Fig 1. (A) The sub-JWNN structure and (B) the proposed FJWNN model structure.

In this figure, where u₁, u₂, …, u_m are inputs, φ₁, φ₂, …, φ_n are the selected dominant wavelet neurons, W₁,W₂, …,W_n+m are weights of the sub-JWNN output layer, JWNN1, JWNN2, …, JWNN_na are n_a sub-JWNN made from n dominant selected wavelets, η₁,η₂, …,η_na are outputs of the n_a sub-JWNN, v₁,v₂,…, v_na are n_a weights of fuzzy rules, ${\stackrel{-}{\mu }}_1,{\stackrel{-}{\mu }}_2,\dots ,{\stackrel{-}{\mu }}_{na}$ are the membership function values of each rule in FJWNN modeling, and PH is the prediction horizon.