The present study aimed to improve the performance of microbial fuel cells (MFCs) by using an intermittent connection period without power output. Connecting two MFCs in parallel improved the voltage output of both MFCs until the voltage stabilized. Electric energy was accumulated in two MFCs containing heavy metal ions copper, zinc, and cadmium as electron acceptors by connection in parallel for several hours. The system was then switched to discharge mode with single MFCs with a 1000-Ω resistor connected between anode and cathode. This method successfully achieved highly efficient removal of heavy metal ions. Even when the anolyte was run in sequencing batch mode, the insufficient voltage and power needed to recover heavy metals from the cathode of MFCs can be complemented by the developed method. The average removal ratio of heavy metal ions in sequencing batch mode was 67 % after 10 h. When the discharge time was 20 h, the removal ratios of zinc, copper, and cadmium were 91.5, 86.7, and 83.57 %, respectively; the average removal ratio of these ions after 20 h was only 52.1 % for the control group. Therefore, the average removal efficiency of heavy metal ions increased by 1.75 times using the electrons stored from the bacteria under the open-circuit conditions in parallel mode. Electrochemical impedance data showed that the anode had lower solution resistance and polarization resistance in the parallel stage than as a single MFC, and capacitance increased with the length of time in parallel.
With the rapid development of urbanization and industrialization, many developing countries are suffering from heavy air pollution. Governments and citizens have expressed increasing concern regarding air pollution because it affects human health and sustainable development worldwide. Current air quality prediction methods mainly use shallow models; however, these methods produce unsatisfactory results, which inspired us to investigate methods of predicting air quality based on deep architecture models. In this paper, a novel spatiotemporal deep learning (STDL)-based air quality prediction method that inherently considers spatial and temporal correlations is proposed. A stacked autoencoder (SAE) model is used to extract inherent air quality features, and it is trained in a greedy layer-wise manner. Compared with traditional time series prediction models, our model can predict the air quality of all stations simultaneously and shows the temporal stability in all seasons. Moreover, a comparison with the spatiotemporal artificial neural network (STANN), auto regression moving average (ARMA), and support vector regression (SVR) models demonstrates that the proposed method of performing air quality predictions has a superior performance. 相似文献