In order to meet increasingly stringent discharge standards, new applications and control strategies for the sustainable removal of nitrogen from wastewater have to be implemented. In the past years, numerous studies have been carried out dealing with the application of fuzzy logic to improve the control of the activated sludge process. In this paper, fuzzy control strategies of predenitrification systems are presented that could lead to better effluent quality and, in parallel, to a reduction of chemicals consumption. Extensive experimental investigations on lab scale plant studies have shown that there was excellent correlation between nitrate concentration and ORP value at the end of the anoxic zone. Results indicated that ORP could be used as an on-line fuzzy control parameter of nitrate recirculation and external carbon addition. The optimal value of ORP to control nitrate recirculation and external carbon addition was - 86 +/- 2 mV and - 90 +/- 2 mV, respectively. The results obtained with real wastewater also showed the good performance and stability of the fuzzy controllers independently from external disturbances. The integrated control structure of nitrate recirculation and external carbon addition in the predenitrification system is also presented. 相似文献
Monthly streamflow forecasting is vital for managing water resources. Recently, numerous studies have explored and evidenced the potential of artificial intelligence (AI) models in hydrological forecasting. In this study, the feasibility of the convolutional neural network (CNN), a deep learning method, is explored for monthly streamflow forecasting. CNN can automatically extract critical features from numerous inputs with its convolution–pooling mechanism, which is a distinct advantage compared with other AI models. Hydrological and large-scale atmospheric circulation variables, including rainfall, streamflow, and atmospheric circulation factors are used to establish models and forecast streamflow for Huanren Reservoir and Xiangjiaba Hydropower Station, China. The artificial neural network (ANN) and extreme learning machine (ELM) with inputs identified based on cross-correlation and mutual information analyses are established for comparative analyses. The performances of these models are assessed with several statistical metrics and graphical evaluation methods. The results show that CNN outperforms ANN and ELM in all statistical measures. Moreover, CNN shows better stability in forecasting accuracy.