A Renewable Energy Assessment Method by Parametric and Non-Parametric Models' Data Analysis

Introduction

Nowadays, if countries can be able to obtain the industrial sector energy at least cost, they can provide a competitive advantage. This situation leads countries to search alternative energy sources, namely renewable energy, due to the limited traditional fossil fuel sources such as oil, natural gas or coal. Renewable energy is an essential alternative energy source because of its environmental utilities and economical efficiencies when compared to traditional energy sources (Dağıstan, 2008). Although there is increasingly widespread use of renewable energy (hydroelectric, wind, solar, biomass, geothermal) in Turkey, a large part of the energy need is still provided from non-renewable (oil, natural gas, coal) energy sources (Koç &Kaya, 2014). On the other hand, compared with European Union countries, climatic conditions of Turkey is an advantage in renewable energy production (Dağıstan, 2008).

This chapter focuses on the stochastic approaches on renewable energy efficiency evaluation in which uncertainty has a technological structure. Decision making problems in energy efficiency field debates with uncertainty, which is affected by prices, energy demand, energy production, consumption, equipment availability, investment and expenditure. Stochastic programming provides an effective framework in which optimization problems under uncertainty are fairly formulated. In energy fields, data envelopment analysis (DEA) and stochastic frontier analysis (SFA) approaches have been used to assess energy sector markets and countries with various types of parameters such as deterministic, stochastic and fuzzy. In such cases, researchers may consider stochastic data as random indicators. By studying on random parameters and analyzing the possibility of uncertain situations, different perspectives of the available information can be obtained in energy markets. The main positive side of studying on random data is determination of accurate energy efficiency policies for future by the help of optimization problems. Parameters of stochastic models are considered random data and this seems to be a reasonable approach for future problems to account for such uncertainties while measuring such data. Analyzing of case study variables can indicates errors and noise. The noise and errors in random indicators generally leads to faulties in frontier production function and in efficiency scores.

When the energy sector researches in the literature are reviewed, it is seen that prominent methods in evaluating renewable energy efficiency are DEA and SFA. DEA was first proposed by Farrell (1957) and developed by evaluating the technical efficiency in the study of Charnes, Cooper and Rhodes. (1978). DEA does not need any assumption of the functional form and can process multiple indicators (Bal & Örkçü, 2005). The following studies can be given as examples to DEA studies in energy field. Jha and Shrestha (2006) measured the performance of hydroelectric power plants by using DEA and they used the current capacity, total number of operations, energy produced by the plant and the number of employees as inputs, total energy produced and winter and summer peak values as outputs in Nepal. In another study Chien and Hu (2007) conducted to compare renewable energy technology among 45 countries level by implementing DEA and stated that Organisation for Economic Co-operation and Development (OECD) members have a greater share of renewable energy resources than non-OECD members. In the study conducted by Barros in 2008, the efficiency of hydroelectric power plants was investigated using DEA. Sözen, Alp and Özdemir (2010) tried to evaluate the performance of thermal electric power plants in Turkey within DEA. The inputs of this chapter were the capacity utilization rate, thermal efficiency, average working time and production capacity, while the outputs were the ton amounts of carbon dioxide, sulfur dioxide, nitrogen dioxide. San Cristobal (2011), is another researcher using DEA to evaluate renewable energy development factors and efficiencies. When this study is examined, it is seen that the inputs were investment rate, implementation period, operating and maintenance costs, while the outputs were composed of power, working hours, service time and carbon dioxide tons. In another study conducted in Turkey by Sözen, Alp and Kilinc (2012), the assessment of efficiency of hydroelectric power plants were examined by DEA. The inputs were the capacity utilization factor, installed capacity, the amount of water collection in the dam, while the output variables were unit cost, operational cost and net energy production.