الفهرس 
INTRODUCTIONOnly 14 pages are availabe for public view
 |  | from | 209 |  |  |
| | | from | 209 | | |
Abstract
Power systems aimed at increasing the share of variable renewable generation (VRG), such as PV and wind, must be more flexible, as one of the most significant challenges associated with this shift is the inherent variability and uncertainty of VRG. In general, the instant balance of temporal inequalities between supply and demand can be reached by many flexibility options. However, there is no accurate quantification of the flexibility needed and available in a power system. Grid operators have successfully balanced the variability of collective power output at a low penetration level of VRG, but as the installations of renewable energy continue growing, the ability to manage these fluctuations becomes a difficult task. At high renewable penetration, more ramps with short durations and small magnitudes occurred, as did the frequency of ramping events.
The power system flexibility is measured by the ability to deal with these ramp events. The literature was based on a binary classification of ramp events, in which the ramp events are similar to each other, which is not true. Therefore, a more accurate classification method will be needed. Power system operator )PSO( should necessarily have statistical information about the power ramping characteristics of VRG, load, and net load. This information helps to mitigate ramping events in the case of a large forecasting error, allowing for improved flexibility, reliability, and cost savings. So far there is no consensus on a precise definition for the ramp event, and few metrics exist to describe the ramping features in a power system, and although several metrics of power system flexibility existed, none of these metrics was taken as a standard due to the dynamics of power systems. Accordingly, for high shares of VRG, it is necessary to have a deeper understanding of the power ramping scale as well as times when the collective ramp events are most likely to occur.
The thesis introduces a method for power ramping analysis. In addition, new metrics describing the ramping features in a power system are introduced. The new metrics are easily applicable in power systems. The thesis also introduces two methods for ramp event classification. The first method depends on the standard deviation scores, and the second depends on the maximum value. Power ramps classification enables PSO to easily know the probability of occurrence of each level or degree of ramps, and the determination of adequate reserves would enhance the efficiency and flexibility of the power system. The thesis proposes a probabilistic methodology to estimate the reserve requirements as a function of historical data and estimates the incremental increase in reserve requirements as the penetration level increases. The methodology is based on the standard deviation of short-term variations and forecast errors as a reserve measure. The thesis investigates the proper distribution of short-term wind variations and forecast errors. Moreover, the thesis estimates the optimal mix of spinning reserve and standing reserve to decrease the number of part-loaded power plants and accommodate more renewable energy. The output power of Belgium’s aggregated wind farms from 2015 to 2020 in a time interval of 15 minutes is used as a practical case study, hour-ahead forecast errors are analysed, and the results of the studied years are compared.