基于知识与AW-ESN融合的烧结过程FeO含量预测

氧化亚铁(FeO)含量是衡量烧结矿强度和还原性的重要指标,烧结过程FeO含量的实时准确预测对于提升烧结质量、优化烧结工艺具有重要意义.然而烧结过程热状态参数缺失、过程参数波动频繁给FeO含量的高精度预测带来巨大的挑战,为此,提出一种基于知识与变权重回声状态网络融合(Fusion of data-knowledge and adaptive weight echo state network, DK-AWESN)的烧结过程FeO含量预测方法.首先,针对烧结过程热状态参数缺失的问题,建立烧结料层最高温度分布模型,实现基于料层温度分布特征的FeO含量等级划分;其次,针对烧结过程参数波动频繁的问题,提出基于核函数高维映射的多尺度数据配准方法,有效抑制离群点的影响,提升建模数据的质量;最后,针对烧结过程数据驱动模型缺乏机理认知致使模型预测精度不高的问题,将过程数据中提取得到的FeO含量等级知识与AW-ESN (Adaptive weight echo state network)结合,建立DK-AWESN模型,有效提升复杂工况下FeO含量的预测精度.现场工业数据试验表明,所提方法能实时准确地预测烧结过程FeO含量,为烧结过程的智能化调控提供实时有效的FeO含量反馈信息.

基于机器视觉的烧结矿FeO含量在线感知

烧结矿FeO的含量是烧结生产的一项综合性指标,影响它的因素较多而且各因素间呈现一种非线性关系,导致对FeO含量的预测难度较大。针对烧结矿FeO含量难以预测的问题,本文提出一种基于机器视觉技术实现烧结矿FeO含量在线感知的系统。该系统通过在烧结台车机尾安装红外热成像设备来获取烧结断面的热成像图片信息并对图片信息特征进行提取和分析,提取的机尾断面图像特征作为Darknet-19算法的输入参数,建立基于改进的Darknet-19算法的烧结矿FeO含量预测模型,实现对烧结矿FeO含量的实时预测。现场实际使用表明,烧结矿FeO含量预测模型的预测值与实际值偏差±0.5时,准确率在82.5%,对稳定和优化烧结生产过程控制有积极作用。

高钙造纸废水中添加FeO提高厌氧颗粒污泥产甲烷能力的研究

本研究向高钙造纸废水中添加FeO,通过分析厌氧反应器中的甲烷累积产量、甲烷最大生成速率和厌氧颗粒污泥中的胞外聚合物组分(蛋白质、多糖和腐殖质)含量变化以及微生物群落变化,探讨了FeO对造纸废水厌氧消化的影响。结果表明,添加FeO对甲烷的产生有促进作用。在模拟高钙造纸废水中,与对照组相比,当FeO的添加量为1 g/L时,甲烷累积产量增加了59.4%;当FeO的添加量为15 g/L时,甲烷累积产量增加了82.3%。分别在实际造纸废水A和B中添加FeO,最多可使甲烷累积产量提高52.0%和85.4%。FeO可导致胞外聚合物组分含量提高,促进微生物的生长代谢。添加FeO改变了微生物群落组成,微生物群落多样性有很明显的提高,增强了厌氧消化反应的稳定性。因此,添加FeO有助于提高厌氧消化产甲烷能力,可以更加高效地处理高钙造纸废水。

基于TCN-DenseNet的烧结矿FeO含量预测

烧结矿FeO含量是烧结工序的一项重要质量和能耗指标,也对高炉冶炼有直接影响。针对目前化学检测法检测烧结矿FeO含量时存在较长时间滞后的现状,本文提出了一种时域卷积网络(Temporal Convolutional Network,TCN)与密集连接卷积神经网络(Densely Connected Convolutional Network,DenseNet)混合的烧结矿FeO含量预测方法。首先采用TCN建立烧结矿FeO含量的时间序列预测模型,同时采集烧结机尾断面红外图像,采用DenseNet建立烧结矿FeO预测模型,通过自适应加权平均方法将两者的输出结果进行整合,获得最终的烧结矿FeO含量预测值。针对烧结矿层断面红外图像的特征,对DenseNet进行了添加注意力层、修改卷积块结构,并修改了浅层卷积层大小和步长等改进措施。在国内某钢铁公司的大型烧结机的实际生产数据上对模型进行了验证,经过数据处理、模型参数优化等操作后,本文所提的TCN-DenseNet混合模型的烧结矿FeO含量预测在测试集绝对误差±0.4%以内命中率可达94.34%,均方根误差为0.21,优于单独使用TCN或者DenseNet进行建模时的预测效果。该方法对提高烧结矿FeO含量预测的准确性和稳定性效果显著,可以为烧结现场的生产操作提供数据支撑。

