Research On Micro-grid Economic Dispatch Considering Micro-generation's Randomness

Faced with the shortage of resources and serious environmental pollution problems,the development and utilization of renewable energies has become an important trend of the development of the society in the future.As an effective way to make full use of renewable energies,micro-grid can solve a series of problems caused by large scale grid connected distributed generations.Micro-grid is a controllable system which is composed of a variety of distributed generations,energy storage devices and loads by power electronic devices.How to rationally dispatch the output of each micro-generation to meet the demand of loads is an important research content of micro-grid economic dispatch.The main research contents of this paper are listed as follows.Firstly,the paper introduces the domestic and foreign research on the economic dispatch models of micro-grid and the uncertainty of micro-generations.Also,the mathematical models of micro-generations are described in detail,including wind power generation,photovoltaic generation,energy storage device,etc.Moreover,economic dispatch optimization model based on the minimum total operation cost and the lowest pollution treatment cost is described,in which constraint conditions include active power balance constraints,thermal power balance constraints,micro-generation output constraints and climbing rate constraints.Secondly,in view of the randomness of micro-generations as well as the inaccuracy of prediction errors,the scenarios analysis method is used.First of all,the forecast error sample set is obtained by Latin hypercube sampling technique,different scenarios reduction methods are used to reduce the scenarios of samples,finally typical wind power prediction error scenarios are obtained to characterize the random characteristics of the original wind power.The simulation results show that the set of scenarios obtained by bidirectional reduction technique can not only characterize the wind power fluctuation characteristics,but also can forecast the possible scenarios in the future.Also,the influence of wind power randomness on economic dispatch results under different scenarios are discussed.The results show that the higher the quality of scenarios reduction is,the lower the operating cost of the system is,and the more the guidance to the future scheduling is.Finally,the improved PSO optimization algorithm is used to solve the problem of multi-variable and multi-constraint for economic dispatch.In improved PSO,the inertia weight and learning factors are modified,Gaussian disturbance factor is adopted,and the violated particles is re-initialized to improve the performance of particle swarm optimization algorithm.Then,the improved PSO algorithm is used to solve two kinds of economic scheduling models,one is the model only considering the electric load demands,the simulation results verify the effectiveness of the improved PSO algorithm;the other one is the model combining heat and power in micro-grid economic dispatch,the electric boiler and the heat storage tank in the model are involved which not only strengthens the connection between thermoelectric dispatch,but also frees the micro-gas turbine from the previous "heat-set" working model,the simulation results prove the validity of the model and the availability of the improved PSO.