Study On Short-Term Hydrothermal Scheduling Based On Hybrid Particle Swarm

Abundant electric power and environment protection are needed with the development of the national economy. Thus the re-enforcing development effort of hydro electricity is imperative. With the general adoption of the market principle of the power system and the exploitation of the hydropower plants in our country, it is meaning that discussing the scheduling of the cascade hydropower plants in the power market.The short-term hydrothermal scheduling in the power market refers to make fully use of the water resources, release electricity purchasing fee under given consumption of water in every hydro plant and given load capacity of every thermal power station in certain scheduling cycle. Theoretically speaking, the short-term hydrothermal scheduling is a problem with time lag, dynamics, complex constraints. It is difficult to find the optimal solution theoretically. An efficient and high accuracy algorithm can reduce electricity purchasing fee of the hydrothermal, release the system spare, maximize reservoir power capacity under a stable power system, and is a important research direction.Evolutionary algorithm combining particle swarm, genetic algorithm, with evolutionary programming is presented based on the stochastic analysis theory for the model of minimizing the electricity purchasing fee in the power market. The particle swarm optimization algorithm with prior low precision, divergent character and slow late convergence is improved by joining the random variable satisfied the Gaussian distribution to individual extremum, namely searching locally the individual extremum by Gaussian operator. Population's multiplicity is improved by introducing intersection operation for the individual of overlapping the local optimum to avoid trapping in local optimum. It exploit new search space. A satisfied solution in velocity and precision is achieved by adjusting continuously velocity and location to approach optimum.To show its efficiency and effectiveness, the proposed PSO algorithm is applied to the Rastrigin function, Rosenbrock function and a hydro-thermal system including cascade hydro plants. The unit's system load balance, water balance, spinning reserve, operating limits, hydro plant discharge limits,...