Research On Multi-scale And Multi-objective Combined Optimal Dispatching Of Cascaded Hydropower Junctions

Oriented to demands of human society and drainage basin development, researchon multi-scale and multi-objective combined optimal dispatching of cascadedhydropower junctions is developed in this paper. And all the results are fully appliedto Three Gorges and Gezhouba cascaded hydropower junctions. Main research workandconclusionsareasfollows:Firstly,cluster genetic algorithm is put forwardto improvingselectionoperatorandcrossover operator. Mutation operator is also improved by bringing in changinggradient of units'operation efficiency. Results of real case show that theimprovements can reduce premature possibility, and optimization is also faster andwithhigherefficiency.Secondly, short-term optimal hydropower dispatching for cascaded hydropowerstations is developed with two different optimization models, which considerefficiency, stability and generation benefit as multi aims. Two-stage optimizationstrategyand synchronous achievement strategyis also proposedtoreducecomplexity.Optimal solutions show that constructed models and optimization strategies ispragmatic. Besides above, dispatching solutions realize all the aims and constraintswithmoregenerationbenefitandhigherefficiency.Thirdly, scheduling solutions of different typical years in drawdown season areconcluded after optimization for Three Gorges and Gezhouba cascaded hydropowerjunctions. Oriented to power market, long-term risk scheduling model is proposedwith min waste water risk, min default risk and max generation benefit as multi aims.Forecasting error of monthly inflow and electricity bidding strategy is simulated byprobabilitydistribution and Monte-Carlo method. Finally, risk scheduling solutions indrawdown season are optimized for Three Gorges and Gezhouba cascadedhydropowerjunctions.Fourthly, couple strategy and intelligent fast decision for navigation andhydropower dispatching is developed. Three coefficients are put forward to evaluatenavigation condition or operation characteristics, which are navigation guaranty rate,level guaranty rate and average level. Fast forecasting model for them is alsoconstructed based on typical simulations. Finally, realization strategies for intelligentfast decision with navigation and hydropower dispatching as multi aims are proposedforcascadedhydropowerjunctions. Fifthly, intelligent fast forecasting of water quality and couple strategy for waterqualityand hydropower dispatchingis developed. Waterflow andwater qualityunderdifferent scheduling modes of Three Gorges hydropower station is simulated by onedimension water and water quality model. Taken daily average change rate ofcontamination concentration as evaluation parameter, intelligent fast forecasting BPneuron network model of water quality is constructed with high precision. Finally,realization strategies for combined optimal dispatching with water qualityimprovement and hydropower dispatching as multi aims are proposed for cascadedhydropowerjunctions.Sixthly, based on the research work of five parts above, multi-scale andmulti-objective combined optimal dispatching model and optimization strategies areproposed and constructed for Three Gorges and Gezhouba cascaded hydropowerjunctions. Finally, dispatching solutions in drawdown season are optimized andconcludedforitafterrealsimulation.