Deep learned recurrent type-3 fuzzy system: Application for renewable energy modeling/prediction
Yan Cao, Amir Raise, Ardashir Mohammadzadeh, R. Sakthivel, Shahab S. Band, Amirhosein Mosavi
Abstract
A deep learned recurrent type-3 (RT3) fuzzy logic system (FLS) with nonlinear consequent part is presented for renewable energy modeling and prediction. Beside the rule parameters, the values of horizontal slices and membership function (MF) parameters are also optimized. The stability of suggested learning scheme is guaranteed. The proposed method is applied for modeling of both solar panels and wind turbines. By the use of experimental setup and generated real-world date sets, the applicability of suggested approach is shown. Comparison with convectional FLSs demonstrates the superiority of the suggested scheme.
Topics & Concepts
Renewable energyFuzzy logicNonlinear systemScheme (mathematics)Computer scienceStability (learning theory)Artificial intelligenceType (biology)Function (biology)Wind powerEnergy (signal processing)Control theory (sociology)EngineeringMachine learningMathematicsGeologyControl (management)Mathematical analysisPhysicsQuantum mechanicsBiologyEvolutionary biologyPaleontologyStatisticsElectrical engineeringEnergy Load and Power ForecastingSolar Radiation and PhotovoltaicsEnergy and Environmental Sustainability