Method On Forecast Operation Of Cascade Hydropower Stations Considering Streamflow Uncertainty And Its Application

With the development of hydropower energy,several clean energy bases with cascade hydropower stations as the main body have been formed,which effectively alleviates the shortage of traditional fossil energy,and ensures the stable attainment of China’s energy transformation and "carbon peak and carbon neutral" goals.Combining the streamflow forecast with the optimal model can effectively guide the power generation of the existing cascade hydropower station and increase the comprehensive benefit.However,in this way,the cascade hydropower stations operation is inevitably affected by the forecast uncertainty which will lead to a deviation between the operation plan and the actual execution result and result in a loss of power benefit or operational risk.Meanwhile,the medium-long-term and short-term optimal models are connected,but the decision-making of the two models is different.How to handle the negative impact of forecast uncertainty on the power generation of cascade hydropower stations in different operation horizons,and give full use to the advantages of the joint operation of cascade hydropower stations,has become a technical problem that needs to be solved urgently.Therefore,focusing on the key scientific issues in the optimal operation of cascade hydropower stations under the streamflow forecast uncertainty,this paper starts from the two aspects of medium-longterm and short-term operation respectively.For medium-long-term streamflow forecasts with large forecast uncertainty,a forecast model that can provide reliable runoff information is developed,and the direction of improvement of the forecast model and the decision-making method combined with the forecast information is studied.For the shortterm streamflow forecast sequence with low forecast uncertainty,the forecast uncertainty analysis is carried out,and the uncertainty analysis results are used to construct a shortterm optimal scheduling model that balances benefits and risks.The main research contents and innovative results are as follows:(1)Since the existing methods focus on improving the accuracy of a single forecast model,this study proposes a deterministic forecast model for medium-long-term reservoir inflow forecasts based on an ensemble Kalman filter.Firstly,prediction models widely used in the hydrological prediction are selected to predict reservoir inflow.Then,the ensemble Kalman filter is used to fuse the ensemble forecast results to obtain the posterior forecast.Results show that the fusion forecast based on the ensemble Kalman filter is superior to all kinds of single forecast models and traditional information fusion forecast methods,and can obtain more accurate and reliable deterministic streamflow forecast.(2)To investigate the impact of long-term streamflow forecasts on the operation of cascade hydropower systems,and guide the improvement of the forecast and decisionmaking,this paper analyzes the influence of different medium-long-term deterministic runoff forecasting methods and optimal models on the power generation of cascade hydropower stations.Firstly,the generalized maintenance of variance extension model is used to generate hypothetical runoff forecasting with different forecast quality.Then,the annual average power generation revenue(CAPR)and system reliability(SR)indicators of cascade hydropower stations under different optimal models are evaluated.The results show that even when using the forecast with the largest uncertainty and bias,the stochastic optimization strategies increase CAPR and SR compared with a reference strategy that uses no forecast information.The SDP performs best with forecast systems that have a negative bias and high accuracy.Compared with the SDP,BSDP increases CAPR and SR and is better able to handle forecast uncertainty,and is insensitive to forecast bias.(3)Previous studies usually use the traditional Copula to describe uncertainty information in the short-term forecasting of streamflow sequences,but such methods are ineffective and unsatisfactory.We adopt Vine Copula to characterize the uncertainty of streamflow forecasting and realize a quantitative evaluation of the uncertainty in the conditions of different streamflow levels and different lead times.And the effect of prior information on the subsequent forecasting uncertainty can also be analyzed.Application of this new method to the Jinxi Reservoir shows that Vine Copula can pass the hypothesis test and achieve the best fitting effect compared to the traditional Copula model,and statistically its simulation error is the lowest against the measured data.When the prior information in the same planning period is used,the expectation of its relative forecasting errors and its 90%confidence level interquartile range can be significantly reduced,thus lowering forecasting uncertainty.(4)For the negative impact of the streamflow forecast uncertainty on the short-term schemes of a cascade hydropower system,we propose a short-term cascade hydropower station optimal operation model based on conditional value-at-risk(called the CVaRSCHOM model).We first establish the conditional probability function to describe the uncertainty of the forecasted streamflow sequence.We next propose the short-term operation target of the cascade hydropower station at the greatest expected benefit.We then introduce the conditional value-at-risk theory to analyze the risk of power shortage in different operating schemes under a certain level of confidence.Finally,based on the risk attitude of the decision-makers,we establish the CVaR-SCHOM model.Research shows that the hydraulic relationships between the cascade hydropower stations are complicated and therefore,the power deviation and forecast error can show positive or negative correlations.Compared with the traditional deterministic optimization model,the CVaR-SCHOM proposed in this paper performs well in real-world applications.