is applied for feature learning and pattern recognition. A multi-stage fusion process will initially combine sensor data at the feature level, and subsequently applies PCNN-based fusion to enhance damage detection accuracy. This integration strategy effectively utilize information from heterogeneous data sources. This structural model shows that the proposed method significantly improves the precision and reliability of damage identification compared to conventional models. Therefore, by using this integrating model, damage is detected. This research provides an approach to identify structural damage based on vibration response evaluation, using parallel convolutional neural network (PCNN) and data fusion. Furthermore, this model is able to enhance the visualization of identification in civil structure. As a result, this model shows better time, accuracy, and damage detection results.