渐变刚度板簧式轻卡悬架阻尼参数仿真设计

为了有效提高渐变刚度钢板弹簧式轻型卡车的乘坐舒适性,根据弹簧受力与刚度和变形之间的关系,利用等频原理,建立了渐变刚度钢板弹簧力学特性解析计算模型,并利用钢板弹簧加载-卸载试验对所建立的模型的正确性进行了验证。基于1/2车辆行驶振动模型,利用MATLAB的Simulink工具箱,建立了考虑渐变刚度钢板弹簧力学特性的轻卡行驶振动仿真分析模型。最后,以人体振动舒适性最佳为目标,利用多岛遗传算法,建立了轻卡悬架系统阻尼参数仿真设计方法,并利用实车试验对所建立的轻卡悬架系统阻尼参数仿真设计方法的正确性进行了验证。上述研究为渐变刚度钢板弹簧式轻卡悬架系统参数的设计提供了有效参考。     

基于深度学习的番茄叶部病害识别模型

番茄叶病种类多、成因复杂,其预防和识别难度较大。传统基于机器学习的方法多靠人工识别,需要一定的专家经验,且具有主观性强、识别准确率不高等缺点。为实现番茄叶病特征的自动提取,并提高识别准确率,提出一种基于深度学习的番茄叶病识别模型。该模型基于卷积神经网络对番茄叶部病害特征进行自动提取,获得高维特征后,采用PCA降维算法去除冗余特征;从增大类间距离并减小类内距离的角度改进了softmax分类器,提高了识别准确率。将该模型在CrowdAI提供的数据集上进行了仿真验证,结果表明,该模型能够对番茄叶部常见10种病害进行自动提取特征和识别,综合识别准确率达到95%以上。     

Ethanolic Noni (Morinda citrifolia L.) leaf extract dechlorophyllised using sedimentation process: Antioxidant,antibacterial properties and efficacy in extending the shelf-life of striped catfish slices

Antioxidant and antimicrobial activities of ethanolic noni leaf extract (ENLE) without and with chlorophyll removal by sedimentation method were comparatively investigated. Total chlorophyll content was reduced by 82% in the top fraction (CR-ENLE) collected after 24 h at 4 °C as compared to that of ENLE. Antioxidant and antimicrobial activities were lower in the bottom fraction rich in chlorophyll (Chlo-ENLE) than others (P < 0.05). Based on the microbiological limit, the shelf-life of striped catfish slices pre-treated with 400 mg kg⁻¹ C-R-ENLE was extended to 9 days as compared to the 3 days recorded for the control (without pre-treatment). Slices treated with CR-ENLE had lower lipid oxidation than those treated with ENLE during refrigerated storage (P < 0.05). The sedimentation process was therefore a potential green method for producing ENLE having improved antioxidant and antimicrobial activities without green colour. It can be used as a natural additive for shelf-life extension of fish slices.     

基于层次分析法的叶用甜菜种质资源观赏性评价

根据叶用甜菜品种的观赏特性,运用层次分析法初步建立叶用甜菜品种观赏价值的评价模型,包含叶梗、叶片和整体感3个方面共11个评价因子。结果表明:叶梗是评价叶用甜菜观赏价值的核心要素;叶梗色、叶梗厚、叶丛型是重要因子,分别以红、粉红、粉白叶梗,叶梗厚度厚和直立型叶丛为优良性状。利用该评价模型对8个甜菜品种进行观赏价值综合评价,筛选出‘Clnx-blush2018'、'Clnx-fen2018’和‘荷兰必久'3个观赏价值高、适于市场推广叶品种。     

Mechanochemically assisted fabrication of ultrafine Pd nanoparticles on natural waste-derived nitrogen-doped porous carbon for the efficient formic acid decomposition

Utilizing natural waste as carbon source to prepare porous carbon with ultrahigh surface area and developing a facile protocol to synthesize supported metal nanoparticles toward an efficient formic acid (FA) decomposition are vital but remains challenging. Here, discarded ginkgo leaves were utilized as carbon source to prepare ginkgo leaf-derived porous carbon (GLPC) with an ultrahigh surface area of 3851 m²/g. Based on the as-prepared nitrogen-doped GLPC (N-GLPC) after “soft” nitriding, a facile solid-state reduction strategy with mortar-pestle grinding and without the use of any organic solvent and stabilizing ligand was developed to synthesize ultrafine and well-distributed Pd nanoparticles (NPs) with a diameter of 2.7 ± 0.7 nm. The “soft” nitriding temperature and addition of base during preparation played vital roles in the activity of the fabricated catalysts. The Pd/N-GLPC-350 exhibited the highest catalytic activity toward decomposing FA, achieving a high turnover frequency of 2952 h^?1 at 333 K. The Pd/N-GLPC-350 was quite stable and could be reused at least five times without evident activity loss. This study provides a facile solid-state reduction protocol with mortar-pestle grinding to synthesize metal NPs by using natural waste-derived porous carbon as support toward efficient FA decomposition.     

护顶层保护下采场结构参数优化研究

为了实现护顶措施控制下采场的结构参数优化，采用大型有限元分析软件——ANSYS对康家湾矿深部典型采场顶板的稳定性进行数值模拟、分析，根据结果对当前采矿方案提出修正和调整建议，以便更清楚地掌握深部地压发展变化规律，有效进行地压管理，实现深部顶板不稳固矿体的安全开采。     

NIR光谱法快速分析总黄酮醇苷和萜类内酯

采用NIR法和化学计量学方法构建了银杏叶片中的总黄酮醇苷和萜类内酯含量的定量分析模型。通过PLS建立数学模型,并对预测集样品进行预测。39份样品经交叉验证建立校正模型,RSMECV分别为0.104和0.063。R分别为0.978 8和0.961 9。用13份样品进行预测,R达0.919 9和0.931 8,RSMEP分别为0.107和0.105。该方法简便快速,结果准确,可应用于不同批次银杏叶片样品的进行快速检查或质量控制。