A Methodology to Allow Comparison among Different Energy Systems

A Methodology to Allow Comparison among Different Energy Systems

Colli, A.

Ale, B.J.M. (promotor)

Technology, Policy and Management

Engineering and Reflection

2009-05-06

The work proposes a methodology to overcome diversity of risk expressions and allow comparison among different energy systems, being them connected to traditional or innovative technologies. The investigation started with the analysis of ten different energy chains along their steps, which has led to the development of a common frame applicable to all fuel and life cycles. This initial analysis has also defined the most dominant risk scenarios associated to each step of the considered energy chains. Secondly, a set of seventeen risk characterisation indicators (RCIs) has been developed on the backbone of the causal structure for hazard progression in an energy system. The RCIs have been tailored to serve the main characteristics of risk in the most convenient general form that makes the majority of them applicable to all energy systems. An innovative approach for grouping the indicators, and ranking the events to which they are applied, is used, based on the concepts of Boolean logic applied in the form of event and fault trees. This application, based on the mathematical meaning of the probabilistic safety assessment (PSA) technique, has been proven to be satisfactory, leading to the limitation of the subjectivity injection in the grouping and ranking procedure in comparison with the usually adopted multi-criteria decision analysis (MCDA) methods. The uncertainty analysis on the performed numerical application of the described methodology has identified in the input data the major source of uncertainty, while the method itself has an acceptable level of uncertainty that makes it reliable for information processing. In conclusion, the problem of missing consistent risk expressions and information concerning new energy systems adopting new technologies has been addressed. Again, the method proposed relies on the use of PSA. A PSA study developed on a manufacture line for multi-crystalline silicon photovoltaic cells has been performed. Due to lack of specific data from the photovoltaic industry, only a general approach could have been performed, but sufficiently reliable to obtain results in line with the main safety studies in the sector. Moreover, the model is ready for further implementation and could be easily adapted to a real case. Though having the results from a PSA study, the possibility to compare them with possible results coming from the application of RCIs requires a further step involving the transformation of numerical results from both sides into a verbal ranking (such as high, medium, low ranking categories) which makes them comparable. This requires the involvement of expert judgment. The proposed methodological approach is promising. It can be easily adapted to various needs, it could be also used with a different set of indicators. Anyhow, in parallel to the benefits that this methodology could bring, it is also very important to be aware of its limitations, like, for example, the need to interpret the results of the ranking process in a relative view, and not as absolute values.

http://resolver.tudelft.nl/uuid:713e4a00-eb17-49cf-8285-8ac60ea858fb

9789056382131

Institutional Repository

doctoral thesis

(c) 2009 Colli, A.