| dc.description.abstract | The increasing penetration of renewable resources raised a new challenge in power system planning and operation to maintain system reliability. Several serious operational issues may take place when significant amount of renewable power is penetrated into a weak power grid. Short circuit ratio with some modifications has been used to study system strength. The existing short circuit ratio calculation measures have some limitations, and do not consider the realistic electrical connections among multiple renewable resources. The system strength evaluation results obtained from using these measures may not accurately reflect the impact of the interactions among multiple renewable resources at different locations on the system strength. To take the realistic electrical connections among multiple renewable resources into consideration of system strength evaluation study, in this dissertation, a novel Site Dependent Short Circuit Ratio (SDSCR) measure is proposed by analyzing the relationship between the system strength and static voltage stability.
The proposed measure evaluates the system strength based on the power grid structural characteristics and considering the amount of renewable power generation at a point of interconnection. It can be used for various studies concerning system strength evaluation, such as identifying weakest combination of points of interconnection. As another application of the proposed measure, an approach is developed to identify the weakest combination of points of interconnection through structural analysis.
In order to improve the system strength evaluation studies using the proposed measure, this dissertation also proposed an algorithm for more accurate estimation of wind power generation, and an approach for estimation of solar PV power generation drops.
An algorithm for more accurate estimation of wind power generation: an accurate estimation of power curves is essential for assessing the actual output characteristics of a wind farm. The power curve can be estimated using the measured power output data comprising wind power generation and wind speed. However, these measured data are generally ill-distributed due to significant number of outliers, which impose a serious bias challenging estimation of power curves. In this dissertation, an intelligent algorithm is proposed for estimating power curves using the measured data while minimizing modeling and bias errors caused by outliners in the data. Particularly, the proposed algorithm is designed based on statistical analysis software (SAS) package in order to facilitate the analysis of a big dataset.
An approach for estimation of solar PV power generation drops: it becomes more and more clear that dramatic variation of solar power generation will impact the grid reliability and voltage stability. Particularly, in a high penetration scenario, estimation of solar power generation drops is very important for reliability management and operation. In this dissertation, the impact of clouds distribution on solar irradiance, and thus the solar power generation drops are investigated. Then an approach is developed to forecast the clouds distribution using weather data. The approach utilizes time-series weather data including several weather parameters around the solar PV plant of interest to forecast the clouds distribution, and therefore sudden solar power generation drops. | en_US |
