Fluorinated semi-interpenetrating polymer networks for enhancing the mechanical performance and storage stability of polymer-bonded explosives by controlling curing and phase separation rates

Herein, the effect of fluoropolymer binders on the properties of polymer-bonded explosives(PBXs) was comprehensively investigated. To this end, fluorinated semi-interpenetrating polymer networks(semiIPNs) were prepared using different catalyst amounts(denoted as F23-CLF-30-D). The involved curing and phase separation processes were monitored using Fourier-transform infrared spectroscopy, differential scanning calorimetry, a haze meter and a rheometer. Curing rate constant and activation energy were calculated using a theoretical model and numerical method, respectively. Results revealed that owing to its co-continuous micro-phase separation structure, the F23-CLF-30-D3 semi-IPN exhibited considerably higher tensile strength and elongation at break than pure fluororubber F2314 and the F23-CLF-30-D0 semi-IPN because the phase separation and curing rates matched in the initial stage of curing.An arc Brazilian test revealed that F23-CLF-30-D-based composites used as mock materials for PBXs exhibited excellent mechanical performance and storage stability. Thus, the matched curing and phase separation rates play a crucial role during the fabrication of high-performance semi-IPNs;these factors can be feasibly controlled using an appropriate catalyst amount.

Rock Mass Quality Rating Based on the Multi-Criteria Grey Metric Space

Assessment of rock mass quality significantly impacts the design and construction of underground and open-pit mines from the point of stability and economy.This study develops the novel Gromov-Hausdorff distance for rock quality(GHDQR)methodology for rock mass quality rating based on multi-criteria grey metric space.It usually presents the quality of surrounding rock by classes(metric spaces)with specified properties and adequate interval-grey numbers.Measuring the distance between surrounding rock sample characteristics and existing classes represents the core of this study.The Gromov-Hausdorff distance is an especially useful discriminant function,i.e.,a classifier to calculate these distances,and assess the quality of the surrounding rock.The efficiency of the developed methodology is analyzed using the Mean Absolute Percentage Error(MAPE)technique.Seven existing methods,such as the Gaussian cloud method,Discriminant method,Mutation series method,Artificial neural network(ANN),Support vector machine(SVM),Grey wolf optimizer and Support vector classification method(GWO-SVC)and Rock mass rating method(RMR)are used for comparison with the proposed GHDQR method.The share of the highly accurate category of 85.71%clearly indicates compliance with actual values obtained by the compared methods.The results of comparisons showed that the model enables objective,efficient,and reliable assessment of rock mass quality.

Comparative Analysis of Building Rating Systems and Occupant Rating Systems

To begin with, rating systems are a beneficial tool in determining the efficiency of a building’s ability to utilise its resources effectively. In this study, the two elements under comparison are the Building Rating Systems (BRSs) and Occupant Rating Systems (ORSs). The main objective of this paper is to be able to examine the most commonly applied international and national BRS and ORS and, based on that, discover the possibility of developing an integration of both the BRS and ORS into one rating system. Quite simply, a BRS is a method by which buildings are assessed and given a score based on numerous features such as the efficiency of each of the services, total energy consumption, and alternate options of consumption. There are various BRSs that are implemented globally, each with its own set of criteria and specifications. Thus, based on the analysis of the benefits and drawbacks of both types of rating systems, it could be deduced that a well-rounded rating system with all technical and non-technical aspects combined would be beneficial to both the efficiency of the building as well as the building occupants’ health and well-being.

Exploring the thermal stability of lithium-ion cells via accelerating rate calorimetry:A review

Given the importance of lithium-ion cell safety,a comprehensive review on the thermal stability of lithium-ion cells investigated by accelerating rate calorimetry(ARC),is provided in the present work.The operating mechanism of ARC is discussed first,including the usage and the reaction kinetics.Besides that,the thermal stability of the cathode/anode materials at elevated temperatures is revealed by examining the impacts of some significant factors,i.e.,the lithium content,particle size,material density,lithium salt,solvent,additive,binder and initial heating temperature.A comparison of the common cathode materials indicates that the presence of Mn and polyanion could significantly enhance the thermal stability of cathode materials,while the doping of Al also helps to restrain the reactivity.Except for their high capacity,some alloy materials demonstrate more competitive safety than traditional carbon anode materials.Furthermore,the thermal behaviors of full cells under abusive conditions are reviewed here.Due to the sensitivity of ARC to the kinetic parameters,a reaction kinetic modeling can be built on the basis of ARC profiles,to predict the thermal behaviors of cell components and cells.Herein,a shortcircuit modeling is exampled.

