Search

In this study novel integrative machine learning models embedded with the firefly algorithm (FA) were developed and employed to predict energy dissipation on block ramps. The used models include multi-layer perceptron neural network (MLPNN), adaptive neuro-fuzzy inference system (ANFIS), group method of data handling (GMDH), support vector regression (SVR), linear equation (LE), and nonlinear regression equation (NE). The investigation focused on the evaluation of the performance of standard and integrative models in different runs. The performances of machine learning models and the nonlinear equation are higher than the linear equation. The results also show that FA increases the performance of all applied models. Moreover, the results indicate that the ANFIS-FA is the most stable integrative model in comparison to the other embedded methods and reveal that GMDH and SVR are the most stable technique among all applied models. The results also show that the accuracy of the LE-FA technique is relatively low, RMSE=0.091. The most accurate results provide SVR-FA, RMSE=0.034. [ABSTRACT FROM AUTHOR]

Significant wave height (H s) is a critical parameter in the design, operation, and maintenance of nearshore and offshore wind and wave farms. In this study, the original pyramid neural network (PNN) was introduced and applied to predict H s for North Sea renewable energy farms. The performance of PNN was compared with state-of-the-art algorithms. The results show that the machine learning (ML) models trained by the network datasets can accurately predict Hs at any point of nearshore or offshore energy farms, significantly reducing wave monitoring costs. The PNN models predict H s with high accuracy. The results predicted by PNN, with SRMSE of 0.256 and R 2 of 0.873, are more accurate than the corresponding results obtained by the deep neural network (DNN) model, which show that PNNs with their far simpler structures and lower number of parameters are attractive alternatives for complex and time-consuming DNN models. The ensemble of regular models, with SRMSE of 0.247, is ranked as the most accurate method. The results also show that the applied state-of-the-art algorithms including MLR, SNN, DNN, ANFIS, and SVR models predict non-physical negative values of H s. The PNNs developed in this study predict H s with positive values. Moreover, the analysis shows that the scatter of the results obtained by the derived nonlinear models is less pronounced than the scatter of corresponding results obtained by other methods. The results also show that to evaluate the performance of nonlinear ML methods for large datasets, the analysis of standard statistical indices is not sufficient. For large datasets standard statistical indices provide inadequate or insufficient information and additional calculations, and statistical analysis are required. • The pyramid neural network (PNN) was introduced. • The performance of PNN was analyzed in predicting significant wave height (H s) for North Sea renewable energy farms. • The PNN with fewer unknown parameters has the potential to deliver results even superior to those of a deep neural network. • Even the trained models can predict the H s of neighboring farms. [ABSTRACT FROM AUTHOR]

In this study novel clustered machine learning (CML) models were developed by applying different base learners including multi linear regression (MLR), adaptive neuro-fuzzy inference system (ANFIS), classification and regression trees (CART), least-squares boosting (LSBoost), and random forest (RF) to predict scour profile induced by the jet of a propeller. Moreover, the results of CML models were compared with regular ML algorithms. The results show that the tree-based ensemble models, RF and LSBoost, are more accurate than CART. The ANIFS has the worst performance compared with other regular nonlinear methods, RMSE = 17.414 mm. The performance of MLR is also weak, RMSE = 35.63 mm and R2 = 0.088. However, the implementation of CML drastically increases the performance of MLR, RMSE = 15.053 mm, and R2 = 0.871. The CANFIS with RMSE = 8.757 mm, R2 = 0.976, and SI = 0 is ranked as the most accurate model. The results show that CLM enhances the performance of regular ML by up to 59.90%. The application of clustered ML is recommended for other complex problems. Finally, the results show that the derived models trained by other datasets very well predict measured scour profiles which datasets were not used in the training process. • Novel clustered machine learning (CML) models are developed. • Several CML models were used to predict scour profile induced by the jet of a propeller. • The performance of CML were compared with regular models. • The results show that CLM drastically enhances the performance of regular ML. [ABSTRACT FROM AUTHOR]

• A new methodology is derived to determine sediment particle velocities over a sandy rippled bed. • The analysis shows that the derived method nearly perfectly predict observed wave-induced sediment particle velocity. • The results indicate that the ANFIS and CART methods are more accurate than regression equations. • SPBO as a training method is more accurate and PSO. The physics and quantification of sediment transport are still a challenge for scientists and engineers. Measurements of wave-induced sediment velocities may be conducted only in selected laboratories and require weeks of pre-tests and a very experienced team. In this study, student psychology-based optimization (SPBO) algorithm was applied to develop new integrated machine learning methods for the determination of wave-induced non-cohesive sediment particle velocities over a rippled bed by incorporating the outcome of particle image velocimetry. Easily measurable data comprising sediment characteristics, bedform details, and hydrodynamic conditions were used to train the machine learning models. The developed techniques determine well the sediment particle horizontal velocity over a horizontal profile. The analysis shows that the derived models nearly perfectly predict observed data. The proposed methodology provides insight into the physics of sediment processes and may also be applied to interpret measurement data and verify sediment transport models. [ABSTRACT FROM AUTHOR]