Loading...

Operation Planning in Restructured Power System Integrating Wind /Solar Energy

Siahkali, Hassan | 2009

669 Viewed
  1. Type of Document: Ph.D. Dissertation
  2. Language: Farsi
  3. Document No: 40518 (05)
  4. University: Sharif University of Technology
  5. Department: Electrical Engineering
  6. Advisor(s): Vakilian, Mehdi
  7. Abstract:
  8. In power system operation, the term "operation planning" carries a wide range of meaning. For some utilities, operation planning denotes primarily short-term planning tasks (up to several days), such as load forecasting, unit commitment, hydro-thermal coordination, transaction pricing, fuel allocation, and security analysis while for other companies operational planning is interpreted in a wider context, and include mid-term planning activities (up to several months), such as maintenance planning, fuel budgeting, rate forecasting, network planning, relay coordination, etc. The mid-term planning provides the link between the long-term and short-term planning programs. The duration of this planning may be one to two years which can be divided into monthly or weekly intervals. Wind power introduces a new challenge to system operators. Unlike conventional power generation sources, wind power generators supply intermittent power because of uncertainty in its resource. In a power system involved large-scale wind power generation scenario, wind intermittency could oblige the system operator to allocate a greater reserve power, in order to compensate the possible mismatch between predicted and the actual wind power output. This would increase the total operation cost. In contrast, generation scheduling (GS) under the restructured power system is more complex and more competitive than in a regulated power system. It is used to create the decision criteria for the generation company. GS objective would be maximizing the profit, rather than minimizing the total production costs. In the past, utilities were obligated to serve all the customers, meeting all the demands and maintaining the reserves. However, under the deregulated system, the GenCo can choose to sell lower amount of power and reserve, than the predicted values, if a higher profit is realized in reduction of the sale amount. Some parameters such as: demand and prices are not known in advance; however are functions of the time, during a day (day of week, and holidays). These parameters are usually predicted. However, forecasts are not easily made because many participants exist in a market. These parameters are associated with a degree of uncertainty that induces risk for GenCo. Therefore, the GenCo must include them in its decision process. This thesis reviews the related works. Different methods for modeling the uncertainty in wind power generation output of a wind farm are discussed. To model this uncertainty, the probabilistic approach and the fuzzy logic approach are implemented. The type-2 fuzzy set is proposed for modeling the linguistic uncertainty. This method is based on application of the different wind power outputs modeling by export persons. This research has illustrated that the fuzzy approach can be more effectively utilized in both vertically integrated and deregulated utility systems, if compared with application of the probabilistic approach. The proposed methods in this thesis are tested against some test systems, realizing the system load, the reserve power and the wind power output uncertainties. Simulations were performed using a GAMS interface to employ the Mixed Integer Nonlinear Programming solver. Other results are obtained using a Matlab-based code which is developed by employing particle swarm optimization (PSO) method. Some of the major contributions of this work are as follows: 1. Presenting a stochastic unit commitment method which integrates wind power generation to the power system. Scenario reduction method is employed to solve the problem with lower computation time. A new decision approach is developed to select the units’ operating state. This can be employed efficiently in unit scheduling. These results are compared with the results obtained using the conventional stochastic approach, that contains some shortcomings in selection of units status to create the feasible solution.
    2. A fuzzy GS method is introduced for a GenCo that integrates the wind power generations. Uncertainty in the power system parameters is simulated by fuzzy sets which are used to model the load, the reserve power, and the wind generation uncertainties. The results of this fuzzy GS problem are compared when expecting different level of profit and different width of membership function for the uncertain parameters. 3- GS is used to determine the generation schedule and the generation maintenance schedule for a GenCo, based on using the given window of applicable maintenance period. In this part of work, the uncertainty in wind generation output is simulated by type-2 fuzzy set to model the linguistic uncertainty of different expert persons. The results of type-1 and type-2 fuzzy modeling are compared and proved that the type-2 model leads to a better GS results. 4- Particle swarm optimization (PSO) method is applied to solve GS problem using a random approach which selects the binary variables. A sharing method is proposed to determine each section of a particle which is implemented in the equality function. A different PSO method, employing a new approach on movement of the particles, is implemented to this GS problem, and the results are compared.
  9. Keywords:
  10. Wind Turbine ; Uncertainty ; Fuzzy Logic ; Fuzzy Set Theory ; Restructured System ; Intermittent Energy Resources ; Type 2 Fuzzy Set

 Digital Object List

 Bookmark

No TOC