SCIM: Incorporating secure communication and interference management in one operation

Due to the broadcast nature of wireless communications,users’data transmitted wirelessly is susceptible to security/privacy threats.Meanwhile,as a result of the limitation of spectrum resources,massive wireless connections will incur serious interference,which may damage the efficiency of data transmission.Therefore,improving both efficiency and secrecy of data transmission is of research significance.In this paper,we propose a wireless transmission scheme by taking both Secure Communication(SC)and Interference Management(IM)into account,namely SCIM.With this scheme,an SCIM signal is generated by the legitimate transmitter(Tx)and sent along with the desired signal,so that the SCIM signal can interact with and suppress the environmental interference at the legitimate receiver(Rx).Meanwhile,the SCIM signal may interfere with the eavesdropper in the coverage of legitimate transmission so as to deteriorate the eavesdropping performance.Therefore,the secrecy of desired transmission is improved.In this way,both the transmission efficiency and privacy are enhanced.Then,by taking various transmission preferences into account,we develop different implementations of SCIM,including Interference Suppression First SCIM(ISF-SCIM),Data Transmission First SCIM(DTF-SCIM),Anti-Eavesdropping First SCIM(AEF-SCIM),and Secrecy Rate Maximization SCIM(SRM-SCIM).Our in-depth simulation results have shown the proposed methods to effectively improve the efficiency and secrecy of the legitimate transmission.

The laparoscopic rating scale for the evaluation of working conditions for surgical treatment of super-obesity

In this technical note,a novel rating scale(abdominal integral index)was introduced for assessing the conditions of the working laparoscopic space based on linear measurements to select the optimal one or two-stage surgical treatment for super-obesity.Patients with the same height and similar BMI values had different rating scale scores,reflecting different conditions of laparoscopic bariatric surgery.The rating scale helps surgeons and patients make a safe option for surgery,depending on the experience of the surgeon and technical laparoscopic conditions.

Design of Online Vitals Monitor by Integrating Big Data and IoT

In this work,we design a multisensory IoT-based online vitals monitor(hereinafter referred to as the VITALS)to sense four bedside physiological parameters including pulse(heart)rate,body temperature,blood pressure,and periph-eral oxygen saturation.Then,the proposed system constantly transfers these signals to the analytics system which aids in enhancing diagnostics at an earlier stage as well as monitoring after recovery.The core hardware of the VITALS includes commercial off-the-shelf sensing devices/medical equipment,a powerful microcontroller,a reliable wireless communication module,and a big data analytics system.It extracts human vital signs in a pre-programmed interval of 30 min and sends them to big data analytics system through the WiFi module for further analysis.We use Apache Kafka(to gather live data streams from connected sen-sors),Apache Spark(to categorize the patient vitals and notify the medical pro-fessionals while identifying abnormalities in physiological parameters),Hadoop Distributed File System(HDFS)(to archive data streams for further analysis and long-term storage),Spark SQL,Hive and Matplotlib(to support caregivers to access/visualize appropriate information from collected data streams and to explore/understand the health status of the individuals).In addition,we develop a mobile application to send statistical graphs to doctors and patients to enable them to monitor health conditions remotely.Our proposed system is implemented on three patients for 7 days to check the effectiveness of sensing,data processing,and data transmission mechanisms.To validate the system accuracy,we compare the data values collected from established sensors with the measured readouts using a commercial healthcare monitor,the Welch Allyn®Spot Check.Our pro-posed system provides improved care solutions,especially for those whose access to care services is limited.

The Impact of Credit Ratings on Financial Performance (ROA) and Value Creation (Tobin’s Q)

