MSWNet: A visual deep machine learning method adopting transfer learning based upon ResNet 50 for municipal solid waste sorting

● MSWNet was proposed to classify municipal solid waste. ● Transfer learning could promote the performance of MSWNet. ● Cyclical learning rate was adopted to quickly tune hyperparameters. An intelligent and efficient methodology is needed owning to the continuous increase of global municipal solid waste (MSW). This is because the common methods of manual and semi-mechanical screenings not only consume large amount of manpower and material resources but also accelerate virus community transmission. As the categories of MSW are diverse considering their compositions, chemical reactions, and processing procedures, etc., resulting in low efficiencies in MSW sorting using the traditional methods. Deep machine learning can help MSW sorting becoming into a smarter and more efficient mode. This study for the first time applied MSWNet in MSW sorting, a ResNet-50 with transfer learning. The method of cyclical learning rate was taken to avoid blind finding, and tests were repeated until accidentally encountering a good value. Measures of visualization were also considered to make the MSWNet model more transparent and accountable. Results showed transfer learning enhanced the efficiency of training time (from 741 s to 598.5 s), and improved the accuracy of recognition performance (from 88.50% to 93.50%); MSWNet showed a better performance in MSW classsification in terms of sensitivity (93.50%), precision (93.40%), F1-score (93.40%), accuracy (93.50%) and AUC (92.00%). The findings of this study can be taken as a reference for building the model MSW classification by deep learning, quantifying a suitable learning rate, and changing the data from high dimensions to two dimensions.     

The role of lipids in fermentative propionate production from the co-fermentation of lipid and food waste

● Lipid can promote PA production on a target from food waste. ● PA productivity reached 6.23 g/(L∙d) from co-fermentation of lipid and food waste. ● Lipid promoted the hydrolysis and utilization of protein in food waste. ● Prevotella , Veillonella and norank _f _Propioni bacteriaceae were enriched. ● Main pathway of PA production was the succinate pathway. Food waste (FW) is a promising renewable low-cost biomass substrate for enhancing the economic feasibility of fermentative propionate production. Although lipids, a common component of food waste, can be used as a carbon source to enhance the production of volatile fatty acids (VFAs) during co-fermentation, few studies have evaluated the potential for directional propionate production from the co-fermentation of lipids and FW. In this study, co-fermentation experiments were conducted using different combinations of lipids and FW for VFA production. The contributions of lipids and FW to propionate production, hydrolysis of substrates, and microbial composition during co-fermentation were evaluated. The results revealed that lipids shifted the fermentation type of FW from butyric to propionic acid fermentation. Based on the estimated propionate production kinetic parameters, the maximum propionate productivity increased significantly with an increase in lipid content, reaching 6.23 g propionate/(L∙d) at a lipid content of 50%. Propionate-producing bacteria Prevotella, Veillonella, and norank_f_Propionibacteriaceae were enriched in the presence of lipids, and the succinate pathway was identified as a prominent fermentation route for propionate production. Moreover, the Kyoto Encyclopedia of Genes and Genomes functional annotation revealed that the expression of functional genes associated with amino acid metabolism was enhanced by the presence of lipids. Collectively, these findings will contribute to gaining a better understanding of targeted propionate production from FW.     

Economic simulation modeling of reprocessing alternatives for corn masa byproducts

Increasing production of corn masa for tortillas, chips, and related snack foods is resulting in large quantities of organic residuals requiring environmentally sound management. These byproduct streams appear suitable for use as livestock feed material, thus eliminating landfilling costs. Possibilities for developing livestock feed include direct shipping to livestock feeding facilities, blending prior to shipping, extrusion processing, pellet mill processing, and dehydration. To assess the viability of these options for reprocessing masa byproducts as livestock feed materials, an economic model was developed and applied to each of these alternatives. Through a series of simulation runs with this model, it was determined that direct shipping was by far the most inexpensive means of recycling masa processing residuals (10–57 $/Mg). Other alternatives examined in increasing order of costs included blending prior to shipping, extrusion, pellet mill processing (3–15, 5–18, and 4–18 times greater than direct shipping, respectively), while dehydration was clearly cost-prohibitive (33–81 times greater). Bagged feed was slightly more expensive to produce than bulk feed (1.1 times greater), and reprocessing costs increased as delivery distance increased, due to increased labor, equipment, and fuel costs, but decreased as byproduct generation rate increased, due to the development of the economies of scale. Alternately, based on a tipping fee of 50 $/Mg, the total estimated cost to landfill ranged from 65 to 112 $/Mg. Based on this cost analysis, direct shipping and feeding to livestock is the recycling option of choice for masa processing byproducts. Although specific details of process configurations and associated costs will vary, similar results are likely for other high moisture food processing residuals destined for utilization as livestock feed or components thereof.     

