Interpretable gradient boosting based ensemble learning and African vultures optimization algorithm optimization for estimating deflection induced by excavation

Intelligent construction has become an inevitable trend in the development of the construction industry.In the excavation project,using machine learning methods for early warning can improve construction efficiency and quality and reduce the chances of damage in the excavation process.An interpretable gradient boosting based ensemble learning framework enhanced by the African Vultures Optimization Algorithm(AVOA)was proposed and evaluated in estimating the diaphragm wall deflections induced by excavation.We investigated and compared the performance of machine learning models in predicting deflections induced by excavation based on a database generated by finite element simulations.First,we exploratively analyzed these data to discover the relationship between features.We used several state-of-the-art intelligent models based on gradient boosting and several simple models for model selection.The hyperparameters for all models in evaluation are optimized using AVOA,and then the optimized models are assembled into a unified framework for fairness assessment.The comprehensive evaluation results show that the AVOA-CatBoost built in this paper performs well(RMSE=1.84,MAE=1.18,R2=0.9993)and cross-validation(RMSE=2.65±1.54,MAE=1.17±0.23,R2=0.998±0.002).In the end,in order to improve the transparency and usefulness of the model,we constructed an interpretable model from both global and local perspectives.

Vultures as a model for testing molecular adaptations of dietary specialization in birds

Vultures are the only obligate scavengers among extant vertebrates.They provide valuable ecological services in ecosystems through removing carcasses,thus preventing the growth of other scavenger populations and the spread of pathogens.Moreover,their specific diets expose them to various deadly pathogens,which makes them potential candidates for studying molecular adaptations required to survive this extremely specialized scavenging habit.In this review,we summarize the morphological characteristics and behavioral habits,origin and phylogeny,and molecular adaptations to scavenging in both Old and New World vultures.The two groups of vultures share a similar appearance,indicative of convergent evolution.Vultures have experienced different degrees of specialization in their sensory organs;Old World vultures depend on sight,while New World ones depend on both smell and sight.Combined fossil records and molecular data suggest that vultures evolved independently,with distinct phylogenetic positions.We also explored their adaptation to scavenging in facial and intestinal microbiomes,gastric acid secretion and immunity.Compared with the facial microbiome,the intestinal microbiome had a lower diversity,dominated by Fusobacteria and Clostridia.The phages and single invertebrate species Adineta vaga,which feeds on dead bacteria and protozoa,present in the gut suggest a possible alternative defense mechanism.Several genes involved in gastric acidic secretion(including ATP4B,SLC26A7 and SST)and immunity(including BCL6,STING,and TLRs) undergoing positive selection likely have essential roles in eliminating invasive pathogens and initiating an innate immune response.Taken together,this review presents the current research status of vultures and highlights the use of vultures as a model for exploring molecular adaptations of dietary specialization in birds.It also provides a theoretical basis for the study of the genetic mechanisms of vultures to scavenging,and contributes to the formulation of vulture conservation strategies.

Opportunistic infection of Aspergillus and bacteria in captive Cape vultures (Gyps coprotheres)

Objective:To describe clinical signs,pathology,diagnosis and treatment of Cape vultures in which Aspergillus fumigatus(A.fumigatus)and mixed species of bacteria were isolated.Methods:Six Cape vultures sourced from South Africa for exhibition at Al Ain Zoo developed illness manifesting as anorexia,dyspnea,polyuria and lethargy.Three vultures died manifesting"pneumonia-like syndrome".These three vultures were necropsied and gross lesions recorded,while organ tissues were collected for histopathology.Internal organs were swabbed for bacteriology and mycology.From live vultures,blood was collected for hematology and biochemistry,oropharyngeal and cloacal swabs were collected for mycology and bacteriology.Results:A.fumigatus was isolated from the three dead vultures and two live ones that eventually survived.One of the dead vulture and two live vultures were co-infected with A.fumigatus and mixed species of bacteria that included Clostridium perfringens,Pseudomonas,Staphylococcus,Escherichia,Proteus,Enterococcus and Enterbacter.One of the Cape vulture and a Lappetfaced vulture,however,were free of Aspergillus or bacterial infections.At necropsy,intestinal hemorrhages were observed and the lungs were overtly congested with granulomas present on caudal air sac.Histopathologtcal examinations demonstrated granulomatous lesions that were infiltrated by mononuclear cells and giant cells.Conclusions:Aspergillosis is a persistent threat to captive birds and we recommend routine health assessments so that early diagnosis may prompt early treatment.It is likely that prompt prophylaxis by broad spectrum antibiotics and antifungals medication contributed to the survival of some of the vultures.

