Research On Optimal Dispatch Of Hydropower Station Based On Model Predictive Control

Entering the new era,my country’s hydropower development business has ushered in great development,but my country’s problems in river basin operation monitoring and dispatch management have become increasingly prominent.How to effectively improve the water energy utilization efficiency in the operation of reservoirs,reduce the power generation costs of hydropower stations,and improve the optimal operation of hydropower stations This is of great significance to my country’s energy adjustment and transformation in the new era.The model predictive control introduced in this article is one of the most interesting feedback control strategies in recent years.As a prediction-based algorithm,it does not require high model accuracy and is convenient for modeling.Rolling optimization strategies are used to replace the global one.Optimization effectively overcomes the influence of model mismatch and external interference.It has good dynamic performance and has good performance in dealing with multi-variable and constrained problems.In recent years,model predictive control has developed rapidly at home and abroad.It has broken through the limitations of the early framework,and has gradually merged with algorithms in other fields to evolve many types,which are widely used in energy,power systems and other fields.Although model predictive control has developed rapidly and many research results have been achieved,the application of model control methods to hydropower station dispatching has not been refined research.This article describes the research background and research status of model predictive control and hydropower station optimal dispatching.On the basis of the significance of the research,this paper conducts a research on the optimal dispatching of hydropower stations based on model predictive control,and conducts an example verification in conjunction with the Tongzilin Hydropower Station.The main research contents and results are summarized as follows:(1)Study on water level simulation model of Tongzilin Hydropower Station.Due to the relatively small adjustment storage capacity of the Tongzilin Hydropower Station,the large difference between the inflow and outflow flows will have a greater impact on the water level of the hydropower station.The safety of the reservoir water level is controlled by regulating the load of the power station.In addition to following the basin scheduling,Deduction of the water level on the dam of Tongzilin Power Station is essential.This paper establishes a water level simulation model based on water balance and a short-term water level simulation model based on ARIAM.Through simulation analysis of the water level in different time periods of Tongzilin,it is verified that the two models are effective in the deduction of the water level on the Tongzilin Dam.The good simulation accuracy lays the foundation for the subsequent optimization of the Tongzilin Hydropower Station.(2)A brief overview of the in-plant economic operation of hydropower stations is given,and the commonly used space optimization criteria and time optimization criteria in the economic operation of hydropower plants are introduced.Aiming at the specific engineering background of the Tongzilin Hydropower Station,the corresponding economic operation model of the Tongzilin Hydropower Station is established,and the optimization criterion is "powered by electricity",with the minimum water consumption of the power station as the goal,dynamic programming algorithm and successive approximation of dynamics are adopted.The planning algorithm solves the power plant unit combination and optimal load distribution.The feasibility and effectiveness of the above method are verified through examples,and the problem is optimized under the premise of meeting the minimum flow consumption.(3)Study on optimal operation of Tongzilin Hydropower Station Based on model predictive control.This paper describes the basic mechanism of model predictive control and its three basic characteristics,as well as the design process of its predictive model,rolling optimization and feedback correction applied to Tongzilin hydropower station.In the rolling optimization of the optimal control sequence in the future,this paper uses the specific optimization algorithm differential evolution algorithm to solve the optimization problem,and expounds the basic principle of differential evolution algorithm,and uses several typical benchmark functions to verify the convergence speed and convergence performance of differential evolution algorithm in solving the minimum value problem.Combined with the measured data of Tongzilin in 2018,the 24 period rolling optimization verifies that the MPC rolling optimization control decision scheme can effectively control the water level compared with the actual process,improve the safe operation level of the power station,and improve the water energy utilization efficiency of the power station.