This study employs a bibliometric and systematic approach to examine the impact of credit ratings as a measure of financial performance for companies listed in the S&P 500 index.The study identified a knowledge gap as only two researches were found,one suggesting and another using credit ratings to measure financial performance.Most researches use leverage,profitability,liquidity,and Share Return measures to explain financial performance.The empirical analysis uses the data of 2,398 observations of 240 companies rated by S&P Global Ratings for the period 2009-2013,applying a Generalized Method of Moments(GMM)methodology to estimate the models due to its ability to address potential endogeneity issues.The study considers Return on Assets(ROA)and Tobin’s Q as dependent variables.It incorporates credit ratings(CRWLTA)along with variables such as Total Debt to Total Assets(TDTA),Total Shareholder Return(TSR),EBITDA Interest coverage(EBITDAICOV),Quick Ratio(QR),Altman’s Z-Score(AZS),as well as macroeconomic factors like Gross Domestic Product(GDP)growth,inflation(Consumer Price Index-CPI),and the Federal Reserve Interest Rate(FDRI)as independent variables.The study argues that credit ratings,which incorporate historical data and confidential information about companies’strategies,provide reliable forward-looking creditworthiness assessments to the market.It is supported by specialized rating agencies that employ their methodologies.However,the findings suggested that CRWLTA,had a negative relationship with Q Tobin,although it was not statistically significant,and a negative relationship with ROA that was on the verge of significance.

Prediction of film ratings based on domain adaptive transfer learning

This paper examines the prediction of film ratings.Firstly,in the data feature engineering,feature construction is performed based on the original features of the film dataset.Secondly,the clustering algorithm is utilized to remove singular film samples,and feature selections are carried out.When solving the problem that film samples of the target domain are unlabelled,it is impossible to train a model and address the inconsistency in the feature dimension for film samples from the source domain.Therefore,the domain adaptive transfer learning model combined with dimensionality reduction algorithms is adopted in this paper.At the same time,in order to reduce the prediction error of models,the stacking ensemble learning model for regression is also used.Finally,through comparative experiments,the effectiveness of the proposed method is verified,which proves to be better predicting film ratings in the target domain.

Aspect Level Songs Rating Based Upon Reviews in English

With the advancements in internet facilities,people are more inclined towards the use of online services.The service providers shelve their items for e-users.These users post their feedbacks,reviews,ratings,etc.after the use of the item.The enormous increase in these reviews has raised the need for an automated system to analyze these reviews to rate these items.Sentiment Analysis(SA)is a technique that performs such decision analysis.This research targets the ranking and rating through sentiment analysis of these reviews,on different aspects.As a case study,Songs are opted to design and test the decision model.Different aspects of songs namely music,lyrics,song,voice and video are picked.For the reason,reviews of 20 songs are scraped from YouTube,pre-processed and formed a dataset.Different machine learning algorithms—Naïve Bayes(NB),Gradient Boost Tree,Logistic Regression LR,K-Nearest Neighbors(KNN)and Artificial Neural Network(ANN)are applied.ANN performed the best with 74.99%accuracy.Results are validated using K-Fold.

Study of the Effect of Operating Temperatures (1320 K, 1420 K and 1520 K) on the Vacant Sites of TiN Alloy in B2 Structure at 45%, 50% and 55% N by MEAM Method

In this work, we have studied the vacancy formation energy of TiN alloy of structure B2 of size 10 × 10 × 10 for nitrogen percentages of 45%, 50% and 55% under the influence of temperature at 1320 K, 1420 K and 1520 K using the Modified Embedded Atom Method MEAM under the calculation code LAMMPS version 2020. This study has enabled us to understand the behavior of the TiN alloy under different nitrogen percentages in terms of total energy, vacancy formation energy, crystalline parameter, occupancy rate and order parameter. For total energy, we have shown that as temperature increases, total energy decreases, making it easier to obtain TiN at higher temperatures;reaching the value of -7344.9169 eV for the 55% nitrogen structure for the temperature of 1420 K. For the energy of formation, we have shown that the compounds obtained at 1320 K and 1520 K have a more considerable energy of formation than that obtained at 1420 K. The study of fractions and the order parameter showed us that the structure of TiN with 55% nitrogen is less likely, as the composition obtained is at most 53.35%.

巴旦木分级破壳机设计与试验

为解决巴旦木破壳机械生产效率低、破壳率低和核仁损伤量大等问题,通过研究坚果破壳原理,选用适宜机械加工的大巴旦木(‘莎车3号’)为研究对象,设计了巴旦木分级破壳机,同时进行巴旦木尺寸、破壳力加载方向及加载速度、巴旦木含水率的物料特性分析。结果表明:破壳机破壳辊间隙和含水率对巴旦木破壳力和变形量影响显著。使用试验法优化了巴旦木分级破壳机构工作参数,当两轧辊间距离调整为Ⅰ级14 mm、Ⅱ级13 mm和Ⅲ级12 mm,含水率控制在9.8%~10.5%范围内,转速为1 000 r·min^(-1)时,测得平均破壳率为95.57%,平均破损率为3.37%,满足实际生产需求。