基于分子动力学研究纳米孔洞对多晶FeO/Fe中FeO断裂的影响

孔洞缺陷是影响表面氧化皮断裂失效的重要因素.本文结合分子动力学软件Lammps和Atomsk软件建立了半径相同位置不同(模型1)和孔洞位置相同半径不同(模型2)两种模型,为了探究两者对FeO/Fe多晶模型拉伸失效的影响,文章从抗拉强度、断裂韧性、中心对称参数(CSP)三方面进行了分析,模拟结果表明:1.当孔洞位于晶界和三叉点时,模型1和2的抗拉强度随着孔洞半径的增大而减小,断裂韧性在抗拉强度降低的同时增强.2.孔洞位于晶内时,模型2的抗拉强度呈现“先下降-后上升-再下降”的变化趋势,断裂韧性整体降低;模型1既增强了模型的抗拉强度,又降低了断裂韧性.3.裂纹扩展受孔洞半径影响较大,半径达到1 nm以上时,模型2的裂纹扩展速度加快,模型1的裂纹扩展速度减慢.本研究为进一步认识氧化皮的断裂过程提供了帮助.

FeO@C/MAX复合材料的制备及电化学动力学性能

将α-Fe_(2)O_(3)@C与钛粉和铝粉一同进行高温煅烧,制备了Fe O@C/MAX(FCM)复合材料。通过XRD、SEM、TEM表征了FCM复合材料在不同Ti/C与Al/C物质的量比下的结构、组成及形貌变化,采用电化学动力学分析方法定量计算了FCM复合材料的赝电容占比,推测可能的电荷储存机理。结果表明,随着Ti/C与Al/C物质的量比的增大,FCM复合材料中MAX相(Ti_(2)Al C和Ti_(3)Al C_(2))的含量随之变化,而α-Fe_(2)O_(3)转变为不稳定的Fe O。当n(Ti)∶n(Al)∶n(C)=3∶1∶2时,制得的FCM-312样品在1 m V/s扫描速率下的比电容最大,为125.09 F/g,约为α-Fe_(2)O_(3)@C的4.76倍。FCM复合材料中部分MAX相在电化学过程中发生氧化还原反应,为离子间电子快速输运提供了条件,增加了FCM复合材料的赝电容占比。其中,FCM-312样品在10 m V/s扫描速率下的赝电容占比为22.12%。

ε-Fe_(2)O_(3)/FeO界面结构与相变机理研究

本文利用脉冲激光沉积技术在SrTiO_(3)(111)衬底上生长了ε-Fe_(2)O_(3)薄膜,并应用高分辨X射线衍射(HRXRD)、X射线光电子能谱(XPS)和透射电子显微镜(TEM)对薄膜的显微结构进行了系统表征。XRD和XPS的研究结果表明ε-Fe_(2)O_(3)(001)薄膜在SrTiO_(3)(111)衬底上外延生长,薄膜中Fe离子为+3价。TEM的结果表明,在TEM样品制备过程中,由于高能Ar离子束轰击,ε-Fe_(2)O_(3)/SrTiO_(3)界面上容易发生相变形成FeO。ε-Fe_(2)O_(3)/FeO/SrTiO_(3)的外延关系为ε-Fe_(2)O_(3)(001)[110]//FeO(111)[112]//SrTiO_(3)(111)[112]。基于ε-Fe_(2)O_(3)/FeO界面结构与取向关系,分析了离子束辐照诱导ε-Fe_(2)O_(3)→FeO相变的微观机制。

