Research On Unit Combination Optimization Modeling Of Cascade Hydropower Under Complex Conditions

With the cascade development and utilization of water resources,hydropower generation in Southwest China is one of the main power generation methods.In addition to providing single grid load demand,it also meets the load demand of multiple power grids in flood season.This complex load demand,the construction conditions of the power station itself and the fixed output characteristics of the unit,as well as the hydraulic connection between the cascade power stations and the time-varying restricted operation area of the unit It increases the difficulty of load distribution.In this paper,aiming at the highly coupled power plant unit commitment problem,taking Xiluodu and Xiangjiaba cascade power plants as the engineering background,the mixed integer linear programming(MILP)and mixed integer nonlinear programming are established from 24 hours a day,MINLP)and the sensitivity analysis of the two models has achieved the following results:1)A mixed integer linear programming model for unit commitment optimization is proposed.The hydraulic connection between upstream and downstream power stations is described by multi variable hierarchical piecewise linear method,and the unit output function is described by rectangular multi variable piecewise linear method.At the same time,the upper limit of forbidden operation area and water level storage capacity curve of the unit are described by single variable piecewise linear method.Finally,the branch and bound algorithm is used to solve.The first mock exam is based on the load distribution of 26 large units in Xiluodu and Xiangjiaba,and the results are compared with the calculation efficiency of the classical MILP model.The results show that the computation time is approximately the same,but the water consumption is much smaller than the classical model.The final error between the optimal solution and the target lower limit is only 1.4%,which is 1/8 of the classical model error.It is shown that the model has high calculation efficiency.2)A mixed integer nonlinear programming model for unit commitment optimization is proposed.Aiming at the non-linear relationship in the unit combination problem of cascade hydropower stations,polynomial fitting is used to deal with the functional relationship.For the tail water level of hydropower stations considering hydraulic coupling,0-1 variable is used to establish the functional relationship under the influence of water level in front of different downstream hydropower stations.For other non-linear functions,MTLAB is used tofit high-precision curves or surfaces,and MINLP solver is used for calculation.The model verifies its reliability by load distribution of Xiluodu and Xiangjiaba cascade hydropower stations in different situations,and compares it with traditional MILP model in terms of calculation efficiency and accuracy,highlighting the feasibility of the model.3)For the above MILP model and MINLP model sensitivity analysis.From different time scales and error settings,the sensitivity of the above two models is analyzed.Using the same background and the same settings,changing a single variable,from the model variables,constraints to the calculation results,calculation time to analyze the impact of the model,providing a reference value for the establishment of the model later.