GASEL: Genetic algorithm-supported ensemble learning for fault detection in autonomous underwater vehicles

Artificial Intelligence Models for Predicting Ground Vibrations in Deep Underground Mines to Ensure the Safety of Their Surroundings

“…GA is a class of algorithms inspired by natural selection, crossover, and mutation processes that enables the generation of new solution candidates. These algorithms are therefore ideal for solving complex optimization problems with large search spaces [36].…”

“…(5) The fitness of the next generation is calculated based on selection, crossover, and mutation. Steps 1-4 are repeated, and, finally, the optimal solution is outputted based on the termination condition [36,37]. The mechanism of GA is shown in Figure S2.…”

Machine learning for fault analysis in rotating machinery: A comprehensive review

“…Such test beds are also denoted as machine fault simulators. Some other built an experimental setup to represent the rotating machines or directly consider the real-world rotating machines, including wind turbines [ 13 , 99 , 125 , 130 , [158] , [159] , [160] ], rotor test rigs of aircraft [ 8 ], compressors [ [161] , [162] , [163] ], gas turbines [ 164 , 165 ], steam turbines [ 166 , 167 ], pumps [ [168] , [169] , [170] ], train bogie [ 171 ] and other rotating machine-included systems such as unmanned underwater or aerial vehicles [ 8 , 172 ] for machine learning-based fault analysis. Some recent works regarding the fault diagnosis of the rotating machines mentioned above are presented as follows.…”

“…Some studies in the literature comprise the real-time assessment of their proposed approach by developing an interface where a simulation including data acquisition, data processing, and fault analysis procedures are conducted sequentially regarding the same machine in real-time after training the model [ 22 , 38 , [172] , [173] , [174] ].…”

“…They assessed the performance of their proposed approach by employing Flask environment in Phyton and Postman API to receive the raw data and send it to the fault diagnosis module. They achieved an average accuracy of 99.33% including the prediction of the healthy state and all faulty conditions of the vehicle [ 172 ]. Zhao et al combined transfer learning and embedded convolutional LSTM with digital twin technology for bearing life prediction in wind turbines.…”