沙钢5800 m^(3)高炉所用烧结矿FeO含量研究

这是一篇冶金工程领域的文章。为给烧结矿FeO成分控制、质量优化提供理论依据,考查了沙钢5800 m^(3)高炉所用烧结矿转鼓强度、低温还原粉化、中温还原性等冶金性能,并通过统计数据分析探究了FeO含量对烧结矿冶金性能的影响以及影响FeO含量的因素,结果表明:基于沙钢现行原料质量、配矿结构以及工艺操作条件下:FeO含量增加可提升烧结矿转鼓指数、RDI+3.15及RI,当RI大于85%时,FeO含量才成为制约RI因素,FeO含量适宜值为9.5%左右;R、SiO_(2)含量、精粉及燃料配比增加会导致FeO含量升高,提升MgO含量,负压会造成FeO含量下降;数据分析表明:控制R为1.9,SiO_(2)、MgO含量分别为5.2%、1.8%左右,精粉配比在11%以内,燃料配比在4.5%~5%之间,负压为16 KPa以内,可使烧结矿FeO含量均值维持在9.5%左右,对应转鼓指数均值为80%左右,RDI+3.15、RI均值分别为75%、80%以上。

Mechanical properties and acoustic emission characteristics of soft rock with different water contents under dynamic disturbance

Uniaxial compression tests and cyclic loading acoustic emission tests were conducted on 20%,40%,60%,80%,dry and saturated muddy sandstone by using a creep impact loading system to investigate the mechanical properties and acoustic emission characteristics of soft rocks with different water contents under dynamic disturbance.The mechanical properties and acoustic emission characteristics of muddy sandstones at different water contents were analysed.Results of experimental studies show that water is a key factor in the mechanical properties of rocks,softening them,increasing their porosity,reducing their brittleness and increasing their plasticity.Under uniaxial compression,the macroscopic damage characteristics of the muddy sandstone change from mono-bevel shear damage and‘X’type conjugate bevel shear damage to a roadway bottom-drum type damage as the water content increases.Dynamic perturbation has a strengthening effect on the mechanical properties of samples with 60%and less water content,and a weakening effect on samples with 80%and more water content,but the weakening effect is not obvious.Macroscopic damage characteristics of dry samples remain unchanged,water samples from shear damage and tensile–shear composite damage gradually transformed into cleavage damage,until saturation transformation monoclinic shear damage.The evolution of acoustic emission energy and event number is mainly divided into four stages:loading stage(Ⅰ),dynamic loading stage(Ⅱ),yield failure stage(Ⅲ),and post-peak stage(Ⅳ),the acoustic emission characteristics of the stages were different for different water contents.The characteristic value of acoustic emission key point frequency gradually decreases,and the damage degree of the specimen increases,corresponding to low water content—high main frequency—low damage and high water content—low main frequency—high damage.

基于新型Contention网络的滚动轴承早期故障诊断方法研究

针对滚动轴承早期故障诊断问题,为了同时建模振动信号中的高频和低频特征,实现高精度诊断,提出了一种新的模型Contention。它以一种互补的方式集成了空洞卷积和自注意力机制的优点,具有同时捕捉高频和低频信息的能力。为了验证其诊断能力,首先,在完整信息原则下将振动信号连续构造成数据集;其次,搭建Contention网络并训练,其最终测试集准确度可达100%,t-SNE显示随网络层次的深入特征被明显聚类;最后,设置对照实验,将该模型与传统RNN、CNN、Transformer模型对比,证明该模型具备突出的诊断能力。

Incorporating empirical knowledge into data-driven variable selection for quantitative analysis of coal ash content by laser-induced breakdown spectroscopy

Laser-induced breakdown spectroscopy(LIBS)has become a widely used atomic spectroscopic technique for rapid coal analysis.However,the vast amount of spectral information in LIBS contains signal uncertainty,which can affect its quantification performance.In this work,we propose a hybrid variable selection method to improve the performance of LIBS quantification.Important variables are first identified using Pearson's correlation coefficient,mutual information,least absolute shrinkage and selection operator(LASSO)and random forest,and then filtered and combined with empirical variables related to fingerprint elements of coal ash content.Subsequently,these variables are fed into a partial least squares regression(PLSR).Additionally,in some models,certain variables unrelated to ash content are removed manually to study the impact of variable deselection on model performance.The proposed hybrid strategy was tested on three LIBS datasets for quantitative analysis of coal ash content and compared with the corresponding data-driven baseline method.It is significantly better than the variable selection only method based on empirical knowledge and in most cases outperforms the baseline method.The results showed that on all three datasets the hybrid strategy for variable selection combining empirical knowledge and data-driven algorithms achieved the lowest root mean square error of prediction(RMSEP)values of 1.605,3.478 and 1.647,respectively,which were significantly lower than those obtained from multiple linear regression using only 12 empirical variables,which are 1.959,3.718 and 2.181,respectively.The LASSO-PLSR model with empirical support and 20 selected variables exhibited a significantly improved performance after variable deselection,with RMSEP values dropping from 1.635,3.962 and 1.647 to 1.483,3.086 and 1.567,respectively.Such results demonstrate that using empirical knowledge as a support for datadriven variable selection can be a viable approach to improve the accuracy and reliability of LIBS quantification.

