Application of artificial neural network for electrical energy management of air-conditioning systems

In this paper, an artificial neural network (ANN) is employed to control the speed of compressor for accomplishing the electrical energy management of air-conditioning systems. On the basis of "the effective temperature figure" by ASHRAE, the theory of enthalpy was utilized to obtain both comfortable temperature point and minimizing the enthalpy of air-conditioning systems. The technology of the Levenberg-Marquardt (LM) procedure and radial basis function (RBF) neural network were introduced to perform the training and learning features. A practical air-conditioner with variable frequency control was used to measure the on-line room temperature and humidity change. The simulations of the ANN network have a good result with this standard and make possible an anticipated and optimal control strategy to balance thermal comfort, energy saving, and reliability.