Electricity Demand Forecasting in Decentralised Demand Side Response for Blocks of Buildings

A downturn of centralised conventional fossil fuel fired power plants and increased proportion of distributed power generation adds to the already troublesome outlook for operators of low-inertia energy systems. The role of an emerging decentralised energy sector allows separation of local energy needs from major power networks. However, the diversion of energy generation means diminishing levels of system inertia on the network needs to be better monitored and controlled. Reducing the overall energy consumption at times of peak demand through demand-side response (DSR) is conducive to intelligent use of electrical power. However, in the absence of reliable demand forecasting measures, effective decentralised demand-side energy planning is often problematic. In this work we formulate a simple yet highly effective approach for forecasting univariate quantitative time series by utilising electricity demand in a decentralised demand-side optimisation model. The forecasting session is constructed initially through analysis of a chronological sequence of discrete observations. Interpretations concerning the generalisation of demand data shows behaviour that allows symbolic representation of the time series. Calculation of short-term forecasting problems have been obtained. Results for medium-term forecasts that extend beyond 12-months are also very promising. In addition to motivating the construction of a forecasting method, the main intention of this article is to derive a practical solution that will evolve a novel proactive approach to existing demand response mechanisms.