转炉终渣FeO含量的控制研究

研究了影响转炉终渣FeO含量的因素,主要从吹炼末期矿石加入量、停吹碳含量、停吹游离氧、炉渣碱度、渣中MgO含量以及钢包磷含量来探索渣中FeO含量的控制方法。研究结果表明,吹炼末期矿石加入量与渣中FeO含量增量呈正相关关系,停吹氧含量与渣中FeO活度成正比例关系,停吹碳含量与渣中FeO活度成反比例关系,渣中碱度与渣中FeO含量呈正相关关系,渣中MgO含量与渣中FeO呈负相关关系。为了控制渣中w(FeO)小于18%,后期将矿石加入量控制在1000 kg以内,停吹w(C)大于0.045%,游离w(O)小于0.060%,渣中w(MgO)控制在8.0%~12.0%,渣中碱度控制在2.8~4.0,对于钢中w(P)上限0.012%以下的钢种,碱度控制在4.0~5.0。

Porous FeP/FeO Deposited on Carbon Cloth with Accelerated Kinetics for the Hydrogen Evolution Reaction

FeP/FeO was prepared on carbon cloth(CC)via hydrothermal method,heat treatment in air,and phosphorization in argon.FeP/FeO/CC presents a porous and loose morphology which is conducive to the exposure of active sites and the transfer of reactants.FeP/FeO/CC requires the low overpotentials of 257 and 117 mV(vs.reversible hydrogen electrode(RHE))to achieve the current density of 10 mA_·cm^(-2)for oxygen evolution reaction(OER)and hydrogen evolution reaction(HER)in alkaline KOH solution,respectively.The small Tafel slope values of 36.1 mV_·dec^(-1)(for OER)and 96.2 mV_·dec^(-1)(for HER)indicate that FeP/FeO/CC exhibits the fast electrocatalytic reactive kinetics for OER and HER.In particular,the reaction kinetics of FeP/FeO/CC accelerated with the progress of HER.The charge-transfer resistance(R_(ct))of FeP/FeO/CC is only 11Ω.Excellent bifunctional electrocatalytic performances of FeP/FeO/CC should be attributed to the porous morphology and the lower charge-transfer resistance.

Digital twin driven and intelligence enabled content delivery in end-edge-cloud collaborative 5G networks

The rapid development of 5G/6G and AI enables an environment of Internet of Everything(IoE)which can support millions of connected mobile devices and applications to operate smoothly at high speed and low delay.However,these massive devices will lead to explosive traffic growth,which in turn cause great burden for the data transmission and content delivery.This challenge can be eased by sinking some critical content from cloud to edge.In this case,how to determine the critical content,where to sink and how to access the content correctly and efficiently become new challenges.This work focuses on establishing a highly efficient content delivery framework in the IoE environment.In particular,the IoE environment is re-constructed as an end-edge-cloud collaborative system,in which the concept of digital twin is applied to promote the collaboration.Based on the digital asset obtained by digital twin from end users,a content popularity prediction scheme is firstly proposed to decide the critical content by using the Temporal Pattern Attention(TPA)enabled Long Short-Term Memory(LSTM)model.Then,the prediction results are input for the proposed caching scheme to decide where to sink the critical content by using the Reinforce Learning(RL)technology.Finally,a collaborative routing scheme is proposed to determine the way to access the content with the objective of minimizing overhead.The experimental results indicate that the proposed schemes outperform the state-of-the-art benchmarks in terms of the caching hit rate,the average throughput,the successful content delivery rate and the average routing overhead.

