The batch reaction processes have many features such as strong nonlinearity，non-steady-state and fixed reaction time.By taking theadvantage of the latter particularity，a new modeling approach is put forward utilizing a combined B-spline neural networks.The output of controlled objects are often defined as the functions of both manipulated variable and time variable，the description task of the two relationships are undertaken by two neural networks respectively.Using the time variable as B-spline neural network’s input，let its share the function of describing object’s dynamic characteristic.The relationship between input and output variables are operating by another B-spline neural network.Two neural networks combined to establish the nonlinear batch reactor dynamic model.Not only can it simplify the architecture of each network and reduce the number of trained parameters and duration of train，but also can provide a convenient solution for model-based control.The design procedure of the method has been introduced and applied to a modeling process of bulk polymerization of styrene.The feasibility of the approach is studied by simulating experiments.Finally，the control strategies based on the model are derived.