Author Details

As a nation, it is not possible to continue to build new power stations to cater for theever increasing demand for power by residential, commercial, and industrial consumers. Demand-side management provides a positive approach to manage the existing power being supplied by the power utility company, reducing the peak demand through load rescheduling, and to also save cost. Key strategies for the accomplishment of this objective were investigated. Artificial Neural Network was introduced to learn the power demand history of key loads in a university environment in order for the artificial neural network to make accurate decisions based on such learning experiences. Artificial neural network structure used which contains threelayer network having one input neuron, hidden layer with ten (10) neurons and one output neuron, gives better results in terms of mean square error of 0.008 which makes the model developed to be ideal for the representation of the university's load profile. Diversity factor was also taken into account to effectively train the artificial neural network for the major power demand items on campus with such units' load rescheduling carried out and analyzed with very encouraging reduction in peak load demand and overall cost saving.
This work clearly presents benefits in terms of economy of usage and longevity of power equipment. Most importantly, long term cost saving is visible.

Keywords

Power Demand, ANN, Load Profile, Epoch, Forecast, Management, PHCN

Full Text

Username
Password
Remember me