Emotional neural networks (ENNs) are a learnable structure that have been inspired from the physiological features of human's emotional brain. In this paper, a single layered ENN is modified in a way to be differentiable for nonlinear system identification problems. In the proposed ENN, thalamus-amygdala expansion link is modeled by a sine and cosine basis function. That is why is named the harmonic emotional neural network (HENN). Due to this proposed expansion, presented method is well-suited for the model based identification methodologies. Simulation results indicate the validity of this claim besides that the supremacy of HENN in terms of accuracy, easy learning and simplicity of model in system identification is shown.