Tensile creep monitoring of basalt ber-reinforced polymer plates via electrical potential change and arti cial neural network

In this research, Long-Term Tensile Creep (LTTC) failure in Basalt Fiber-Reinforced Polymer (BFRP) composites under ambient conditions was predicted and detected via an expert system in order to monitor the LTTC of BFRP laminated composites. This was accomplished by using a highly accurate, easy to use, and low-cost monitoring method incorporating an Electrical Potential Change (EPC) technique that employs an Electrical Capacitance Sensor (ECS) in conjunction with an Arti cial Neural Network (ANN) to improve the process of detecting and predicting LTTC. A Finite Element (FE) simulation model for Tensile Creep (TC) detection was generated by ANSYS to obtain groups of data for the training of ANNs. The proposed method was then applied to minimize the extent of FE analysis in order to reduce the time required for the monitoring of creep behavior to a minimum. First, creep monitoring at di erent levels of TC (%σc) as a percentage of Ultimate Tensile Strength (UTS) equal to 25%, 50%, and 75% was studied. Subsequently, the trained ANN was utilized to predict the creep behavior at di erent TC levels (%σc) of 15%, 35%, 60%, and 85%, excluded from the FE data. The results showed excellent agreement between FE and predicted results.