Membrane distillation is an emerging separation technology with a high separation factor in water desalination. Ceramic membranes are increasingly used in membrane distillation because of high thermal and chemical stabilities. Coal fly ash is a promising ceramic membrane material with low thermal conductivity. In this study, three hydrophobic coal-fly-ash-based ceramic membranes were prepared for saline water desalination. The performances of different membranes in membrane distillation were compared. The effects of membrane pore size on permeate flux and salt rejection were researched. The coal-fly-ash-based membrane showed both a higher permeate flux and a higher salt rejection than the alumina membrane. As a result, using coal fly ash as the material for membrane fabrication can effectively increase the performance when applied to MD. Increasing the membrane pore size improved the permeate flux, but reduced the salt rejection. When the mean pore size increased from 0.15 μm to 1.57 μm, the water flux rose from 5.15 L·m−2·h−1 to 19.72 L·m−2·h−1, but the initial salt rejection was reduced from 99.95% to 99.87%. The hydrophobic coal-fly-ash-based membrane with a mean pore size of 0.18 μm exhibited a water flux of 9.54 L·m−2·h−1 and a salt rejection of higher than 98.36% in membrane distillation.

In this paper, we mainly focus on the stability of Nash equilibria to any perturbation of strategy sets. A larger perturbation, strong δ-perturbation, will be proposed for set-valued mapping. The class of perturbed games considered in the definition of strong δ-perturbation is richer than those considered in many other definitions of stability of Nash equilibria. The strong δ-perturbation of the best reply correspondence will be used to define an appropriate stable set for Nash equilibria, called SBR-stable set. As an SBR-stable set is stable to any strong δ-perturbation and, various perturbations of strategy sets are not beyond the range of strong δ-perturbation, it has the stability that various stable sets possess, such as fully stable set, stable set, quasistable set, and essential set. An SBR-stable set is stable to any perturbation of strategy sets, so it will provide convenience for study in strategic stability, which is even used to study any noncooperative game.

In recent years, significant progress has been made on the research of crowd counting. However, as the challenging scale variations and complex scenes existed in crowds, neither traditional convolution networks nor near recent Transformer architectures with fixed-size attention could handle the task well. To address this problem, this paper proposes a sceneadaptive attention network, termed SAANet. First of all, we design a deformable attention in-built Transformer backbone, which learns adaptive feature representations with deformable sampling locations and dynamic attention weights. Then we propose the multi-level feature fusion and count-attentive feature enhancement modules further, to strengthen feature representation under the global image context. The learned representations could attend to the foreground and are adaptive to different scales of crowds. We conduct extensive experiments on four challenging crowd counting benchmarks, demonstrating that our method achieves state-of-the-art performance. Especially, our method currently ranks No.1 on the public leaderboard of the NWPU-Crowd benchmark. We hope our method could be a strong baseline to support future research in crowd counting. The source code will be released to the community.

No abstract

No abstract