The Egyptian Vulture (Neophron percnopterus): record of a new bird in China

In 2 April 2012 a birdwatcher,Mr.Hong Guo,found a strange white bird near a garbage dump in Wuqia County,(39°55′N,75°15′E),at an elevation of 2100m.This location is approximately 80km west of Kaxgar.After experts checked the photographs,it was identified as an

First Characterization of <i>Escherichia coli</i>Strains Isolated from Wildlife Griffon Vulture (<i>Gyps fulvus</i>) in the Southeast of Spain

The aim of the present study was to characterize a collection of Escherichia coli strains isolated from asymptomatic griffon vulture (Gyps fulvus) during a reintroduction program in the southeast of Spain, in order to establish if griffon vulture could play a role in the spread of resistant or potentially pathogenic E. coli strains. For this purpose, 14 E. coli strains obtained from 10 griffon vulture were studied to establish their serotypes, phylogroups, virulencegene profiles and antimicrobial resistances. High heterogeneity was observed within the 14 strains isolated which belonged to three phylogroups (A, B1 and D), 8 serogroups (O2, O21, O29, O60, 073, O78, O103 and O141) and 13 different serotypes. Out of 34 genes screened, we have detected eight virulence genes that are typical of extraintestinal pathogenic E. coli (ExPEC) (fimH, fimAvMT78, iroN, iucD, cvaC, iss, traT and tsh);however, none of the studied strains showed the ExPEC status. The 14 strains were also analyzed for the production of extended-spectrum beta-lactamases (ESBLs) and for antimicrobial resistances. None of the 14 strains were ESBL-producing E. coli, but high resistance-prevalences to ampicillin and cotrimoxazole were detected. To our knowledge, this is the first characterization of E. coli strains isolated from griffon vulture and although they did not show high virulencegene scores, they showed cotrimoxazole resistance.

Modelling the nesting-habitat of threatened vulture species in the caucasus:An ecosystem approach to formalising environmental factors in species distribution models

Abiotic factors play an important role in species localisation,but biotic and anthropogenic predictors must also be considered in distribution modelling for models to be biologically meaningful.In this study,we formalised the biotic predictors of nesting sites for four threatened Caucasian vultures by including species distribution models(wild ungulates,nesting tree species)as biotic layers in the vulture Maxent models.Maxent was applied in the R dismo package and the best set of the model parameters were defined in the R ENMeval package.Performance metrics were continuous Boyce index,Akaike's information criterion,the area under receiver operating curve and true skill statistics.We also calculated and evaluated the null models.Kernel density estimation method was applied to assess the overlap of vulture ecological niches in the environmental space.The accessibility of anthropogenic food resources was estimated using the Path Distance measure that considers elevation gradient.The availability of pine forests(Scots Pine)and wild ungulates(Alpine Chamois and Caucasian Goat)contributed the most(29.6%and 34.3%)to Cinereous Vulture(Aegypius monachus)nesting site model.Wild ungulate distribution also contributed significantly(about 46%)to the Bearded Vulture(Gypaetus barbatus)model.This scavenger nests in the highlands of the Caucasus at a minimum distance of 5–10 km from anthropogenic facilities.In contrast,livestock as a food source was most important in colony distribution of Griffon Vulture(Gyps fulvus).The contribution of distances to settlements and agricultural facilities to the model was 45%.The optimal distance from Egyptian Vulture(Neophron percnopterus)nesting sites to settlements was only 3–10 km,to livestock facilities no more than 15 km with the factor contribution of about 57%.Excluding the wild ungulate availability,the ecological niches of studied vultures overlapped significantly.Despite similar foraging and nesting requirements,Caucasian vultures are not pronounced nesting and trophic competitors due to the abundance of nesting sites,anthropogenic food sources and successful niche sharing.

