IJSRP, Volume 6, Issue 11, November 2016 Edition [ISSN 2250-3153]
The most used thing in today’s world is energy (power). We use energy in various forms in our daily life like electricity, LPG, solar energy, wind energy, chemical energies in form of batteries and many other forms. Energy performance and its usage have direct impact on consumer’s life. Electrical load forecasting has much importance in the field of smart electrical grids and it plays a vital role in proper planning of electrical load and power systems. It helps in optimization of generating units and dispatching in real time. Forecasting of electricity load demand is a key task for the effective operation and planning of power systems. It was concluded that a comparative study of different model types seems to be necessary. Several models were developed and tested on the real load data .Most of them use a MLP (Multilayer perceptron) network, regressive models, genetic algorithms etc. We carried out short term electrical load forecasting for California Energy Market data, using ANN (Artificial Neural Network) and Plant Identification technique, both techniques implemented in MATLAB software. Error is calculated between actual and forecasted load as MAPE (Mean Absolute Percentage Error).