• Energy Research

Abstract Accurate forecasting of electricity prices can provide significant benefits to energy suppliers when allocating their assets and to energy consumers for defining an optimal portfolio. There are numerous methods that efficiently support the forecasting of time series, such as electricity prices, which have high volatility. However, the performance of these approaches varies depending on data sets and operational conditions. In this work, the concept of composite forecasting is presented and implemented in a retrospective study, in real industrial forecasting conditions to show the potential of forecast performance improvement and comparable high consistency of a forecast performance across different ‘Day Peak’ and ‘Day Base’ electricity price data sets for different seasons. As individual methods support vector regression, artificial neural networks and ridge regression are implemented. The forecast performances of these methods are evaluated and compared with their forecast combination using different error measures. The results show that composite forecasting processes with ‘inverse root mean squared error’ combination approach can generate, on average, a more accurate and robust forecast than using an individual methods or other combination schemas.

Abstract Significant progress has been done in the last decades to characterise and define uncertainty in model based planning and decision making in general and in areas like integrated assessment or water resource management. However, existing uncertainty typologies are only partially shared. In city or territory integrated energy planning literature less attention has been paid to uncertainty aspect. Integrated energy planning and model building process have been defined on the base of literature review and the need for consideration of uncertainty is highlighted at the beginning of this work. Using this planning and modelling framework, a conceptual basis of uncertainty showing the allocation of different types of uncertainty according to each planning and modelling stage is provided. Uncertainty concepts proposed in existing typologies of uncertainty from different domains are harmonised into a framework and adapted to the special energy modelling and planning conditions, in a holistic way. Based on this framework, a review of practices in energy planning and modelling shows the gap between needs and practices and raises the question of methodological supports for fulfilling it. The suggested framework can be used to identify and classify different types of uncertainty in context of sustainable model based integrated energy planning in cities or territories, or develop methods to address them.

Abstract Integrated, model-based energy planning particularly in cities and territories involves different planning and modelling activities, which, from a methodological point of view, can be divided into four phases. The analysis and findings of this study focus on planning “phase I”, which is devoted to preparation and orientation. Despite the importance of this planning phase, which is underlined in several papers, only a few studies have addressed planning phase I partially using a systematic methodology. A brief review of planning activities, problems and methods enables mapping the applicability of these methods to their purpose in planning context. The review reveals that no methodological support is provided to fulfil all of the requirements and tasks of this phase. Thus, a methodology for supporting “phase I” activities is presented and illustrated using Singapore as a case study. The methodology combines methods that are either already used in energy planning or borrowed from the area of inventive problem solving, and a specially developed method. The methodology can explicitly reveal problems, key and hidden contradictions, which allows a better understanding of the situation and requirements for the next planning phase especially when looking for solutions beyond common optimality (innovative solutions).