混沌自适应非洲秃鹫优化算法训练多层感知器

针对训练多层感知器(MLP)时,算法对初始值敏感、易陷入局部最优和收敛速度慢等问题,对新型启发式算法非洲秃鹫优化算法提出改进算法IAVOA。在初始化种群时引入Logistic混沌映射,增加种群的多样性;对最优秃鹫和次优秃鹫增加自适应权重系数,自动调整这两类秃鹫对普通秃鹫的引导作用;IAVOA用于MLP的训练,采用均方误差的平均值作为适应度函数寻找MLP的连接权重和偏差的最佳组合。选取4个不同复杂度的分类数据集,比较IAVOA算法与现有启发式算法对MLP训练后,MLP对数据分类的性能,仿真结果表明,IAVOA算法训练的MLP在数据分类准确率、全局搜索能力、收敛速度和稳定性方面均具有良好的性能。

改进非洲秃鹫算法的汽车充电站选址方法

充电站选址需要综合考虑多个因素,如充电需求密度、车流密度、车速等,这些因素之间存在复杂的相互关系和约束条件,如何将它们综合考虑是一个难点。为了获取用户满意度较高的选址方案,提出一种考虑交通拥堵的新能源汽车充电站选址优化方法。考虑交通拥堵指数对新能源汽车用户出行和充电站选取的影响,以运营商投资成本最小和用户满意度最大为目标,建立新能源汽车充电站选址优化模型。利用改进非洲秃鹫优化算法对模型求解,确定新能源汽车充电站选址最优解决方案。仿真实验结果表明,所提方法可以有效提升用户满意度,减少网损率和路径长度,获取更加贴合实际需求的新能源汽车充电站选址方案。

基于非洲秃鹫算法优化卷积神经网络的重叠峰解析方法

利用光谱仪器检测土壤中重金属时,由于仪器分辨率较低,峰位相近元素的特征峰会产生重叠。光谱重叠峰严重影响定量分析结果的准确性,为了得到准确的重金属含量需要进行光谱重叠峰分解。提出利用非洲秃鹫算法优化卷积神经网络(AVOA-CNN)的重叠峰解析方法。首先,利用高斯函数模型模拟出150个双高斯含噪光谱重叠峰和43个三高斯含噪光谱重叠峰,选择不同小波基函数进行光谱数据去噪,以信噪比和均方根误差(root mean square error, RMSE)为评价指标,最终确定coif 3小波基函数,使用导数法进行光谱重叠峰预处理。然后,使用AVOA-CNN获得卷积神经网络(convolutional neural network, CNN)预测结果,解析结果表明:AVOA-CNN成功分解重叠峰且准确率高,双高斯光谱重叠峰和三高斯光谱重叠峰参数(峰强度,峰位,峰宽)的最大相对误差平均值分别为3.15%和5.90%。最后对比麻雀搜索算法优化CNN、CNN与AVOA-CNN,结果显示AVOA-CNN模型预测准确率最高。

基于组稀疏学习与AVOA-XGBoost的轴承故障分级诊断

针对工业设备中轴承振动信号在噪声环境下故障分级诊断准确率低的问题,提出一种基于组稀疏学习与非洲秃鹫优化算法优化极端梯度提升树(African vultures optimization algorithm-extreme gradient boosting,AVOA-XGBoost)的轴承故障分级诊断方法。首先,利用组稀疏学习对轴承振动信号进行重构,以降低噪声水平并更有效地表征故障脉冲。然后,对重构后的信号提取时域、频域和熵值特征并构建特征集。最后,利用AVOA自适应优化XGBoost超参数以建立稳健的XGBoost模型,进而高效实现轴承故障分级诊断。试验结果表明,经过组稀疏学习重构的信号具备更强故障特征表示能力,相较于传统机器学习模型,采用AVOA-XGBoost模型进行分类能够取得更高准确率,所提方法能够有效诊断轴承故障类型及故障程度。

