Abstarct: A wide variety of scientific and engineering problems can be formulated as nonlinear optimization problems (NOPs). One promising approach to solve the NOPs with high dimension is to employ artificial neural networks. In this thesis we present two neural network models for solving NOPs. The first model can solve convex nonlinear programming problems (CNPPs) with the constraints of equality and inequality. The second model is derived baxsed on projection function for solving nonconvex nonlinear programming problems (NCNPPs). In addition, we also analyze the existence and the convergence of the trajectory, and the stability properties of the neural networks models. The validity and efficiency of the proposed neural networks are demonstrated by using numerical examples.