On the proper use of net present value for capacity expansion planning with renewable generation

Generation investment costs are significantly decreasing for some renewable technologies as wind or solar power, making them an attractive option for generation capacity expansion.
Nevertheless, most of the available capacity expansion models, consider a single year in the future, or consider a set of them but annualizing investment costs, arbitrarily distributing them
along the plant life-span. This approach is misleading if investment cost varies along the years. This paper introduces a deterministic cost minimization model for capacity expansion planning under the infinite life hypothesis of companies, taking into account simultaneously finite life span of assets by properly including a residual value and considering overall investment costs (overnight costs). This model distributes the recovery cost along the years in an optimal way, different from the uniform annualization alternative.
It is shown that under certain hypothesis the point where the company breaks even (i.e. net present value equals zero) is obtained. The results suggest that uniform annualization overestimates optimal investment and may not be the most adequate option in this context, opening the possibility of a deeper analysis of the optimal recovery cost path for any electrical asset taking into account uncertainty of renewable sources, demand elasticity, corporate taxes and network constraints.