With increasing restrictions on ship carbon emis-sions,it has become a trend for ships to use zero-carbon energy such as solar to replace traditional fossil ***,uncer-tainties of solar energy and load affect safe and ...
详细信息
With increasing restrictions on ship carbon emis-sions,it has become a trend for ships to use zero-carbon energy such as solar to replace traditional fossil ***,uncer-tainties of solar energy and load affect safe and stable operation of the ship *** order to deal with uncertainties and real-time requirements and promote application of ship zero-carbon energy,we propose a real-time energy management strategy based on data-drivenstochasticmodelpredictive ***,we establish a ship photovoltaic and load scenario set consid-ering time-sequential correlation of prediction error through three *** steps include probability prediction,equal probability inverse transformation scenario set generation,and simultaneous backward method scenario set ***,combined with scenario prediction information and rolling op-timization feedback correction,we propose a stochasticmodelpredictivecontrol energy management *** each scenario,the proposed strategy has the lowest expected operational cost of control ***,we train the random forest machine learn-ing regression algorithm to carry out multivariable regression on samples generated by running the stochasticmodelpredictive ***,a low-carbon ship microgrid with photovoltaic is *** results demonstrate the proposed strategy can achieve both real-time application of the strategy,as well as operational cost and carbon emission optimization performance close to stochasticmodelpredictive *** Terms-data-driven stochastic model predictive control,low-carbon ship microgrid,machine learning,real-time energy management,time-sequential correlation.
暂无评论