Research On Multi-objective Day-ahead Economic Dispatch Considering Flexible Load

With the rapid development of the national economy,non-renewable fossil energy sources such as coal are largely consumed.At the same time,a large amount of polluting gases is generated during the combustion process,which causes irreversible harm to the environment.Wind energy has been widely researched and utilized due to its high cleanliness and abundant reserves.However,wind power has the characteristics of volatility and peak inversion.Therefore,the access of wind power to the grid brings some challenges to the safe operation and dispatch of the grid.In addition,the increasing daily peak load has further increased the difficulty of power grid security dispatch.According to the characteristics of wind power,combined with the characteristics of demand side resources,the multi-objective day-ahead economic dispatch problem considering flexible load is studied.The main research contents are as follows:First,the uncertain factors are described in the dispatching system according to the probability distribution functions of wind power forecasting error and load forecasting error.Then,three types of demand-side flexible loads are introduced into the dynamic economic dispatch system,and a multi-objective day-ahead economic dispatch model considering flexible load is established.At the same time,in order to ensure the safe operation of the system,opportunistic constraint planning is used to evaluate the risks caused by the positive and negative rotation reserves that cannot meet the actual wind power fluctuations and load fluctuations,also to optimize the flexible load plan values and the output values of each thermal power unit.In view of the above-established model,the problem is solved and the corresponding Pareto solution set is obtained based on the SMPSO algorithm and the CMOPSO algorithm,respectively.Then,a dual strategy CMOPSO algorithm is further proposed for the improvement of the calculation efficiency and optimization effect of the CMOPSO algorithm.Based on the original learning strategy of the CMOPSO algorithm,an archive set is introduced and an evolutionary search strategy is adopted for the search issues of the archive set.The simulation study is carried out with a classic 10-machine system and the comparison of simulation experiment results show that the dual strategy CMOPSO algorithm has a significant improvement in convergence speed and optimization effect compared with the original CMOPSO algorithm.The simulation results also verify the feasibility of the proposed model.At the same time,the simulation experiment results of flexible load participating in system scheduling optimization and non-participating system scheduling optimization are compared and analyzed.The results show that the flexible load participates in the economic scheduling of the system is feasible and effective.In order to further improve the optimization performance of dual strategy CMOPSO algorithm in solving this problem,adaptive improvements are made to the characteristics of the algorithm optimization,and a two-stage dual-strategy CMOPSO algorithm with synchronous population update is proposed.On the one hand,the original particle update strategy is replaced with a two-stage particle update strategy.On the other hand,the evolutionary search strategy of the archive set is further improved such that the set of particles obtained by the random selection mechanism after the cross mutation can be added not only to the archive set but also to the entire population.A 10-machine system is also employed for simulation research,which verifies that the two improvements of the algorithm are both reasonable and effective.