Research Article

Application of BP Neural Network Improved by Fireworks Algorithm on Suspender Damage Prediction of Long-Span Half-Through Arch Bridge

Table 9

Prediction results of each model for comparative analysis.

Test no.ModelAE (%)AEmax (%)RT/s

1BPNN5.7420.810.90260.49
GA-BPNN4.2613.260.94852.36
PSO-BPNN4.0313.190.95951.42
FWA-BPNN3.599.790.96351.73

2BPNN5.6222.350.91359.16
GA-BPNN4.1112.790.96243.69
PSO-BPNN3.9811.360.96942.58
FWA-BPNN3.249.140.97540.48

3BPNN5.4920.100.92651.38
GA-BPNN3.8311.510.96441.67
PSO-BPNN3.7211.340.97341.09
FWA-BPNN3.078.820.98240.16