Performance Prediction of Room Air Conditioners and Optimization of Control Strategy for Energy Conservation

摘要

Long-term performance of room air conditioners (RAC) is an important evaluation index to estimate RAC actual energy conservation efficiency. The environmental condition change and long-term use will result in degradation of RAC performance, which increases energy consumption. To solve the problem, an optimization method was proposed. The operating data was collected by a RAC online monitoring system. The hidden relationship among operating parameters was discovered by data mining method, and then an artificial neural network prediction model was developed. After training, this model had been validated to be effective in predicting energy efficiency ratio (EER) with absolute fraction of variance of 0.990 and root mean square error of 0.291. Moreover, the prediction results of return air temperature, relative humidity and EER agreed well with measured values. Based on the prediction results and historical operating database, an optimal control strategy was proposed to improve the operating performance of RAC. By applying the control strategy the energy saving rate was theoretically about 9% in the stable operation case and 2.61% in all actual operating period.