Mechanical Properties of Wood Prediction Based on the NAGGWO-BP Neural Network

The existing original BP neural network models for wood performance prediction have low fitting accuracy and imprecise results. We propose a nonlinear, adaptive grouping gray wolf optimization (NAGGWO)-BP model prediction. Firstly, the (GWO) algorithm is optimized. CPM mapping (the Chebyshev method combined with piecewise followed by mod operation) to generate initial populations improve population diversity, an ‘S’-type nonlinear control parameter proposed balance exploitation exploration capabilities of algorithm; strategy also proposed, based on which wolves are divided into predator, wanderer, searcher groups. improved differential evolution strategy, stochastic opposition-based learning oscillation perturbation operator used update positions in different groups convergence speed GWO. Then, weights thresholds optimized using NAGGWO algorithm. Finally, we separately predicted heat-treated wood’s five main mechanical property parameters models. experimental results show that NAGGWO-BP significantly mean absolute error (MAE), square (MSE), percentage (MAPE) specimens, compared BP, GWO-BP, TSSA-BP algorithms. Therefore, this has strong generalization ability good reliability, can fully meet practical engineering needs.