Self-forecasting energy-load stakeholders

The emergence of the Smart Grid brings new opportunities and challenges for all involved stakeholders. Integration of distributed energy resources, in particular renewables, introduces uncertainties in traditional load forecasting which is pivotal towards capitalizing upon the Smart Grid opportunities. This calls for an active contribution of the grid stakeholders and involvement of many locally available assets that can help achieving such goals. However, resources that can actively contribute to reduce the load uncertainties also need to be measurable and therefore predictable. This works presents a system that enables the realisation of Self-Forecasting EneRgy-load Stakeholders (SFERS) that can achieve highly-predictable loads on its own and report them as such to external parties. Accuracy in self-forecast is achieved by absorbing their unpredictability within locally available assets. We investigate the key performance indicators of such systems and the capability of electric vehicles residing on SFERS premises to absorb the forecasting errors. A detailed assessment of various operational conditions is realised by utilizing real-world data and simulating the main system components.