Research On Uncertainty Multi-objective Optimization And Multi Attribute Decision Making Of Low Carbon Power Planning

The scientific composition of power supply is the key factor to effective implementation of low carbon development with Chinese power industry. With the rapid development of low carbon power technology, such as hydropower, wind power, photovoltaic power and other forms of renewable energy and carbon capture and storge technology, there will be a profound impact on the future development of the power industry. Therefore, it is very necessary to study the new pervasive planning model of power supply for the target to low carbon power.The thesis carries on research planning from two aspects with siting and sizing of the low carbon power. It mainly includes the research on modeling and simulation of the carbon capture power plant siting, low carbon power supply planning optimization with containing a variety of power supply forms, uncertainty planning and multi attribute decision making. The main research contents of this thesis are as follows:(1)The study on site optimization modeling of new carbon capture power plant is carried out. The optimization model is established of carbon capture power plant location select. The objective function is minimizing the annual cost in model, considering the constraints, such as CO2 emissions limit, CO2 pipeline transmission limit and CO2 storage conditions. The improved bacterial colony chemotaxis algorithm is used to optimizing. The interval linear programming carbon capture power plant location model is proposed, which aiming at the problem of the uncertainty influence factors of coal source and price, transportation difference, CO2 transmission and storge costs etc. The simulation analyses with different scene mode of the changes of the original parameters are carried out to determine the optimal location of carbon capture power plant. It can provide a scientific basis for decision makers. The evaluation index system of carbon capture power plant site selection is proposed. It is composed with 4 first level indexes and 17 two level indexes including technical, economic, environmental and social indexes. The comprehensive evaluation is developed with using the method of fuzzy comprehensive evaluation based on entropy weight to the optimized scheme.(2)The dynamic programming model is established for a variety of low carbon power supply form including hydropower, wind power, photovoltaic and other renewable energy and carbon capture power plant in the planning period. The target is the minimum cost of investment and operation. The constraint condition includes the CO2 emission reduction. Optimizing for stages in planning period is to determine the optimal decision solution. It could be get of the investment plan of each stage in the planning period, including the type and volume of power supply, in order to achieve the required system emissions reduction target in the at the same time to meet the load demand.(3)The multi-objective robust optimization model of low carbon power planning is put forward in order to solve the uncertainty problem of model and parameters from forms of power. The objective functions are to minimize the annual comprehensive cost and CO2 emission in planning period within boundary system. Considering output uncertainty of wind power and photovoltaic, the effective objective functions are established. The effective objective functions are calculated with using the Latin hypercube sampling of the Monte Carlo method. The Pareto optimal solution meet the condition of robustness is seeked by the discrete bacterial colony chemotaxis algorithm. Aimed at the power supply planning in a region in the next five years, power random operation scheme, discrete bacterial colony chemotaxis optimization scheme and robust optimization scheme are calculated respectively. The optimization analysis results show that when there are some uncertainties in the power supply planning, robust optimization scheme has better anti disturbance ability.(4)The low carbon power planning index system of multi attributes decision-making is put forward, including safety, reliability, economy and environmental protection. There are 4 first level indexes and 17 two level indexes. With application of grey correlation theory to analyze the established index system, we can obtain the indexes which correlation is greater than the threshold value. The evidence theory to determine is used to decide whether screening for index. It realizes the combination of the expert subjective experience and objective theory including the grey relation theory and evidence theory. Using TOPSIS method for multiple attribute decision-making, and considering the effect of each index on the alternative superior degree, the optimal solution in the whole is found with caculating the relative degree of approximation from different schemes to the positive ideal solution.