PERFORMANCE MODELLING AND DECISION ANALYSIS OF DECENTRALIZED ENERGY SYSTEMS AND THEIR IMPACT ASSESSMENT

  • Ting Wu

Student thesis: Phd

Abstract

Renewable energy contributes to attaining the general goal of energy security, affordability and sustainability in a balanced way, and the development trend of future energy should also aim to transfer centralized energy systems to clean and decentralized energy systems while using more renewable energy. Decentralized energy (DE), also called distributed energy, is usually produced close to where it is consumed, in contrast to centralised energy, which is produced at large power plants and transported through the national grid. DE is regarded to be central to the world’s future energy strategies, and it plays an increasingly important role in renewable energy development and economic strategies in many countries. In this thesis, a comprehensive literature review is first conducted on decentralized energy systems and micro-grids, their development status, benefits and challenges, the performance assessment of DE systems, the applications of multiple criteria decision analysis (MCDA) in renewable energy, existing MCDA methods in the performance assessment of DE systems and their merits and limitations. Second, a set of data envelopment analysis (DEA) models are constructed to evaluate the energy efficiency on the country level which takes into account not only energy input and economic output but also non-energy input and undesirable output. The use of DEA models can help decision maker evaluate the efficiency objectively and take effective measures to improve the energy and environmental efficiency of enterprises, industries or regions, and promote energy conservation and achievement of emission reduction goals. Third, a performance modelling and decision analysis model is developed for decentralized renewable energy systems, and this requires the systematic and consistent handling of multiple factors of both a quantitative and qualitative nature under uncertainty. Among alternative MCDA methods, the evidential reasoning (ER) approach is a generic evidence-based MCDA approach and uses a belief structure or so called an extended probability distribution to represent the assessment of an alternative on each attribute as a piece of evidence, regardless whether it is qualitative or quantitative. The aggregation of multiple criteria in the ER approach is through the combination of the extended probability distributions. The weights and reliabilities of assessment information collected from multiple sources can be taken into account consistently. In this way, the ER approach can deal with various types of uncertainty, form a solid basis for sensitivity analysis and provide a panoramic view for informative decision analysis. Thus in this research the ER approach is implemented systematically in the context of analysing the performance and impact of DE systems. Furthermore, two real case studies are conducted respectively to validate the practicality of the proposed performance modelling and decision analysis methods. One is a small-scale micro-grid in an industrial park, which includes different kinds of renewable energies. The other one is a large micro-grid cluster project in Inner Mongolia, located at the northwest of China. The key findings are discussed from the systematic performance modelling and impact analysis of DE systems on the above case studies. It is believed that multiple stakeholders can potentially benefit from these research findings, including policy makers, energy suppliers and consumers, energy network owners, and DE investors and stakeholders in local communities, who have direct interests in the generation, transition and consumption of renewable energy. In the future work, this research can be linked closely with specific decision contexts in order to support informed decision-making from multiple dimensional renewable energy performance evaluation. In addition, more detailed and comprehensive evidence combination rules can be developed to better characterise various types of uncertain data and information in the evaluation of various DE systems.
Date of Award1 Aug 2021
Original languageEnglish
Awarding Institution
  • The University of Manchester
SupervisorJian-Bo Yang (Supervisor) & Dong Xu (Supervisor)

Keywords

  • Decentralized energy sytstem
  • Evidential Reasoning
  • Energy efficiency
  • Data Envelopment Analysis
  • Performance modelling
  • Decision making

Cite this

'