Study of Pb-based and Pb-free perovskite solar cells using Cu-doped Ni1-xO thin films as hole transport material

SCAPS solar cell simulation program was applied to model an inverted structure of perovskite solar cells using Cu-doped Ni_1-xO thin films as hole transport layer. The Cu-doped Ni_1-xO film were made by co-sputtering deposition under different deposition conditions. By increasing the amount of the Cu-dopant, the film crystallinity enhanced whereas the bandgap energy decreased. The transmittance of the thin films decreased significantly by increasing the sputtering power of copper. High quality, uniform, compact, and pin-hole free films with low surface roughness were achieved. The structural, chemical, surface morphology, optical, electrical, and electronic properties of the Cu doped Ni_1-xO films were used as input parameters in the simulation of Pb-based (MAPbI_3-xCl_x) and Pb-free (MAGeI₃) perovskite solar cells. Simulation results showed that the performance of both Pb-based and Pb-free perovskite solar cell devices significantly enhanced with Cu-doped Ni_1-xO film. The highest power conversion efficiency (PCE) for the Pb-free perovskite solar cell is 8.9% which is lower than the highest PCE of 17.5% for the Pb-based perovskite solar cell.     

Comprehensive numerical investigations on the thermal performance of heat recirculating micro combustor with pin fins for micro-thermal voltaic system applications

Aiming at improving the relatively low energy output and energy conversion efficiency of the micro-thermal voltaic (MTPV) system, an innovative heat recirculating micro combustor with pin fins is designed. The effects of pin fins arrangement, hydrogen/air equivalent ratio on the energy output and performance of CHMC, HMCP and HMCI are compared and investigated. The result shows that when the V_in is 6 m/s and Φ is 1.0, the emitter power of CHMC is 72.76W, and that of HCMP and HCMI micro combustor are 75.99W and 76.35W. and the emitter efficiency of CHMC, HCMP and HCMI is 41.93%, 43.26% and 44.01%. HMCI has better energy output capability compared with CHMC and HMCP. Even though, HMCI brings a higher pressure drop, it is within the acceptable range. When the V_in is 6 m/s, the pressure drop from the pin fins only accounts for 26.4% of the total pressure drop for HMCI. Through the study of equivalent ratio, it is found that HMCI has good adaptability in different equivalent ratio range. This work provides new ideas for the development of MTPV system in the future.     

Investigation of stress fields for non-standard sized glass plates loaded by ring-on-ring

A ring-on-ring (ROR) test is a prevailing test method for evaluating the equi-biaxial strength of glass materials. However, current ROR test standards limit the strength and size of glass to prevent a nonlinear behavior. In this study, the feasibility of ROR testing for non-standard, high-strength glass, such as tempered or ion-exchanged rectangular glass is investigated. To this end, ROR simulation based on theory and experiment is conducted for thirty non-standard glasses with widths of 100–300 mm and aspect ratios of 1.0–2.0. As a result, the maximum measurable stress was about 215.6 MPa for 100 × 200 mm glass and 481.3 MPa for 300 × 600 mm glass with a 3% deviation, which is well above the strength of regular tempered glass. The main purpose of this work is to understand the range of aspect ratio of horizontal and vertical widths of a glass plate that can be evaluated by the standard ROR test.     

Machine fault diagnosis with small sample based on variational information constrained generative adversarial network

In actual engineering scenarios, limited fault data leads to insufficient model training and over-fitting, which negatively affects the diagnostic performance of intelligent diagnostic models. To solve the problem, this paper proposes a variational information constrained generative adversarial network (VICGAN) for effective machine fault diagnosis. Firstly, by incorporating the encoder into the discriminator to map the deep features, an improved generative adversarial network with stronger data synthesis capability is established. Secondly, to promote the stable training of the model and guarantee better convergence, a variational information constraint technique is utilized, which constrains the input signals and deep features of the discriminator using the information bottleneck method. In addition, a representation matching module is added to impose restrictions on the generator, avoiding the mode collapse problem and boosting the sample diversity. Two rolling bearing datasets are utilized to verify the effectiveness and stability of the presented network, which demonstrates that the presented network has an admirable ability in processing fault diagnosis with few samples, and performs better than state-of-the-art approaches.     

