Deep learning ensembles of Long Short-Term Memory Networks to predict ammonia dynamics for aeration control in WWTPs: AdaBoost versus Bagging
Description
To cite the provided dataset for building prediction models, please reference the following paper: Shi H, Wei A, Zhu Y, Tang K, Hu H, Li N. Accurate and robust ammonia level forecasting of aeration tanks using long short-term memory ensembles: A comparative study of Adaboost and Bagging approaches. J Environ Manage. 2024;371:123173. doi: 10.1016/j.jenvman.2024.123173. This data article provides the data set and all the codes for the study. The dataset is the ammonia nitrogen concentration of the aerobic tank in the wastewater treatment plant, which was recorded every 2 minutes from 04 July 2022 to 2010 July 2022, with a total of 5041 samples. The codes included the Adaboost and Bagging techniques, LSTM, CNN-LSTM and the seasonal decompose function.