2015~2019年中国居民脑卒中发病与死亡特征分析

目的:分析2015~2019年中国居民脑卒中发病、死亡特征及其变化情况。方法:利用2014年在全国31个省级行政区100个县(市、区)建立的中国心脑血管事件登记系统中2015~2019年脑卒中发病和死亡数据,以监测地区常住人口为分母,计算年度脑卒中粗发病率;以第七次全国人口普查数据为标准人口,采用直接标化法计算年龄标化发病率;运用Joinpoint模型分析脑卒中发病率的年度变化百分比(APC)及变化趋势;以年度脑卒中病例的死亡数量(例)与报告发病数量(例)的比值计算死亡发病比。结果:2019年脑卒中粗发病率为468.48/10万,比2015年升高9.41%(APC=2.12%,95%CI:1.43%~2.82%,P_(trend)<0.01),且男性脑卒中粗发病率高于女性,农村高于城市。男性和女性的粗发病率分别升高11.26%(APC=2.53%,95%CI:1.83%~3.24%,P_(trend)<0.01)和7.26%(APC=1.63%,95%CI:0.81%~2.46%,P_(trend)<0.01)。2019年脑卒中年龄标化发病率为523.82/10万,比2015年下降7.47%(APC=-1.72%,95%CI:-3.23%~-0.20%,P_(trend)<0.05),女性脑卒中年龄标化发病率降低9.56%(APC=-2.27%,95%CI:-3.99%~-0.52%,P_(trend)<0.05),城市脑卒中年龄标化发病率降低13.19%(APC=-3.49%,95%CI:-5.34%~-1.61%,P_(trend)<0.05),城市女性脑卒中年龄标化发病率下降幅度(15.82%,APC=-4.27%,95%CI:-6.34%~-2.15%,P_(trend)<0.01)大于城市男性(11.40%,APC=-2.95%,95%CI:-4.56%~-1.31%,P_(trend)<0.05)。随着年龄的增长脑卒中粗发病率升高。相比2015年,2019年45~49岁年龄段的脑卒中粗发病率升高了12.48%(APC=3.18%,95%CI:1.67%~4.72%,P_(trend)<0.01),而80~84岁年龄段则降低15.76%(APC=-4.39%,95%CI:-7.63%~-1.04%,P_(trend)<0.05)。2015~2019年总体脑卒中死亡发病比为0.19。总体死亡发病比的年龄分布呈现“U”型,50~54岁及55~59岁人群的死亡发病比最低(0.10),之后随年龄增长逐渐升高,≥85岁最高(0.45),且各年龄段的城市死亡发病比均低于农村。结论:2015~2019年间监测地区脑卒中发病形势严峻,存在城乡、年龄、性别的差异化分布。未来有必要加强脑卒中防控,并在年龄、地区广泛覆盖等方面进一步优化。

云南西双版纳澳洲坚果开花与结实特性

【目的】对云南西双版纳地区的澳洲坚果开花结实特性展开观测及分析,为澳洲坚果高产优质提供理论依据。【方法】以12个澳洲坚果品种为观测对象,采用挂牌记录法观察其开花物候及坐果规律,以坐果率、坐果花序率、每果序平均坐果数、落果率等参数为评价指标,分析澳洲坚果不同品种的结实特性。【结果】不同澳洲坚果品种的开花物候复杂多样,品种D4、O.C、863、820和842花期较长,分2批次开花;品种A16、A4、Special、828、816、508和246的花期较短且较集中。受花期和品种特性共同影响,澳洲坚果不同品种的坐果率、坐果花序率、每果序平均坐果数及落果率存在明显差异,品种O.C、828、863和D4的第2批次末花期花序几乎没有坐果能力。所有参试品种的最终坐果率与最终坐果花序率呈正相关,可使用坐果花序率代替坐果率评价产量。花期对不同品种每果序最终平均坐果数量的影响不大,每个果序上仅1~2个果实。所有参试品种均存在严重的落果现象,但同一花期不同品种的落果率差异不显著(P>0.05)。【结论】根据开花物候决定采收模式,花期长的品种D4、O.C、863、820和842需分批次采收,花期短的品种A16、A4、Special、828、816、508和246可集中采收。根据第2批次末花期花序几乎无坐果能力的结实特性,可疏除2批次开花品种O.C、863、828和D4的第2批次末花期刚萌发的花序,节约母树资源。

