Personified Behavioural Demand Response Model for the Reduction of Peak Time Energy Consumption Coincidence of Domestic Sector with the Utility

Curtailment of discrete customer’s demand coincidence with utility demand during peak time ends up in good benefits to the utility at different levels as this coincidence is very expensive due to additional requirement of demand. Though few Demand Response(DR) programs are working towards this peak time energy coincidence reduction, they are not that successful due to either requirements of technological installations near customer premises or penalising the customer or lack of encouraging the customer to achieve the reduction. This work proposes a Personified Behavioural Demand Response (P-BDR) model especially for residential customers as they are good contributors of peak time demand. Rather than coaxing or compelling the customer, the proposed model relies on customer’s motivation regarding the peak time energy conservation, setting targets based on their monthly contribution to utility peak time demand and measuring their achievements through feedback models. P-BDR model comprises of Target/Goal setting model based on forecasted data and feedback model based on real time data of individual customer. This model is observed on synthetic smart meter data of 20 discrete domestic customers. For the better application of the model, customers are clustered into 4 categories using K-Means Machine learning algorithm. The model sets an individual target of 5%-15% energy consumption reduction during utility peak time based on the customer classification. The model achieves an overall consumption reduction of 14.9% during peak time with the proposed model.