Research on Rust Damage Diagnosis Method for Street Light Poles Based on WAESN and CT-IAOA-LightGBM

Street light poles may fall down at any time after rusting, which may threaten the safety of vehicles and pedestrians and cause traffic accidents. In this article, we design and implement a rust diagnosis system based on the wireless acoustic emission sensor network (WAESN) with the main objective of nondestructive detection of rust damage of street lamp tubes: first, the WAESN node samples the acoustic emission (AE) signal of the rusted surface of the lamp pole at high speed, then calculates the characteristic parameters and transmits them wirelessly to the gateway, and then uses the conditional tabular generative adversarial network (CTGAN) to augment the collected AE features, followed by the proposed eXtreme Gradient Boosting-Boruta Shapley additive explanation (XGBoost-BorutaShap) algorithm for feature optimization of the augmented hybrid dataset; furthermore, the chebyshev-t-improved arithmetic optimization algorithm-light gradient boosting machine (CT-IAOA-LightGBM) strategy is proposed to establish an accurate inferred model for identifying rust damage of street light tubes. The experimental comparison results show that the CT-IAOA-LightGBM model constructed based on the preferred feature subset is the most effective in enhancing the performance of lamp pole rust damage diagnosis, with the accuracy, precision, recall, F1-score, macroaverage, microaverage, weighted average, and AUC of 99.32%, 99.43%, 98.25%, 98.61%, 0.9863, 0.9813, 0.9975, and 0.9954, respectively, which are higher than those of the optimization algorithm, such as seahorse combined with the evaluation index of LightGBM model. Finally, the system can reach 99.40% accuracy in the diagnosis and identification of rust damage of the lamp post to be detected, indicating that this method has excellent monitoring efficacy.