不同治疗方案对卵巢储备功能正常年轻女性夫精人工授精妊娠结局的影响

目的探讨不同治疗方案对年龄<35岁卵巢储备功能正常女性夫精人工授精(AIH)妊娠结局的影响。方法回顾性分析2015年1月至2021年8月在本中心行AIH助孕的年龄<35岁卵巢储备功能正常女性的3024个周期的临床资料。根据治疗方案分为6组:自然周期组(NC组,n=380)、克罗米芬组(CC组,n=57)、CC联合促性腺激素(Gn)组(CC+Gn组,n=78)、Gn组(n=1712)、来曲唑组(LE组,n=430)以及LE联合Gn(LE+Gn组,n=367),比较各组患者的一般情况和妊娠结局,并采用多因素Logistic回归分析影响AIH临床妊娠率及活产率的相关因素。结果一般情况比较:6组间体质量指数(BMI)、窦卵泡数(AFC)、基础卵泡刺激素(FSH)水平、优势卵泡数、不孕因素整体比较有统计学差异(P<0.05),而女方年龄、不孕年限、不孕类型及精液处理方式比较均无统计学差异(P>0.05)。妊娠结局比较:6组患者间临床妊娠率和活产率整体比较有统计学差异(P<0.05),其中LE+Gn组的临床妊娠率和活产率显著高于NC组(分别为20.7%vs.9.5%;16.3%vs.7.4%)(P<0.05);6组间流产率、异位妊娠率及多胎率比较无统计学差异(P>0.05)。多因素Logistic回归分析结果显示,矫正混杂因素后,LE+Gn治疗方案和优势卵泡数均是临床妊娠和活产的保护因素(OR>1,P<0.05),而不孕年限是临床妊娠和活产的危险因素(OR<1,P<0.05)。结论LE联合Gn方案可显著提高AIH的临床妊娠率和活产率,可作为年龄<35岁卵巢储备功能正常女性AIH助孕的首选治疗方案。

分离禽蛋壳膜粉碎装置刀片的设计与试验

为高效分离回收蛋壳膜,设计了一款采用剪切和碾压的粉碎方式专用于分离禽蛋壳膜的粉碎装置。采用干燥的蛋壳膜进行试验,通过理论分析和预试验确定刀片类型、刀片间隙、转子转速、筛网目数为主要影响因素,根据实际生产情况确定各因素的试验水平,以分离率和损耗率为评价指标进行四因素三水平响应面试验;利用Design-Expert 12.0对试验结果进行响应面分析,研究不同影响因素对评价指标的影响规律。验证试验结果表明:当调整转子转速为1374r/min、安装2目筛网、刀片间隙为1mm的50°剪切式刀片时,分离率为86.72%,损耗率为3.71%,大粒径蛋膜较多。试验装置有效提高了粉碎蛋壳膜后蛋膜的分离效果,降低了损耗率,具有极大的优越性。

我国不同付费模式信用评级校验机制研究

我国债券市场刚性兑付被打破后,不断出现高评级债券违约事件。为此,监管部门积极发展多种评级,以期提高债券评级质量。通过理论建模与实证分析,研究了不同付费模式信用评级如何综合发挥校验作用。基于博弈模型的理论分析发现,不同付费模式评级发挥校验作用的机制在于投资者的反应:当投资者付费和发行人付费两种模式评级差值增大,投资者不以单一评级结果进行债券定价,而是根据评级差值调整债券价格。基于2011—2020年企业债评级数据的经验证据显示,随着评级差值增大,发行人付费评级机构会降低后续评级,且评级差值对单一信用评级与债券信用利差之间的负向关系具有抑制作用。进一步检验发现,发行人付费评级机构提供的偏正面的私有信息会降低债券信用利差,而以中债资信为代表的投资者付费评级机构提供的私有信息则不会显著影响债券信用利差,均体现出单一评级模式在定价方面的弊端。因此,监管部门应继续鼓励发展投资者付费评级模式,并鼓励机构投资者发展内部评级,以充分发挥多种评级模式的交叉校验作用。

