|
Ref. | Year | EC approach | Problem (optimization) | Characteristics |
|
[107] | 2015 | Review: artificial intelligence- (AI-) based techniques. | Forecasting of short-term load. | Comprehensive and systematic literature review of AI-based short-term load forecasting techniques. |
[102] | 2015 | Hybridizing the support vector machines (SVMs) with firefly algorithm (FFA). | Forecasting of solar radiation. | A comparison with artificial neural networks (ANN) and genetic programming (GP) models is provided. |
[114] | 2016 | Ensemble empirical mode decomposition (EEMD) and GA-backpropagation neural network. | Forecasting of wind speed. | EEMD can effectively handle the mode-mixing problem and decompose the original data into more stationary signals with different frequencies. Each signal is taken as an input to the GA-BP neural network model. |
[115] | 2015 | Mutual information, wavelet transform, evolutionary particle swarm optimization (EPSO), and adaptive neurofuzzy inference system. | Forecasting of wind power. | Integration of existing models and algorithms, which jointly show an advancement over the present state of the art, is provided. Results show a significant improvement over other reported methodology. |
[116] | 2016 | Least square support vector machine (LSSVM) and adaptive neurofuzzy inference system (ANFIS). | Forecasting/prediction of dew point temperature of moist air at atmospheric pressure. | GA was applied to optimize the corresponding parameters of these models. Predictions are performed over an extensive range of temperature and relative humidity. |
[92] | 2016 | Multistep approach: cuckoo search (CS) algorithm, fuzzy system (FS), weather research and forecasting (WRF), and ensemble forecast (CS-FS-WRF-E). | Forecasting of wind speed. | CS optimization is used to construct the final model adjusting and correcting the results obtained based on physical laws. The final model yields to best forecasting performance and outperforming all the other models used for comparison. |
[117] | 2017 | Artificial neural network (ANN) coupled with fuzzy clustering method (FCM). Additive linear interdependent fuzzy multiobjective optimization (ALIFMO). Second version of the nondominated sorting genetic algorithm (NSGA-II). | Exergetic optimization of continuous photobiohydrogen production process from syngas by Rhodospirillum rubrum bacterium. | Multiobjective optimization. Objectives: minimization of the normalized exergy destruction and maximization of the rational and process exergetic efficiencies. The solutions of the proposed approach were also compared with conventional fuzzy multiobjective optimization procedures with independent objectives. |
|