基于改进AVOA的小型自由活塞斯特林制冷机控制算法研究

针对小型自由活塞斯特林制冷机在温度控制系统中建模复杂以及易受环境温度和热负载变化的影响等问题,提出一种基于改进非洲秃鹫优化算法的模糊PID控制算法。通过在一台10 W@77 K自由活塞斯特林制冷机的温度控制系统中加入滞环偏置继电器辨识制冷机的数学模型,以此构建闭环控制系统,着重针对基于改进非洲秃鹫算法的模糊PID控制算法进行仿真实验。针对非洲秃鹫优化算法存在的容易陷入局部最优、收敛速度慢等问题,提出了一种融合多策略的改进非洲秃鹫优化算法,其中引入Tent混沌映射以提高初始个体多样性,并对次优秃鹫选择策略进行改进,加强算法脱离局部困境的能力,最后利用精英反向学习进一步增强算法全局搜索能力。仿真实验结果表明,该算法与PID算法、模糊PID算法和非洲秃鹫算法优化的模糊PID算法相比,调节时间分别降低79.69%、56.22%、50.26%,超调量分别减少87.70%、59.81%、60.38%;在加入方波扰动、白噪声、脉冲扰动和正弦波扰动下,控制系统的超调量的变化率分别为2.61%、2.35%、2.44%和2.49%。因此,基于改进非洲秃鹫算法的模糊PID控制算法与常规控制算法相比,具有响应速度快、超调小、抗干扰性强和稳定性好等优点。

基于双重动态调整的改进非洲秃鹫优化算法

针对非洲秃鹫优化算法(African vulture optimization algorithm, AVOA)多样性低、探索开发能力不平衡、易发生早熟的现象,提出了一种基于双重动态调整的改进非洲秃鹫优化算法(improvement African vulture optimization algorithm, IAVOA)。改进后的算法分为3个部分,通过引入混沌映射初始化种群,以确保种群在前期寻优中具有较高的多样性;加入动态调整因子来确定当前最优个体,用来平衡前期探索与后期开发的能力;针对AVOA中饥饿率的变化情况加入动态调整的高斯扰动,用于防止早熟问题的发生,提高最终解的质量。改进后的算法在9个标准测试函数上进行测试。结果表明,该算法表现出更佳的求解性能。

一种自适应选取参数的改进变分模态分解方法

针对传统的变分模态分解(VMD)方法中模态数和惩罚参数难以确定的问题,提出了一种自适应选取参数的改进变分模态分解方法。首先,综合考虑了故障的冲击性和周期性特点,以Gini指数和谱峰比指标为基础构建了加权谱峰比(WSPR)指标;然后,采用非洲秃鹫优化算法(AVOA)进行了寻优,得到了最佳的模态数和惩罚参数组合,克服了人为主观选择参数的弊端;最后,在VMD分解信号后,利用加权谱峰比最大原则自适应选取了敏感内涵模态分量,对最佳IMF进行了包络解调分析,提取了滚动轴承早期故障特征,利用仿真信号、单一故障滚动轴承试验信号及复合故障滚动轴承试验信号对所述方法进行了验证。实验结果表明:该方法可以准确地提取出仿真信号的故障频率(100 Hz)、单一故障信号的故障频率(236.4 Hz)和复合故障信号的故障频率(内圈故障频率149.14 Hz、外圈故障频率86.39 Hz),并且在与其他方法和指标的对比中,其最佳IMF的包络谱图中故障特征频率及其倍数频的谱峰更加明显,准确率更高且鲁棒性更强。研究结果表明:该方法能够有效提取轴承早期故障信号的微弱特征,实现故障类型准确识别的目的。

基于测距修正和秃鹫优化的DV-Hop改进算法

针对传统距离矢量跳(DV-Hop)算法在各向异性网络中定位误差大的问题,提出一种基于测距修正和秃鹫优化的DV-Hop改进算法(AVOADV-Hop)。该算法在双通信半径的基础上依据锚节点间的跳数信息引入测距修正阈值,根据锚节点的坐标信息计算基于跳数改进的平均跳距,实现节点间距离估算的分段处理;融合动态调整和加权思想改进非洲秃鹫优化算法取代最小二乘法求解节点坐标。仿真结果表明,在3种不规则网络下,AVOADV-Hop算法相较于传统DV-Hop算法及相关改进算法,定位误差分别下降了43.6%、21.1%和10.5%,定位精度有效提高。

