Reliable neural circuit can address the field energy in exact form and equivalent Hamilton energy is obtained for theoretical models. Furthermore, the energy dependence on firing activities can be discerned by changing the excitability and forcing current applied on the neuron. In this paper, the distinct energy function for the original Hindmarsh-Rose (HR) neuron is obtained by using Helmholtz theorem and then an equivalent two-variable HR model is proposed. By applying additive noise, stochastic resonance is induced. The intrinsic energy function is approached and energy shunting from each electric component is confirmed. The energy proportion for magnetic field and electric field is changed when mode transition is controlled by the external stimulus. Therefore, the physical property for each electric component in the equivalent neural circuit is predicted, and then the reliability of our neural circuit is confirmed in experiment.