Research On Optimal Planning And Operation Of Distributed Generations And Energy Storage Systems

With intensifying environmental problems,the distributed utilization of renewable energy has been paid more and more attention in the world,but its disorderly access will bring negative impacts to the power systems.The energy storage system is regarded as an effective way to solve the randomness and volatility of intermittent distributed generations because of its charging and discharging’s integration and its rapid response for power demand.However,with the consideration of energy storage system investment cost,the economic problem is especially important in the planning stage.At present,under the low market participation of distributed generations and energy storage systems,the return on investment is difficult to achieve,which is not as conducive to the increasing of social capital.Improving the economy of planning method and promoting the participation in market will have a real impact on reducing their investment costs,improving the return on investment and promoting the healthy development of the whole industry.Based on this,the paper studies the planning and operation method of distributed generations and energy storage systems:In view of the fact that time section method is difficult to describe the fluctuation characteristics of loads and distributed generations,a typical-scenario-based method is proposed for distributed generations planning.Such as temperature,light intensity,wind speed,load and so on,these typical scenarios are constructed by the fuzzy clustering method,as the input data of the distributed generations planning model.Taking the cost of investment and maintenance,the cost of power losses and the cost of purchasing electricity as the objective functions,considering the network security constraints and equipment capacity constraints,the distributed generations planning model is established.The multi-group co-evolution genetic algorithm is used to solve the planning model.And the planning method is verified in the IEEE-33 standard distribution network.Through the Pareto analysis,the alternative planning schemes are provided.In order to solve the problem of increasing extreme scenarios after distributed generations’ access,an optimal method for energy storage systems planning with scenarios error distance is proposed.With the error distance between typical scenarios and actual scenarios described by normal distribution,uncertainties in the process of energy storage system planning are modelled.The ability to smooth power fluctuation is considered in operation layer,and the investment and maintenance cost is considered in planning layer.The objective function and constraints are expressed as the form of expected value in this bi-level planning model.A hybrid intelligent algorithm based on quadratic programming and genetic algorithm is used for the solution method.The modified IEEE-33 standard distribution network is introduced to verify the validity of voltage,power fluctuation and other related expectation constraints,as well as the economy and flexibility of the planning method.In view of the market operation problems of distributed generations,energy storage systems and traditional power supply enterprises,a multi-agent market operation mechanism is proposed.Based on game theory and existence proof of Nash equilibrium solution,a multi-agent market game model with supervision or free competition is constructed.In the solution method,the iterative process is accelerated by parallel calculation in the hardware level.Through a series of simulations,the effects of different regulation and profit levels on the transaction behavior of each agent are analyzed.The validity of the model and the effectiveness of the solution method is verified.