Research on the prediction model of cable-stayed bridge force based on attention mechanism

Bridge health monitoring is mainly for long-term automatic measurement of load, environmental effects and structural response of bridges to provide remote moni-toring, online early warning, condition assessment and decision aid services, but there are few services to determine bridge condition for future development trend. In this paper, we use an attention-based long and short-term memory (LSTM) model for predicting the cable-stayed bridge force, which is nonlinear and non-stationary. In this paper, an at-tention mechanism is added to the LSTM model to predict the future cable force values with good accuracy and low mean square error. The similarity between the predicted and real cable force values reaches 0.97, which indicates that this method can be used to predict the cable force values of cable-stayed bridges. The prediction results are better than the LSTM model and the recurrent neural network (RNN) model.