Close  
Figure 1: (a) The proposed hybrid dynamic wavelet neural network modeling structure and (b) the proposed hybrid dynamic fuzzy wavelet neural network modeling structure, in which I(k−D_{I}), M(k−D_{M}), and G(k−D_{g}) are the exogenous insulin rate, carbohydrate, and blood glucose concentration delayed regressors, respectively; u_{1}, u>_{2},…, u_{m}are the useful selected inputs; ϕ(a_{1}, b_{1}), ϕ(a_{2}, b_{2}), …, ϕ(a_{p}, b_{q}) are all wavelet lattice neurons; ϕ_{1}, ϕ_{2}, …, ϕ_{n}are the selected dominant wavelet neurons; W_{1}, W_{2}, …, Wn are the weights attributed to the dynamic wavelet neural network output layer; WNN_{1}, WNN_{2}, …, WNN_{na} are the n_{a}subwavelets made from the n dominant selected wavelets, v_{1}, v_{2}, …, v_{na}are n_{a}weights attributed to the dynamic wavelet neural network output layer; [INSIDE:8] are membership functions of each rule in the dynamic fuzzy wavelet neural network modeling; and PH is the prediction horizon 