电子废弃物中持久性有机污染物分析检测方法的研究进展

针对目前电子废弃物中持久性有机污染物(POPs)尚未形成有效快速分析检测方法的问题,介绍了溶剂萃取、固相萃取、微波萃取和超临界流体萃取等POPs前处理技术的研究进展,总结了POPs的分析方法,如:气相色谱、液相色谱、气相色谱-质谱法、高分辨率气相色谱-高分辨质谱法等化学分析方法,以及生物传感器测定法、表面胞质团共振检测法、酶联免疫检测法等生物分析方法。分析了各方法的优劣,并提出了展望。     

化学吸收法采集固定污染源氨气的影响因素探究

固定污染源氨气的手工监测,干扰因素较多,其中采样环节尤为关键。实验通过催化氧化-化学发光法考察不同采样管线材质对氨气的吸附效果,离子色谱法考察化学吸收法采集氨气的吸收瓶类型、采样流量、吸收液种类、吸收液浓度及体积等采样条件对氨气吸收效率的影响。实验表明,316 L不锈钢与聚四氟乙烯对氨气的吸附较小,氨气的采样流量不宜超过1.0 L/min,棕色气泡式吸收瓶更适用于氨气样品的采集,磷酸溶液作吸收液对氨气的吸收效率较好,对于低浓度的氨气样品,应采用低流量长时间采集。     

高效液相色谱-串联质谱法同时测定地表水中33种药物

采用固相萃取-高效液相色谱-串联质谱法(SPE-HPLC-MS/MS)建立了地表水中25种抗生素类药物和8种非抗生素类药物的分析方法。通过重点优化质谱参数、色谱条件、样品pH、洗脱溶剂组成及用量等确定了最佳分析条件。水样经过滤、固相萃取柱富集净化后,选择Shim-pack XR-ODS为色谱柱,以乙腈和0.2%甲酸-2 mmol/L乙酸铵-水溶液为流动相进行梯度洗脱,采用电喷雾电离源,在多反应监测模式下(MRM)分析测定,内标法定量。33种药物的仪器定量限为0.012～4.68 ng/L,方法检出限为0.011～7.60 ng/L,地表水加标回收率为53.7%～122%,相对标准偏差为1.22%～32.1%(n=6)。方法成功应用于北京市凉水河12个地表水样分析,共检出32种药物,检出质量浓度为未检出～239 ng/L。利托那韦(RTV)作为新型冠状病毒诊疗方案中推荐的药物在凉水河检出率为100%。     

江苏省钢铁产业绿色低碳技术创新发展路径

钢铁产业作为国民经济的重要基础支柱性产业和高碳排放行业,其绿色低碳转型对实现碳达峰、碳中和目标具有十分重要的作用。在梳理国内外钢铁产业绿色低碳技术进展的基础上,阐述了江苏省钢铁产业绿色低碳技术发展的特点及实现“双碳”目标面临的挑战,提出了由能源结构优化、能效提升、钢化联产协同处置和低碳突破性创新等技术综合集成的绿色低碳技术创新体系发展路径。     

利用粉煤灰处理生活污水的试验研究

以火电厂贮灰场粉煤灰为吸附剂,以生、活污水为吸附对象,通过间歇吸附试验,分析了各因素对粉煤灰吸附性能的影响,并找出了粉煤灰对生活污水中化学耗氧物质(COD)的吸附规律。     

利用木薯渣进行纤维素分解菌混合发酵工艺研究

从自然界中筛选出6株相互间无拮抗作用的纤维素分解菌进行混合发酵,研究了该混菌以工业废弃物木薯渣为原料发酵的最佳产酶条件是:培养基配方为木薯渣4 g,黄豆粉0.75 g,KH2PO40.20 g,NaCl 0.02g,MgSO4·7H2O 0.015 g,蒸馏水1 000 mL;发酵工艺为初始pH=7.0,培养温度40℃,每250 mL锥形瓶装培养液140mL,接种量3%,在140 r·min-1转速下发酵72 h.结果表明,混合菌的发酵可大幅度提高纤维素酶活力及对还原糖的耐受力.     

钻井废液的固液分离工艺与设备

影响钻井废液固液分离的因素包括钻井废液的稀释比例、脱稳处理药剂的种类和加量、固液分离设备。文章介绍了钻井废液处理的基本流程,分析了钻井废液pH值对固液分离效果的影响,论述了卧式螺旋卸料沉降离心机的基本结构、设备参数和操作参数对固液分离效果的影响。