Electricity load forecasting plays an important role in capacity planning, scheduling, and the operation of power systems. Reliable and accurate planning and prediction of electricity load are therefore vital. In this study, an approach for forecasting daily electricity demands by wavelet transform and artificial neural networks (ANNs) is proposed. First, Haar wavelet transform is utilized to decompose the load data and eliminate noise. Cuckoo Search algorithm is then used to optimize the parameters of the neural network. The neural network is further trained by back-propagation algorithm to create the forecasting model. Then the proposed approach is applied to forecast the electricity load. Based on the performance criteria calculated, the constructed models have shown high forecasting performances. The obtained results are also compared with those of several other well-known methods including MLR and ARIMA. It shows that the proposed model outperforms the others and provides more accurate forecasting.