Relative optimization potential: A novel perspective to address trade-off challenges in urban energy system planning

2021 
Abstract Urban energy systems often contain multiple optimization sub-objectives with different dimensions. They often need to be normalized in many multi-objective optimization problems. This process may result in the loss of trade-off characteristics of each sub-objectives. This paper introduces a novel indicator, relative optimization potential, to quantify the difference in sensitivity of sub-objectives to the trade-off process, which is easily ignored in previous studies. An integrated multi-objective optimization framework for urban energy systems with flexible decision-making approaches is further proposed to match the differentiated requirements of various scenarios. This framework includes a partial a posterior approach to tackle high-dimensional multi-objective optimization problems, which reduces the computational cost while obtaining equivalent or better results through a reasonable simplification of the decision process. Moreover, a results evaluation method consistent with the perspective of decision-making is proposed to improve the interpretability and consistency of the entire trade-off process. The case studies illustrate the preference characteristics of the proposed approach under different scenarios. And its feasibility is verified by the cross-sectoral planning problem of integrated energy system and building retrofit in different urban areas and scenarios. The proposed approach favors different sub-objectives under different building environments, demand profiles and grid-connection conditions while seeking the maximization of overall relative benefits. Overall, the proposed approach can be applied flexibly in various optimization scenarios to rationally trade-off multi-objective according to different optimization conditions without causing a significant increase in computational cost. This study provides a new tool for multi-objective optimization of urban energy systems with different perspectives.
    • Correction
    • Source
    • Cite
    • Save
    • Machine Reading By IdeaReader
    60
    References
    0
    Citations
    NaN
    KQI
    []