The vast majority of Aegean Archipelago Islands cover their continuously increasing electricity demand on the basis of oil-fired autonomous thermal power stations, presenting increased operational cost and power insufficiency. On the other hand, this area has a very high wind potential. However, the stochastic behaviour of the wind and the important fluctuations of daily and seasonal electricity load pose a substantial penetration limit for the contribution of wind energy in the corresponding load demand. The problem investigated in the present study concerns the estimation of the maximum wind energy yield, which is acceptable by an autonomous electrical network, on the basis of the probability distribution of the local grid load demand and the corresponding data related to the available wind potential. For this purpose, an integrated numerical method is developed from first principles. More specifically, the proposed calculation method estimates the maximum wind energy contribution on the basis of the existing wind potential data and the information provided by the system operator concerning the corresponding load demand as well as the operational status of the existing thermal power stations. According to the results obtained, one may state that the present situation imposes a quite narrow limit on the wind energy contributing to the fulfilment of the local societies electrical needs. Hence, only by planning and applying an integrated new strategy concerning the incorporation of new wind power in the local networks, including complementary activities, appropriate energy storage installations and improved electrical load management, it is possible to increase the wind energy participation in the autonomous Islands electrical networks.
The Aegean Archipelago is a remote Hellenic area, east of the mainland, including several hundreds of scattered islands. Unfortunately, the electricity production cost for the vast majority of them is extremely high (Kodossakis et al., 1999) approaching 0.25 €/kWh, due to the utilization of aged autonomous (based on diesel electric generators) power stations (APS). At the same time most Aegean Sea islands are characterized by a considerable annual increase of the electrical power demand exceeding the 5% on annual basis (PPC, 2004). In this context, the existing electrification solution cannot meet with reliability the variable load demand; hence in several cases the existing infrastructure cannot fulfil the excessive power demand during the summer period (Kaldellis et al., 2004a).
On the other hand, these islands, along with the mainland coasts, possess a very high wind potential, since in many locations the average annual wind speed exceeds the 9 m/s. Thus wind energy may be an economic attractive solution for their habitants’ urgent electrification problem (Kaldellis et al., 2005). Unfortunately, the stochastic behaviour of the wind and the remarkable fluctuation of daily and seasonal electricity load, in almost all island grids, lead to substantial wind energy penetration limits (Kabouris and Perrakis, 2000; Kaldellis, 2001), especially during the low consumption periods of the year. In fact, the Island electrical networks manager (i.e. the Greek Public Power Corporation or PPC) defines an instantaneous upper wind energy penetration limit in order to protect the local grid stability in case the wind energy production is suddenly zeroed. This, up to now empirically chosen value, permits the operating thermal power units to replace the wind power contribution without overloading problems or electrical system voltage and frequency fluctuations.
The proposed analysis is concentrated on developing an integrated methodology which can estimate the maximum wind energy contribution to the existing autonomous electrical grids on the basis of stochastic analysis. For this purpose, one takes into account the electrical demand probability density profile of every island under investigation, as well as the operational characteristics of the existing thermal power stations (PPC, 2004). Accordingly, one also uses the corresponding wind potential characteristics, on the basis of the available wind speed probability density profiles (PPC, 1986). Thus, by combining the electrical load with the corresponding wind potential probability values, one may estimate the resulting wind energy contribution to the local network electricity generation. The proposed methodology is applied to a representative Aegean Archipelago island, in order to demonstrate its applicability in similar problems solution. Finally, the proposed analysis is integrated with an appropriate parametrical analysis, investigating the impact of the available wind potential quality on the expected maximum wind energy contribution.
An integrated numerical method, able to estimate the maximum wind energy penetration in the existing autonomous electrical networks, is developed using stochastic analysis, taking into consideration the area wind potential and the corresponding load demand. More precisely, the calculation method developed estimates the maximum wind energy contribution on the basis of the existing wind potential data and the information provided by the system operator concerning the corresponding load demand and the operational status of the existing thermal power stations. For this purpose extensive wind speed and load demand measurements for a considerable time period are taken into account, using the appropriate mathematical tools.
The calculation results indicate that the wind energy absorption by the local network decreases significantly, as the rated power of the installed wind parks increases. Thus, if one wants to amplify the wind energy contribution to covering the energy consumption of autonomous island grids, a considerable part of the wind energy yield cannot be absorbed. The problem is much more intense in regions with high wind potential, since the energy surplus increases with the quality of the available wind potential. Only by finding complementary applications of the wind energy (e.g. desalination or hydrogen production) or by building appropriate energy storage installations, it will be possible to further increase the wind energy participation in similar autonomous electrical markets.
Recapitulating, it is important to note that the proposed methodology gives us the capability to estimate the maximum wind energy contribution to any autonomous electrical network, on the basis of the available wind potential and the operational parameters of the existing thermal power units. Using this model, one may state that the present situation imposes a quite narrow limit for the wind energy contribution to fulfil the electrical needs of the local societies. Unfortunately, this situation minimizes the possibility of new wind parks to be erected in these remote islands without energy storage applications. Hence, only by planning and applying an integrated new strategy, concerning the incorporation of new wind power in the local networks, including complementary activities, appropriate energy storage installations and improved electrical load management, it is possible to increase the wind energy participation in the continuously increasing electricity demand of autonomous islands networks.
