Robust Operational Optimization of a Typical micro Gas Turbine. (February 2019)

• Record Type: Journal Article
• Authors: Paepe, Ward De; Coppitters, Diederik; Abraham, Simon; Tsirikoglou, Panagiotis; Ghorbaniasl, Ghader; Contino, Francesco
• Is Part Of: Energy procedia. Volume 158(2019)
• Page Start: 5795
• Page End: 5803
• Publication Date: 2019-02
• Subjects: micro Gas Turbine -- Deterministic Optimization -- Uncertainty Quantification -- Robust Design; Power resources -- Congresses; Power resources -- Periodicals; Power resources; Conference proceedings; Periodicals; 333.7905
• DOI: 10.1016/j.egypro.2019.01.549
• Languages: English
• ISSNs: 1876-6102

Abstract

Abstract: Due to their high total efficiency and flexibility, micro Gas Turbines (mGTs) offer great potential for use in small-scale distributed cogeneration applications. The economic success of this application; however, fully depends on the optimal usage of the system, which requires careful selection of the number and size of the units in the system and their specific operating strategy. This is only possible if the performance of each individual unit is known precisely. However, in real world operating conditions, the parameters determining the operation and performance of an mGT are only known with a certain uncertainty. Depending on the sensitivity of the model to these parameters, the uncertainties may have a strong negative effect on the performance of the mGT. These uncertainties should thus be taken into consideration by the designers in an early stage of the design process to achieve a so-called robust design. In this paper, we present the robust optimization of a typical mGT, the Turbec T100, operation. This optimization under uncertainties is based on a classical multi-objective optimization scheme linked with an uncertainty propagation technique. In this approach, a robust optimum is found, less sensitive to variations in design and operation parameters. The deterministic optimization results in a Pareto front for maximal electrical efficiency and power output, highlighting that the two objectives are conflicting. The impact of the uncertainties on the parameters is translated into a slight negative shift in this Pareto front. Finally, the most robust operation can be found at a power output of 106.5 kWe, corresponding to a maximal efficiency of 30.6%.