Just-in-time (JIT) learning has been widely used for data-driven soft sensor modeling. However, traditional JIT soft sensors do not always function well when applied to complex industrial processes because they are only equipped with a single learning configuration. Therefore, a novel ensemble JIT (EJIT) learning-based soft sensor, referred to as triple-modal perturbation (TP)-based EJIT extreme learning machine (TP-EJITELM), is proposed. In the method, a set of diverse and accurate base JITELM models are generated by using heterogeneous similarity measures and optimizing the model structure and input variables through an evolutionary multiobjective optimization approach. Then, a selective ensemble learning strategy is used to integrate the base models. Compared with traditional JIT soft sensors, TP-EJITELM can significantly improve prediction performance because of the complementary advantages of heterogeneous similarity measures and a good tradeoff between model complexity and accuracy. The effectiveness of the proposed method is demonstrated through two industrial applications.
For micro direct methanol fuel cell (μDMFC), water flooding on the cathode seriously affects the performance stability. Additionally, the effect of material and wettability of the cathode current collector (CCC) on the drainage capacity is studied to improve the μDMFC’s performance. To this end, a CCC with foamed stainless steel was prepared to assemble the μDMFC due to its absorbency. Further, based on analyzing the gas–liquid two-phase flow characteristics of the μDMFC cathode, it was found that the gradient wettability CCC could accelerate the discharge of cathode water. Hence, the foam stainless steel CCC was partially immersed in a KOH solution to complete the gradient corrosion using its capillary force. Then, four different types of gradient wettability CCC were prepared by controlling the time of chemical corrosion. Finally, the performance of the μDMFC with different gradient wettability CCC was tested at room temperature using electrochemical impedance spectroscopy (EIS) and discharge voltage. The experimental results show that the gradient wettability CCC can improve the performance of the μDMFC by slowing down the rate of cathode flooding. The optimum corrosion time is 5 min at a concentration of 1 mol/L. Under these conditions, the CCC has the best gradient wettability, and the μDMFC has the lowest total impedance. The discharge voltage of the μDMFC with corroded CCC is increased by 33.33% compared to the uncorroded CCC μDMFC. The gradient wettability CCC designed in this study is economical, convenient, and practical for water management of the μDMFC.
The diffusion layer (DL) in the structure of the membrane electrode assembly (MEA) of a micro direct methanol fuel cell (μDMFC) plays an essential role in reactant/product mass transfer and catalyst loading. The material selection and structure design of the μDMFC affects its performance. In this work, a reduced graphene oxide/carbon paper (rGO/CP) was proposed and prepared for the anode DL of a μDMFC. It was prepared using electrophoretic sedimentation combined with an in situ reduction method. The rGO/CP reduced the cell’s ohmic and charge transfer resistances. Meanwhile, it provided more significant mass transfer resistance to reduce the methanol crossover, allowing the cell to operate stably at higher concentrations for a longer duration than conventional μDMFCs. The experimental results showed that the maximum power density increased by 53% compared with the traditional anode DL of carbon paper.
Micro Direct Methanol Fuel Cells (μDMFCs) often have application in moveable power due to their green and portable nature.
With the rapid spread of high-power density equipment, the vapor chamber must adapt to more applicable environments and exhibit a better heat transfer performance. The copper mesh and powder are sintered in this work to make a composite wick vapor chamber (CW-VC). Six sections of copper wire are cut and sintered together with the wick as the inner support column of the CW-VC. The effects of filling ratio and inclination on the heat transfer performance of the vapor chamber are investigated. The experimental results showed that the maximum thermal power of the CW-VC is 23.29 W at the optimal liquid filling ratio of 80% and the lowest thermal resistance is 0.33 °C/W. Below the liquid filling ratio of 60%, when the inclination angle of the CW-VC increases, its thermal resistance increases. At filling ratios of 70%, 80%, and 90%, the inclination angle of 30° can reduce the thermal resistance of the CW-VC. It implies that the inclination angle of the CW-VC can be increased appropriately to reduce the thermal resistance when the filling ratio is high. However, when the liquid filling ratio is low, increasing the inclination angle of the CW-VC will make the liquid film of the evaporator thinner and dry out earlier and its thermal resistance will increase.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
customersupport@researchsolutions.com
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.