Robust Model Predictive Control Based Voltage Regulation Method For Active Distribution Network

With the increasing power demand and the growing environmental pressure around the world,large amount of distributed energy sources(DESs)and the energy storage systems(ESSs)are integrated to distribution networks.And the traditional distribution networks are transforming into active distribution networks.The high-level penetration of DESs in distribution networks brings a lot of problems to the safe operation of distribution networks,such as the harmonic pollution problem,power flow problem,over-voltage problem and three-phase unbalance problem.The voltage regulation problem is more severe than other problems.Due to the low X/R ratio of distribution network and the uncertainty of renewable energy outputs,the voltage regulation problem becomes more serious.The voltage regulation strategy should be made to optimize the voltage distribution of network.In recent researches,the outputs of DESs are assumed to be a deterministic value.But the predicted outputs of DESs are not equal to the actual outputs in real operation because of the prediction errors,decreasing the effectiveness of the voltage regulation strategy based on certain value.Some researchers have considered the influence of the prediction errors having on voltage regulation by generating the day-advanced voltage regulation scheme with different control devices,eliminating the prediction errors in long time scale.The influence of prediction errors in short-time scale is still ignored.Therefore,this paper proposed the voltage regulation strategy based on robust model prediction control theory in order to ensure the safe operation of distribution network.The main work in this paper is as follow:Firstly,the characteristics of different voltage control devices were analyzed,including the traditional reactive power regulation devices and the electronic regulation devices.The principles and application scenarios were summarized.And the characteristics for reactive power output of conventional wind turbines,active power output of energy storage systems were analyzed.Secondly,the shortcomings of the existing deterministic voltage regulation model were analyzed.Establish an existing deterministic voltage control model and build it on the SIMULINK platform.The influences of the RESs output uncertainty and prediction errors on voltage regulation were discussed through the simulation analysis of existing model.Thirdly,the voltage regulation optimization model considering the DESs output uncertainty was established.A multi-time step robust optimization model based on the robust model prediction control theory(RMPC)was proposed,satisfying the constraints of power system safe operation during the range of renewable energy output fluctuation.The RMPC theory was used to deal with the effect of prediction error between different time steps on voltage regulation.And the robust optimization theory was used to deal with the impact of renewable energy output uncertainty having on voltage regulation.The two-layer robust optimization model was linearized and processed with strong duality theory,modelled as a multi-time step optimization model which can be solved quickly through quadratic programming.The simulation results on the Finnish distribution network demonstrated the effectiveness of proposed voltage regulation strategy.Fourthly,the multi-time step robust optimization model proposed in this paper was applied to the Finnish distribution network and compared with the existing one-step voltage control model,the deterministic voltage control model based on model predictive control theory and the multi-step voltage control model based on scenarios,verifying the effectiveness of the model and method proposed in this paper.