Search

Usually, a typical wind turbine system cannot generate electricity power consistently, because the power outputs are heavily depended on wind speed. However, terrain, temperature, humidity and other factors can also affect wind speed. Therefore, wind power forecasting is a complex, multi-dimensional, and highly non-linear system. Neural network is able to learn the relationship between system inputs and outputs without mathematical conversion, and perform complex non-linear mapping, data classification, knowledge processing, and so forth. In addition, neural network also has the ability of parallel processing to reduce computing time, so it is suitable for wind power forecasting. The purpose of this paper is to use neural network technology to design a wind power forecasting system. Moreover, the efficiency analysis of the proposed wind power forecasting system in Kinmen farm is described. Finally, we use MATLAB to implement the proposed wind power forecasting system in Kinmen farm, which is capable of forecast within 48-hours ahead.

Germanium (Ge) thin films and Ge-rich silicon-germanium (SiGe) alloys have potential for lowering the manufacturing cost of photovoltaic (PV) devices especially in tandem solar cells. This thesis focuses on the fabrication and characterization of Ge thin films and Ge-rich SiGe alloys prepared by radio-frequency (RF) magnetron sputtering, an inexpensive, non-ultra high vacuum deposition technique capable of fabricating large area films. The motivation is given firstly, followed by a brief review of Ge in PV applications. Secondly, fabricating polycrystalline Ge (poly-Ge) thin films on glass by RF magnetron sputtering is investigated. In addition, in situ hydrogenation is applied in an attempt to further improve the properties of the Ge films. The influence of hydrogen on the deposition rate, surface morphology, and structural, optical, as well as electrical properties of poly-Ge films is explored. Moreover, to demonstrate the potential of the in situ hydrogenated poly-Ge (poly-Ge:H) films in PV applications, the doping of poly-Ge:H thin-films on glass is studied. P type films were deposited and in situ doped by co-sputtering Ge:H with boron (B) at various power levels in a mixture of argon and hydrogen at 500C followed by a rapid thermal anneal (RTA) process. On the other hand, n type films were deposited and ex situ doped by firstly sputter-depositing a Ge:H layer and then a SiO2/P2O5+SiO2/SiO2 sandwich structure capped with a SiNx layer, and finally followed by a thermal drive-in process with RTA. The evidence for successfully p-type and n-type doping of poly-Ge:H films is presented. Furthermore, the research scope is extended into polycrystalline Ge-rich SiGe alloys to explore the properties of the Ge alloyed with small amounts of Si by RF sputtering. Finally, a novel method for growing thin relaxed single crystalline Ge heteroepitaxial layers on Si substrates by using RF sputtering was developed. By using this method, the need of ultra-high-vacuum condition and the use of costly and extremely toxic germane gas are avoided. Therefore, thin relaxed single crystalline Ge epitaxial layers on Si substrate can be obtained at low cost, making the resultant Ge layer a potential virtual substrate for III-V material growth for tandem cell applications.