Title:
Endogenous technological learning. Experiments with MARKAL. Contribution to Task 2.3 of the EU-TEEM project

Samenvatting:
The experiences gained with including endogenous technology learning in the energy optimisation model MARKAL are reported. The objective of the TEEM project is to provide new insights for the European Union (EU) energy Research and Technology Development (RTD) strategy, focusing on policy or market induced progress of energy technologies. Activity 2.3 provides methodological recommendations, numerical results and sensitivity analyses for use in other activities of the TEEM project, based on experience with the MARKAL model with endogenous learning. First, the formulation of learning curves in MARKAL is described. The basic implementation was carried out by the Paul Scherrer Institute (PSI). Some conceptual issues for characterising energy technologies development and introducing technology learning in an energy system model are discussed and used to select technologies with learning from the MARKAL-Europe database. For these selected technologies learning parameters are estimated and introduced in several model cases to test the new endogenous technology learning feature of MARKAL. The main finding is that a full-scale MARKAL model with learning is able to generate globally optimal solutions efficiently. For an optimal use of the benefits of including technology learning, some model database improvements are recommended, such as re-evaluation of the reference energy system and a well-targeted use of bounds and growth parameters. Sensitivity cases show the large impact of assumed learning parameters on model outcomes. Furthermore, model cases are analysed simulating the impact of R, D and D activities and the introduction of a CO₂ tax on technology learning and thus on the market penetration of technologies. 55 p.

Auteur(s):
Seebregts, A.J.; Kram, T.; Schaeffer, G.J.; Stoffer, A. (ECN-Beleidsstudies, Petten (Netherlands))