The back propagation neural network [23] is a type of multilayer
feedforward network according to the training of error back propagation
algorithm. It embodies the essence of the artificial neural network,
simulates the activities of human brain neurons, and is self-learning,
self-organizing, self-adaptive, and strong in fault tolerance. The artificial
neural network prediction model also has a strong nonlinear processing
ability [24–25]; it can accurately describe the nonlinear relationship
between factors without determining the functional form.