Abstract
For reliability evaluation of power systems, reliability indices are computed. To compute these indices, Generation System Model (GSM) from the generation system and the load model from the load demand information are built, and these two models are convolved to calculate the reliability indices such as Loss of Load Probability (LOLP), Loss of Load Expectation (LOLE), Loss of Load Frequency (LOLF), Expected Unserved Energy (EUE). These indices can be computed for a single area, multi area or composite systems. Similarly in production cost evaluation studies, generation system model is built to evaluate the cost of production of energy or for system expansion purposes. Since the formation of generation system model is basic to these studies, the techniques to build this model are investigated in this thesis. By using the conventional unit addition algorithm, the number of states generated is large and this effects the time taken to build this model The reduction in the number of states results in fewer calculations involved in computing the reliability indices. First this thesis, describes a new method of rounding off the generation capacity states to reduce the number of states in the generation system model without sacrificing the accuracy. The accuracy and the efficiency are studied using three sample systems. Second, the proposed method is compared with the Fast Fourier Transform method in terms of computation time. Since in the literature, there are claims that Fast Fourier Transform method is faster and more efficient than the conventional unit addition algorithm, these claims are investigated. The general expressions to obtain the number of computations involved in each method are derived and they are compared against the real CPU times taken for two sample systems.
Gubbala, Nagalakshmi V. (1994). Investigation of techniques to build generation system models for reliability evaluation. Master's thesis, Texas A&M University. Available electronically from
https : / /hdl .handle .net /1969 .1 /ETD -TAMU -1994 -THESIS -G921.