施氮量对优质稻G优325氮肥利用率和产量的影响

【目的】探明优质杂交籼稻新品种G优325的适宜施氮量。【方法】设5个施氮量处理(0、60、120、180和240 kg/hm^(2),分别记为N0、N60、N120、N180和N240)进行田间试验,研究不同施氮量对G优325主要生育时期茎糵数、干物质积累与转运能力、氮肥吸收利用率和产量及其构成因素的影响。【结果】适当增加施氮量能显著提高G优325的产量,且在N180处理下产量最高。施氮量对产量构成因素、茎蘖数和干物质积累总量影响显著,随着施氮量增加,G优325成熟期的有效穗数、每穗总粒数和干物质积累总量均表现为先增加后降低的趋势,且有效穗数和干物质积累总量均在N180处理达到最高值,而每穗粒数在N120处理最多。随着施氮量增加,抽穗前后的干物质积累向籽粒的转运量均表现为处理N180>N240>N120>N60>N0;拔节期、抽穗期和成熟期的氮素积累总量均表现为先增加后降低的趋势,且均在N180处理下最高;氮素干物质生产效率、氮素稻谷生产效率、氮素收获指数、氮肥吸收利用率、氮肥农学利用率、氮肥偏生产力和氮素生理利用率均表现为处理N0>N60>N120>N180>N240。【结论】适宜提高施氮量(180 kg/hm^(2)),可协同实现优质杂交籼稻新品种G优325高产与较高氮肥利用效率。

不同施氮量对春小麦灌浆速率和产量的影响

【目的】研究不同施氮量对小麦籽粒灌浆进程的调控效应,分析滴灌条件下春小麦产量的提升潜力,为确定不同品质类型小麦品种适宜施氮量提供理论依据。【方法】以11个新疆审定的春小麦常规品种和1个现阶段正在培育的品系为材料,设置4个施氮水平处理,研究施氮量对不同春小麦品种(系)籽粒灌浆特性和产量的影响。【结果】不同施氮水平下小麦籽粒灌浆趋势均呈现为“S”型,随着施氮量的增加,小麦籽粒的干物质积累量达到最高籽粒千粒重后,积累的有效籽粒干物质重不增反降,但下降程度有所减缓。新春6号、新春37号、核春137表现为灌浆强势品种,新春11号、新春21号、新春29号表现为灌浆弱势品种。100 kg/hm^(2)低氮水平下,新春17号和核春615产量达最高水平;200 kg/hm^(2)中氮水平下,新春6号、新春11号、新春26号、新春29号、新春37号、新春39号、核春121和核春514共8个春小麦品种产量达最高水平;300 kg/hm^(2)高氮水平下,核春137和新春21号产量达最高水平。【结论】4个施氮水平下,最大灌浆速率V max与平均灌浆速率Vs均呈极显著正相关,平均灌浆速率Vs与灌浆总天数T均呈负相关。公顷产量Y与千粒重A、最大灌浆速率V max、平均灌浆速率Vs、达到最大灌浆速率所用天数T max、灌浆总天数T均呈正相关。

设施生菜光合和蒸腾速率影响因素分析与预测模型构建

光合速率及蒸腾速率是植物的2个重要生理指标。在全人工环境下,选取意大利生菜作为对象,设计并开展多环境变量对生菜光合速率及蒸腾速率影响的嵌套实验,得到环境因子对生菜光合速率及蒸腾速率的影响规律,应用神经网络构建生菜幼苗期光合速率及蒸腾速率预测模型。针对幼苗期生菜,选择温度、相对湿度、光子通量密度(Photosynthetic photon flux density, PPFD)及CO_(2)浓度共4个环境影响因素,采用随机森林方法对数据进行相关性分析。结果表明,与蒸腾速率相关性由大到小的因素依次为CO_(2)浓度、温度、相对湿度、PPFD,与光合速率相关性由大到小的因素依次为CO_(2)浓度、PPFD、温度、相对湿度;采用枚举法确定隐藏层节点数和训练函数,通过遗传算法优化BP神经网络的初始权值和阈值,构建GA-BP神经网络生理指标预测模型。应用测试数据对模型进行验证,光合速率及蒸腾速率预测值与实测值的决定系数分别为0.962 12、0.979 44,均方根误差(RMSE)分别为2.983 2μmol/(m^(2)·s)、0.001 435 8 mol/(m^(2)·s),表明GA-BP神经网络在模型精度和迭代次数方面性能显著提高。研究结果可为设施生菜生产环境调控提供有效依据。