Application Of The Heuristic Rules And Self-organizing Optimization In Unit Commitment

Unit commitment (UC) is the foundation of the economic scheduling and security assess of power system. It can be the key problem of the optimal short-term operation. The study on unit commitment problem has important economic and social benefit. In this dissertation, the heuristic methods were studied for solving unit commitment problem.Firstly, to handle integer variables in the unit commitment problem, approaches proposed in the literature are studied and then some heuristic rules are proposed. A novel identification method for the integer variables was proposed, which can take count of different unit output levels on its cost, ramping constraints, system capacity reserve and transmission line constraints. By calculating the slacked security-constrained unit commitment model with each linear objective function, the proposed rules can then be adopted to obtain the collection of all the units whose running status may be switched during all periods. Based on this method, the optimization range is narrowed efficiently. Under the premise of optimization precision guarantee, adopt Load Curve Specifically Truncated to accelerate the calculation. Tests on two IEEE test systems show the effectiveness of proposed method.A model of the algorithm was proposed to solve the unit commitment based on Self-organizing Optimization. After the studying on the self-organized criticality theory, the Extremal Optimization and Self-organizing Optimization were introduced as theoretical basis. The methods and steps to obtain the groups were described. And then the fitness value of the internal element was defined. The power-law function of the distribution probability was applied, which is the function of the sorted fitness values'serial numbers. Analysis revealed that the set of parameters can control the size of the search space and directions. The effectiveness and superiority of the proposed method have been verified by numerical examples.