Short term residential load forecasting using long short-term memory recurrent neural network

<span>Load forecasting plays an essential role in power system planning. The efficiency and reliability of the whole can be increased with proper planning organization. Residential load is indispensable due to its increasing smart grid environment. Nowadays, meters deployed at residential level for collecting historical data consumption residents. Although employment ensures large availability, inconsistency makes it challenging taxing forecast accurately. Therefore, traditional techniques may not suffice purpose. However, a deep learning network-based long short-term memory (LSTM) proposed this paper. powerful nonlinear mapping capabilities RNN time series make effective along higher sequences LSTM. method tested validated through available real-world sets. A comparison LSTM then made two traditionally techniques, exponential smoothing auto-regressive integrated moving average model (ARIMA). Real from 12 houses over three months used evaluate validate performance forecasts performed using mentioned techniques. has achieved best results capability memorizing series-based predictions.</span>