Trace Identification and Visualization of Multiple Benzimidazole Pesticide Residues on Toona sinensis Leaves Using Terahertz Imaging Combined with Deep Learning

Molecular spectroscopy has been widely used to identify pesticides. The main limitation of this approach is the difficulty of identifying pesticides with similar molecular structures. When these pesticide residues are in trace and mixed states in plants, it poses great challenges for practical identification. This study proposed a state-of-the-art method for the rapid identification of trace (10 mg·L⁻¹) and multiple similar benzimidazole pesticide residues on the surface of Toona sinensis leaves, mainly including benzoyl (BNL), carbendazim (BCM), thiabendazole (TBZ), and their mixtures. The new method combines high-throughput terahertz (THz) imaging technology with a deep learning framework. To further improve the model reliability beyond the THz fingerprint peaks (BNL: 0.70, 1.07, 2.20 THz; BCM: 1.16, 1.35, 2.32 THz; TBZ: 0.92, 1.24, 1.66, 1.95, 2.58 THz), we extracted the absorption spectra in frequencies of 0.2–2.2 THz from images as the input to the deep convolution neural network (DCNN). Compared with fuzzy Sammon clustering and four back-propagation neural network (BPNN) models (TrainCGB, TrainCGF, TrainCGP, and TrainRP), DCNN achieved the highest prediction accuracies of 100%, 94.51%, 96.26%, 94.64%, 98.81%, 94.90%, 96.17%, and 96.99% for the control check group, BNL, BCM, TBZ, BNL + BCM, BNL + TBZ, BCM + TBZ, and BNL + BCM + TBZ, respectively. Taking advantage of THz imaging and DCNN, the image visualization of pesticide distribution and residue types on leaves was realized simultaneously. The results demonstrated that THz imaging and deep learning can be potentially adopted for rapid-sensing detection of trace multi-residues on leaf surfaces, which is of great significance for agriculture and food safety.     

基于深度强化学习的光储充电站储能系统优化运行

优化储能充放电策略有利于提升光储充电站运行经济性,但是现有模型驱动的随机优化方法无法全面考虑储能系统的复杂运行特性以及光伏发电功率、电动汽车充电负荷的不确定性.因此,提出一种基于深度强化学习的光储充电站储能系统全寿命周期优化运行方法.首先对储能运行效率模型和容量衰减模型进行精细化建模.然后考虑电动汽车充电需求、光伏出力和电价的不确定性,在满足电动汽车充电需求和光伏消纳的条件下,以光储充电站收益最大化为目标,建立了基于强化学习的储能优化运行问题.考虑到储能充放电决策动作的连续性,采用双延迟深度确定性策略梯度算法进行求解.采用实际历史数据对模型进行训练,根据当前时段状态对储能充放电策略进行实时优化.最后,对所提方法及模型进行测试,并将所提出的方法与传统模型驱动方法进行对比,结果验证了所提方法及模型的有效性.     

考虑样本类别不平衡的电网故障事件智能识别方法

电网中不同设备的故障概率存在差异,影响智能诊断技术的准确性.为解决此问题,提出了一种基于代价敏感学习和模型自适应选择融合的电网故障事件智能识别方法.首先,利用Word2vec模型将预处理后的电网告警信息向量化,并搭建2个双向长短期记忆网络作为基础分类器;然后,设计代价敏感损失函数,将交叉熵损失函数与代价敏感损失函数分别应用于2个分类器中;最后,提出一种模型自适应选择融合法,融合上述分类器,得到故障事件识别结果.实际数据测试表明,所提方法能够有效降低故障事件识别中样本类别不平衡的影响.     

MICROSTRUCTURES AND TEXTURES IN COLD-ROLLED ALUMINIUM STRIP

ＭＩＣＲＯＳＴＲＵＣＴＵＲＥＳＡＮＤＴＥＸＴＵＲＥＳＩＮＣＯＬＤＲＯＬＬＥＤＡＬＵＭＩＮＩＵＭＳＴＲＩＰ①ＮｉｅＺｕｏｒｅｎ，ＺｈａｎｇＸｉｎｍｉｎｇ，ＹｉｎＺｈｉｍｉｎＤｅｐａｒｔｍｅｎｔｏｆＭａｔｅｒｉａｌｓＳｃｉｅｎｃｅａｎｄＥｎｇｉｎｅｒｉ...     

