Study On Short-Term Deep Peak-Shaving Operation Of Cascade Hydropower Stations With Strict Hydraulic Connection

Recently,with the increasing peak-to-valley difference of grid load and the large-scale integration of clean energy sources,such as wind power and solar power,characterized by intermittency,uncertainty,and randomness,the peak-shaving pressure of power grid has increased day by day.As the largest clean energy source in China at the moment,hydropower is recognized a high-quality peak-shaving power source due to the advantages such as good regulation performance,low peak-shaving cost,and flexible startup/shutdown of units.Thus,the participation of hydropower in peak shaving is always the research focus and hotspot of power dispatch.To fully utilize the water head drop,most of the river basins in China adopt cascade or single-reservoir and multicascade development mode to build a lot of cascade hydropower stations(CHS)with strict hydraulic connection.How to fully tap peak-shaving potential of this type of CHS and maximize their peak-shaving flexibility in the process of power dispatch is an urgent problem to be solved.This paper takes the CHS in the Hongshui River Basin(HRB)in southwest China as the engineering background,and focuses on the accurate division of peak and valley periods,the dimensionality reduction solution of peak-shaving model,and the peak-shaving strategy of based on water spillage during the wet season in the short-term deep peak shaving process of the cascade reservoirs with strict hydraulic connection.After in-depth research,the main research work and achievements are as follows:(1)A method for dividing peak,flat and valley time periods for short-term peak shaving of CHS is constructed.According to the set calculation interval,adopt the transfer closure fuzzy cluster analysis or the improved fuzzy membership function method based on the membership function as the division method.When the interval is 1h,choose the simple and easy-to-implement but limited by solution scale fuzzy clustering analysis that can realize automatic classification based on the aggregation properties of load data.But when the interval is 15 minutes,the fuzzy membership function method that is not subject to computing scale but is affected by subjective experience is selected,and the threshold evaluation index is improved to select the appropriate threshold.The division results show that the two methods can more truly reflect the actual peak-flat-valley characteristics of the grid load curve.(2)A short-term peak shaving method for CHS under the peak-flat-valley fuzzy recognition mode.Aiming at the problems of complex hydraulic and electrical connections in the short-term peak-shaving model,intertwined various constraints,and serious dimensional disasters in the solution process,a typical peak-shaving characteristic curve method with obvious peak-flat-valley distinction is proposed to realize the dimensionality reduction solution of the peak-shaving model.The determination of the required typical peak-shaving curve is realized based on the fuzzy identification of the peak,flat,and valley periods of the grid load curve and the setting of the peak-flat-valley flow ratios.An example of the dry season dispatching of CHS in the HRB shows that the proposed method can quickly obtain results with better peak-shaving benefits under all constraints,and can flexibly control the peak-shaving depth of hydropower by adjusting the flow ratio.(3)A short-term peak shaving method for cascade reservoirs under prediscarging water mode is proposed.Aiming at the unavoidable water spillage problem of cascade reservoirs in wet season scheduling,a spillage adjustment method based on load characteristics is proposed.A strategy of prediscarging water during valley periods is proposed for cascade peak shaving based on water spillage.And a mixed integer linear programming peak-shaving model with prediscarging water ratio is constructed,and linearization is performed with SOS2 constraint in the solution process.An example of wet season dispatching of CHS in the HRB demonstrates that the listed method can smooth the residual load of power grid to a certain extent,and can flexibly control cascaded peak-shaving capacity by adjusting the prediscarging water ratio of the upstream,middle and downstream hydropower stations.Finally,a brief summary of the full text and the prospects of future research work is proposed.