Surrogate model assisted multi-criteria operation evaluation of community integrated energy systems. (October 2022)

• Record Type: Journal Article
• Authors: Jiang, Jintao; Yu, Hao; Song, Guanyu; Zhao, Jinli; Zhao, Kunpeng; Ji, Haoran; Li, Peng
• Is Part Of: Sustainable energy technologies and assessments. Volume 53:Part C(2022)
• Publication Date: 2022-10
• Subjects: Operational evaluation -- Community integrated energy system (CIES) -- Technique for order preference by similarity to an ideal solution (TOPSIS) -- Surrogate model; Renewable energy sources -- Periodicals; Energy development -- Technological innovations -- Periodicals; Electric power production -- Periodicals; Energy storage -- Periodicals; 333.79
• DOI: 10.1016/j.seta.2022.102656
• Languages: English
• ISSNs: 2213-1388

Abstract

Highlights: A quantified operation evaluation method is developed for CIES to assess its comprehensive performance and the gap to the optimal value. Exergy models are constructed for the CIES to accurately evaluate its energy utilization capability considering energy quality variations. A CNN-based surrogate model is developed to effectively accelerate the evaluation efficiency under massive scenarios. Abstract: Operational evaluation is critical for revealing the operational performance and improvement potential of community integrated energy systems (CIESs). However, complex system configuration and flexible energy conversion in a CIES render the evaluation more challenging. Both the loss in energy quality and quantity should be considered to comprehensively evaluate the energy utilization capability of a CIES. Accurately quantified evaluation requires the optimal solution as the reference, which incurs a heavy computational burden and limits its applicability under massive scenarios. This paper proposes a surrogate model assisted quantified evaluation method for CIES operation considering the variation in energy quality and to show the gap to optimal performance. First, exergy theory is adopted to analyze the matching degree between the input and output energy quality. Second, three indicators are considered as examples to reveal the operational performance in terms of system efficiency, renewable energy penetration, and operation cost. The scores of quantified evaluation are derived using the technique for order preference by similarity to an ideal solution (TOPSIS) method. Third, a CNN-based surrogate model is designed to replace the time-consuming optimization problems for the optimal reference solution, which effectively accelerates the evaluation efficiency under massive scenarios. Case studies demonstrate the effectiveness of the evaluation methods and efficiency of the proposed surrogate model.