Effect of deep cryogenic treatment on mechanical behavior of a Cu-Cr-Zr alloy for electrodes of spot welding

The effects of deep cryogenic treatment on mechanical behavior of a Cu-Cr-Zr alloy for electrodes of spot welding were investigated employing Brinell-hardness testing unit, abrasion examination machine, electronic almighty testing machine and X-ray stress analyzer. Tensile fracture surfaces of the alloy were characterized by scanning electronic microscope （SEM） with energy dispersive X-ray spectroscopy （EDS）. The results showed that, after deep cryogenic treatment, σb and σ0.2 increased 23 MPa and 21 MPa respectively, the wear rate of the alloy exhibited the trend of decrease with the decreasing temperature and increasing time of deep cryogenic treatment, and the surface residual stress of the alloy was partially eliminated by deep cryogenic treatment.     

EFFECTS OF SPECIFIED BLANK SIZE ON BODY WRINKLING DURING HYDRODYNAMIC DEEP DRAWING OF TAPERED RECTANGULAR BOX

1.~nonHydridapawhcdeepdrawing(HDD)belongstothefieldofsheetmetalfoeingandcanbeaPPliedinthefieldofautomotive,aerospace,drilitaryindustryetc.Ithasbeenrecognizedwidelyintheworld["'j.Fig.Ishowsitsprinciple:TheliqUidthatisfilledinthe~ngdiecavitywillbepressedwhenthepunchgoesdownsothattheeffectoffrictionkeepingandoverflowlubricatingcanbeformed,astherestilt,thefoeinglicitofsheetmetalwillbeimprovedareailvandthegualitVofthefinishedpartswillbeup~ed.BecauseofmanyadvantagesofHDDprocees,ithasbeenwidelys…     

Tool design in a multi-stage drawing and ironing process of a rectangular cup with a large aspect ratio using finite element analysis

Tool design is carried out for a multi-stage deep drawing and ironing process of a rectangular cup with the large aspect ratio using the result of the finite element analysis. The analysis incorporates three-dimensional continuum elements for an elasto-plastic finite element method with the explicit time integration scheme using LS-DYNA3D. The analysis simulates the five-stage deep drawing and ironing process with the thickness control of the cup wall. Simulation is performed in order to investigate the failure by tearing during the forming process at the initial state of tool design. The analysis reveals that the difference of the drawing ratio within the cross section induces non-uniform metal flow which causes severe local extension. The irregular contact condition between the blank and the die also induces non-uniform metal flow which causes local wrinkling. This paper identifies such unfavorable mechanism in the rectangular cup drawing with ironing and proposes a new tool design with the guideline for modification in the design of the process and the sequential tool shape. The finite element analysis result with the improved tool design confirms that the proposed design not only reduces the possibility of failure but also improves the quality of a deep-drawn product. The numerical result shows fair coincidence with the experimental result.     

Supported Molybdenum Carbides Lie Between Metallic and Sulfided Catalysts for Deep HDS

Hydrodesulfurization (HDS) of 4,6-dimethyldibenzothiophene on alumina-supported Mo₂C has been studied. These catalysts are stable and active under deep HDS conditions (0-250 wt ppm S). However, although they are well known to have hydrogenation properties, they lead preferentially to a non-hydrogenated product of the HDS reaction: dimethylbiphenyl. For the same reaction, supported platinum and sulfided molybdenum oxide lead to the hydrogenated products dimethyldicyclohexyl and methylcyclohexyltoluene, respectively. The ranking of HDS activity is as follows: MoS₂/Al₂O₃ < Mo₂C/Al₂O₃ < Pt/SiO₂.     

恒压操作下深层过滤阻力模型的研究

分析了流体在深层滤床中所受的阻力,从经典的达西公式和科泽尼——卡曼方程着手,推导出清洁滤层的过滤阻力及过滤速度表达式,在此基础上结合深层过滤物理模型,引入微元概念,得到恒压操作下截污滤层的过滤阻力和过滤速度数学模型,明确了过滤阻力随各种参数的变化关系,对恒压操作深层过滤的应用有指导意义。     

Volatile compounds in oils after deep frying or stir frying and subsequent storage

Soybean oil (900 g) was heated by deep frying at 200°C for 1 h with the addition of 0, 50, 100, 150 and 200 mL water, and then stored at 55°C for 26 weeks. Soybean oil, corn oil and lard were heated by stir frying and then stored at 55°C for 30 weeks. The volatiles and peroxide values of these samples were monitored. All samples contained aldehydes as major volatiles. During heating and storage, total volatiles increased 260-1100-fold. However, aldehyde content decreased from 62–87% to 47–67%, while volatile acid content increased from 1–6% to 12–33%; especially hexanoic acid which increased to 26–350 ppm in the oils after the storage period was completed. Water addition to the oils heated by deep frying tended to retard the formation of volatile compounds. The total amount of volatile constituents of lard heated by stir frying increased more during storage than that of corn oil or soybean oil. Peroxide values did not reflect the changes of volatile content in the samples.