Surface and Content Validity of the Mentoring Function Scale for Novice Nurses

This study aimed to examine the surface and content validity of the Mentoring Function Scale for Novice Nurses, used to assess the mentoring of entry-level nurses, and to refine the scale items. In Study 1, six nurse education researchers, selected using convenience sampling, with five or more years of nursing experience and experience teaching novice nurses, were invited to an expert meeting in July 2015. A group interview was conducted that lasted approximately 120 minutes. Study 2 examined the content validity index. Between September and November 2015, we distributed a self-administered questionnaire survey to 11 participants selected by convenience sampling. The participants included five nurse education researchers with a minimum of five years of nursing experience and experience teaching novice nurses, as well as six clinical nurses with a master’s degree or higher. Finally, 81 questionnaire items were retained from the initial 125 items. The 81-item Mentoring Function Scale for Novice Nurses had higher content validity than the original scale. To further increase the scale’s applicability, future studies should assess its reliability, construct validity, and criterion-related validity.

A Survey of Digital Content Required for STD Prevention Education Taught by School Nurses

Purpose: In recent years, there has been concern in Japan about the increase in STDs among adolescents aged 15 - 19. In order to prevent STDs and guide adolescents toward desirable sexual behavior, this study’s goal was to examine the actual state of STD prevention education taught by school nurses as part of the school curriculum, as well as the kinds of digital content they wish to have for teaching on the subject. Method: An anonymous, self-administered questionnaire was distributed by email and leaflet. Targeting the 100 valid responses received from the surveyed school nurses, descriptive statistics were made for each survey item and comparisons were made between the early- and mid-career groups based on years of experience. Results: 70.0% of respondents reported teaching about STDs in Health & Physical Education classes. School nurses in the early-career group used ready-made materials, while the mid-career group used both ready-made and self-made materials. 95% of respondents reported that they had little or no knowledge of STDs, while 84% reported that they were “good” or “fairly good” at teaching classes on sexuality. Both groups reported difficulty with the topics of “phimosis” and “nocturnal emissions” in the physiological category, “sexual behavior” in the “sex-adjacent” category, and “sexual and reproductive issues” in the psychosocial category. Respondents expressed a need for digital content covering the topics of “sexual violence”, “sexual abuse”, “sexually transmitted diseases”, and “how to turn down sex”. The mid-career group desired digital content for more items than the early-career group, with significant differences in the pathogens involved in sexually transmitted diseases and sex-adjacent items. 32.0% of respondents answered that they had done self-study for sex education classes in the past three years, and both groups desired self-study on “sexually transmitted diseases”, with the mid-career group significantly higher than the early-career group on “cervical cancer & HPV”, “emergency contraceptives”, and “media literacy”. Conclusion: Health & Physical Education is the main subject that covers STDs, and pre-existing materials are commonly used. It is expected that this is due to nurses’ busy schedules and the fact that only one school nurse is assigned to each school. The data collected suggested that nurse teachers want categorized digital content that can be used in the classes they are responsible for, rather than educational content on topics they are not comfortable with. In addition, given the changing environment surrounding sex, it was clear that they wish to understand the realities and needs of the high school students they teach, and desire self-study opportunities to improve their teaching methods.

Quantitative characterizations of anisotropic dynamic properties in organic-rich shale with different kerogen content

Understanding the quantitative responses of anisotropic dynamic properties in organic-rich shale with different kerogen content(KC)is of great significance in hydrocarbon exploration and development.Conducting controlled experiments with a single variable is challenging for natural shales due to their high variations in components,diagenesis conditions,or pore fluid.We employed the hot-pressing technique to construct 11 well-controlled artificial shale with varying KC.These artificial shale samples were successive machined into prismatic shape for ultrasonic measurements along different directions.Observations revealed bedding perpendicular P-wave velocities are more sensitive to the increasing KC than bedding paralleling velocities due to the preferential alignments of kerogen.All elastic stiffnesses except C_(13)are generally decreasing with the increasing KC,the variation of C_(1) and C_(33)on kerogen content are more sensitive than those of C_(44)and C_(66).Apparent dynamic mechanical parameters(v and E)were found to have linear correlation with the true ones from complete anisotropic equations independent of KC,which hold value towards the interpretation of well logs consistently across formations,Anisotropic mechanical parameters(ΔE and brittlenessΔB)tend to decrease with the reducing KC,withΔB showing great sensitivity to KC variations.In the range of low KC(<10%),the V_(P)/V_(S) ratio demonstrated a linearly negative correlation with KC,and the V_(P)/V_(S) ratio magnitude of less than 1.75may serve as a significant characterization for highly organic-rich(>10%)shale,compilation of data from natural organic rich-shales globally verified the similar systematic relationships that can be empirically used to predict the fraction of KC in shales.

