An Improved Decision-making Method For Multi-objective Optimization Generation Scheduling On Power System

With the voice of energy saving,environmental protection and efficient operation increasingly improved,the traditional economic dispatch model only with the purpose of minimizing the primary energy consumption will no longer be a very reasonable criteria.The optimization dispatch model under the environment of electricity market should strive for maximizing the comprehensive economic benefit,which requires decision-makers in the pursuit of minimizing primary energy consumption and taking a number of indicators,such as environmental factor and energy saving into consideration.In this paper, based on analysis of characteristics of hydrothermal operation,from the economy of power system operation,combined with the purpose of energy saving and environmental protection,multi-objective optimization scheduling models of thermal power system and hydrothermal power system are built respectively using the modeling mothod of muti-angle. The objective functions are constituted by three parts,namely minimizing the primary energy consumption,minimizing the emission and minimizing the active power loss. Constraints in the thermal power system consider security and stability of power system as well as the physical and technical limits of thermal units.On the basis of above-mentioned constraints, hydraulic network constraints are joined in hydrothermal power system.From the view of mathematics,the multi-objective optimization scheduling model is a typical multi-objective decision making problem. The processing of muti-objective and selection of optimization algorithm are the keys to the model solution.In the view of the common features of multi-objective problem,an improved multi-objective decision making method is proposed to transform multi-objective to single-objective optimization problem,and then it solved by relevant mathematical optimization algorithm.The incommensurablity and mutual contradiction of each sub-goal are settled by defining single-objective satisfaction degree and comprehensive degree respectively. Drawing on the evaluation function method and the idea of interactive multi-objective decision making,an novel interactive decision making model is built based on evaluation function.Subjective preference information is blended in the model,so scheduling program obtained is easily to meet specific requirements of the desicision making.The Particle Swarm Optimization (PSO) which has the capability of stochastic and global optimization is selected as mathematical algorithm in this paper.Use for reference on the idea of calculating virtual force and acceleration in Artifical Physical Optimization,the basic PSO is improved.The comparision results of several standard test functions between PSO and improved PSO showed the convergence characteristics is improved.In order to verify the validity of the multi-objective optimization constructed and feasibility of the solving method proposed,two examples namely hydro-thermal system and thermal system is selected for simulation analysis.On the one hand,the results of the simulation show that multi-objective optimization scheduling model is better reflect the specific requirement of the decision-maker than traditional economic dispatch model under electricity market and suitable scheduling program is easier obtained in different scheduling criteria. On the other hand, the effective and feasible of the improved multi-objective decision making method combined with PSO for such problem is demonstrated.