基于全U网络的混凝土路面裂缝检测算法

针对现有的混凝土裂缝检测算法在各种复杂环境中检测精度不够、鲁棒性不强的问题,根据深度学习理论和U-net模型,提出一种全U型网络的裂缝检测算法。首先,依照原U-net模型路线构建网络;然后,在每个池化层后都进行一次上采样,恢复其在池化层之前的特征图规格,并将其与池化之前的卷积层进行融合,将融合之后的特征图作为新的融合层与原U-net网络上采样之后的网络层进行融合;最后,为了验证算法的有效性,在测试集中进行实验。结果表明,所提算法的平均精确率可达到83.48%,召回率为85.08%,F1为84.11%,相较于原U-net分别提升了1.48%,4.68%和3.29%,在复杂环境中也能提取完整裂缝,保证了裂缝检测的鲁棒性。     

Learn decision trees with deep visual primitives

In this paper, we strive to propose a self-interpretable framework, termed PrimitiveTree, that incorporates deep visual primitives condensed from deep features with a conventional decision tree, bridging the gap between deep features extracted from deep neural networks (DNNs) and trees’ transparent decision-making processes. Specifically, we utilize a codebook, which embeds the continuous deep features into a finite discrete space (deep visual primitives) to distill the most common semantic information. The decision tree adopts the spatial location information and the mapped primitives to present the decision-making process of the deep features in a tree hierarchy. Moreover, the trained interpretable PrimitiveTree can inversely explain the constituents of the deep features, highlighting the most critical and semantic-rich image patches attributing to the final predictions of the given DNN. Extensive experiments and visualization results validate the effectiveness and interpretability of our method.     

Enhanced wide-activated residual network for efficient and accurate image deblocking

To save bandwidth and storage space as well as speed up data transmission, people usually perform lossy compression on images. Although the JPEG standard is a simple and effective compression method, it usually introduces various visually unpleasing artifacts, especially the notorious blocking artifacts. In recent years, deep convolutional neural networks (CNNs) have seen remarkable development in compression artifacts reduction. Despite the excellent performance, most deep CNNs suffer from heavy computation due to very deep and wide architectures. In this paper, we propose an enhanced wide-activated residual network (EWARN) for efficient and accurate image deblocking. Specifically, we propose an enhanced wide-activated residual block (EWARB) as basic construction module. Our EWARB gives rise to larger activation width, better use of interdependencies among channels, and more informative and discriminative non-linearity activation features without more parameters than residual block (RB) and wide-activated residual block (WARB). Furthermore, we introduce an overlapping patches extraction and combination (OPEC) strategy into our network in a full convolution way, leading to large receptive field, enforced compatibility among adjacent blocks, and efficient deblocking. Extensive experiments demonstrate that our EWARN outperforms several state-of-the-art methods quantitatively and qualitatively with relatively small model size and less running time, achieving a good trade-off between performance and complexity.     

LBM investigation of a Cu-water nanofluid over various configurations of pipes in the mixed convection flow

Increasing the heat capacity of heat exchangers is a crucial need for modern devices. The thermal conductivity of the usual fluids and the Nusselt (Nu) number of flows containing such fluids are two bottlenecks in the way of increasing heat delivery in the heat exchangers. For this reason, nanofluids have been introduced. The effect of utilizing a Cu-water nanofluid as a coolant of two hot pipes in a square cavity is investigated numerically with a two-component lattice Boltzmann method. The volume fraction of nanoparticles is assumed to be constant (0.03) while the Richardson (Ri) number varies from 0.02 to 20. Results show that the effectiveness of nanoparticles is better observed in the natural convection mode. However, sedimentation is also very probable at high Ri numbers, which significantly reduces the effectiveness of the nanoparticles. Configurations which produce a natural convection stream similar to the forced convection one as well as the configurations with high spacing and hence, low heat stream interactions, are the best choices for a uniform heat rate from the pipes.