Longitudinal dependence of the forecast accuracy of the ionospheric total electron content in the equatorial zone

The longitudinal dependence of the behavior of ionospheric parameters has been the subject of a number of works where significant variations are discovered.This also applies to the prediction of the ionospheric total electron content(TEC),which neural network methods have recently been widely used.However,the results are mainly presented for a limited set of meridians.This paper examines the longitudinal dependence of the TEC forecast accuracy in the equatorial zone.In this case,the methods are used that provided the best accuracy on three meridians:European(30°E),Southeastern(110°E)and American(75°W).Results for the stations considered are analyzed as a function of longitude using the Jet Propulsion Laboratory Global Ionosphere Map(JPL GIM)for 2015.These results are for 2 h ahead and 24 h ahead forecast.It was found that in this case,based on the metric values,three groups of architectures can be distinguished.The first group included long short-term memory(LSTM),gated recurrent unit(GRU),and temporal convolutional networks(TCN)models as a part of unidirectional deep learning models;the second group is based on the recurrent models from the first group,which were supplemented with a bidirectional algorithm,increasing the TEC forecasting accuracy by 2-3 times.The third group,which includes the bidirectional TCN architecture(BiTCN),provided the highest accuracy.For this architecture,according to data obtained for 9 equatorial stations,practical independence of the TEC prediction accuracy from longitude was observed under the following metrics(Mean Absolute Error MAE,Root Mean Square Error RMSE,Mean Absolute Percentage Error MAPE):MAE(2 h)is 0.2 TECU approximately;MAE(24 h)is 0.4 TECU approximately;RMSE(2 h)is less than 0.5 TECU except Niue station(RMSE(2 h)is 1 TECU approximately);RMSE(24 h)is in the range of 1.0-1.7 TECU;MAPE(2 h)<1%except Darwin station,MAPE(24 h)<2%.This result was confirmed by data from additional 5 stations that formed latitudinal chains in the equatorial part of the three meridians.The complete correspondence of the observational and predicted TEC values is illustrated using several stations for disturbed conditions on December 19-22,2015,which included the strongest magnetic storm in the second half of the year(min Dst=-155 nT).

Selection and application of four QTLs for grain protein content in modern wheat cultivars

The grain protein content(GPC)is the key parameter for wheat grain nutritional quality.This study conducted a resampling GWAS analysis using 406 wheat accessions across eight environments,and identified four previously reported GPC QTLs.An analysis of 87 landraces and 259 modern cultivars revealed the loss of superior GPC haplotypes,especially in Chinese cultivars.These haplotypes were preferentially adopted in different agroecological zones and had broad effects on wheat yield and agronomic traits.Most GPC QTLs did not significantly reduce yield,suggesting that high GPC can be achieved without a yield penalty.The results of this study provide a reference for future GPC breeding in wheat using the four identified QTLs.

A Study on the Correlation between Salivary Cortisol Content and Anxiety and Depression in Pregnant and Postpartum Women

Objective: To analyze the correlation between salivary cortisol content and anxiety and depression in pregnant and postpartum women. Method: 300 pregnant and postpartum women who were admitted from January 2021 to December 2021 were selected as the research subjects. They were all tested with the Hamilton Anxiety Scale (HAMA) and the Edinburgh Postnatal Depression Scale (EPDS). 40 pregnant women with anxiety and depression were selected as the observation group, and 50 normal pregnant women were selected as the control group, adopting logistic regression analysis to investigate the correlation between salivary cortisol and postpartum anxiety and depression during pregnancy. Result: The salivary cortisol levels in the observation group were significantly higher than those in the control group before bedtime, after waking up the next day, 15 minutes after waking up, 30 minutes after waking up, 60 minutes after waking up, and 1 hour before lunch (p < 0.05). Maternal depression and anxiety were not related to age, weight, gestational age, negative events, or history of adverse pregnancy and childbirth (p > 0.05). Maternal depression and anxiety are closely related to salivary cortisol levels and educational background (p Conclusion: There is a close correlation between anxiety and depression and an increase in cortisol levels, suggesting that the salivary cortisol awakening response during pregnancy has a predictive effect on the occurrence of postpartum depression.