Using electric vehicles, in addition to decreasing the environmental concerns, can play an important role in decreasing the peak and filling the off-peaks of the daily load characteristics. In other words, in smart grids' infrastructure, the load characteristics can be improved by scheduling the charge and discharge process of electric vehicles. In smart grids, the customers are instantaneously informed about the load and its price and are able to react to the prices. This reaction pattern results in a wide range of changes in the load curve of the network. In this paper, a multi-stage model based on neural networks and the neuro-fuzzy network is presented for forecasting the daily electric load in the price-responsive environment of smart grids. Then, in order to determine the load and generation models of the set of electric vehicles based on the forecasted load for the next day, a complete probabilistic model of these vehicles in the range of parking lots is presented by considering three utilization strategies. These utilization strategies are: Uncontrolled Charging Mode (UCM), Controlled Charging Mode (CCM) and Smart Charge/Discharge Mode (SCDM). Finally, the proposed model is applied on the data of a target day in 2015 in NSW region of Australia's National Electricity Market and the charge and discharge schedule of electric vehicles are determined based on the forecasted load for the next day. The results indicate the most improvement in the daily load factor when the SCDM utilization strategy is employed.
- حق عضویت دریافتی صرف حمایت از نشریات عضو و نگهداری، تکمیل و توسعه مگیران میشود.
- پرداخت حق اشتراک و دانلود مقالات اجازه بازنشر آن در سایر رسانههای چاپی و دیجیتال را به کاربر نمیدهد.