基于IAVOA-GRU网络的高频地波雷达电离层杂波预测

电离层杂波的精确预测对提升高频地波雷达的探测性能具有重要作用。提出了一种基于改进非洲秃鹫优化算法优化门控循环单元(Improved African Vultures Optimization Algorithm Optimization Gated Recurrent Unit,IAVOA-GRU)网络的电离层杂波预测方法。首先,依据电离层杂波的混沌特性,通过相空间重构方法对接收到的电离层杂波进行相空间重建,构建GRU网络的输入、输出样本集;然后,利用IAVOA对GRU网络的隐层节点数、迭代次数及初始学习速率3个超参数值执行优选;最后,重新训练优化后的GRU网络,并进行预测。实测结果表明,相较其他6种对比预测模型,所提出的IAVOA-GRU网络模型具有较高的预测精度和可靠性,为有效改善高频地波雷达的探测性能提供了一种思路和方法。

基于EMD-AVOA-BP的逆变器故障诊断方法

以CRH3C型动车组逆变器中的绝缘栅双极型晶体管(IGBT,Insulated Gate Bipolar Transistor)双管开路故障为研究对象,提出了一种基于非洲秃鹫算法(AVOA,African Vultures Optimization Algorithm)和优化的反向传播(BP,Back Propagation)神经网络的逆变器故障诊断方法。在Simulink中搭建列车逆变器的控制模型,取得故障电流;采用经验模态分解(EMD, Empirical Mode Decomposition)对电流信号进行去噪和故障特征提取,再利用AVOA对BP神经网络进行优化,实现了对列车逆变器IGBT双管开路故障的诊断。与传统方法进行对比可知,该方法具有更高的精准度,在测试集中其精准度达到100%。

基于改进鲸鱼算法优化LSTM的化工过程故障诊断方法

鲸鱼优化算法是一种群体智能优化算法,文中针对基本鲸鱼优化算法收敛速度慢、收敛精度低、易陷入局部最优等问题,从几个方面对其进行了改进:通过Tent混沌对种群进行初始化来增加种群多样性;融入非洲秃鹫算法的群体最优和次优策略与探索阶段位置更新策略,以避免算法早熟以及陷入局部最优;采用一种新的非线性收敛因子替代鲸鱼优化算法原本的线性收敛因子,平衡算法的全局探索和局部开发;引入了非线性自适应增量惯性权重,更好地平衡了全局搜索能力与局部搜索能力;最终得到一种混合非洲秃鹫算法的改进鲸鱼优化算法(MAWOA)。在对4种基准测试函数进行的对比试验中显示,MAWOA具有较快的收敛速度和较高的收敛精度。将MAWOA算法应用于长短期记忆(LSTM)网络的超参数寻优中,构建MAWOA-LSTM故障诊断模型。结合田纳西伊斯曼(TE)化工数据集进行故障诊断,通过与LSTM、WOA-LSTM等模型进行准确率对比,验证了所提算法的优越性。

Improved AVOA based on LSSVM for wind power prediction

Improving the prediction accuracy of wind power is an effective means to reduce the impact of wind power on power grid.Therefore,we proposed an improved African vulture optimization algorithm(AVOA)to realize the prediction model of multi-objective optimization least squares support vector machine(LSSVM).Firstly,the original wind power time series was decomposed into a certain number of intrinsic modal components(IMFs)using variational modal decomposition(VMD).Secondly,random numbers in population initialization were replaced by Tent chaotic mapping,multi-objective LSSVM optimization was introduced by AVOA improved by elitist non-dominated sorting and crowding operator,and then each component was predicted.Finally,Tent multi-objective AVOA-LSSVM(TMOALSSVM)method was used to sum each component to obtain the final prediction result.The simulation results show that the improved AVOA based on Tent chaotic mapping,the improved non-dominated sorting algorithm with elite strategy,and the improved crowding operator are the optimal models for single-objective and multi-objective prediction.Among them,TMOALSSVM model has the smallest average error of stroke power values in four seasons,which are 0.0694,0.0545 and 0.0211,respectively.The average value of DS statistics in the four seasons is 0.9902,and the statistical value is the largest.The proposed model effectively predicts four seasons of wind power values on lateral and longitudinal precision,and faster and more accurately finds the optimal solution on the current solution space sets,which proves that the method has a certain scientific significance in the development of wind power prediction technology.

基于改进非洲秃鹫算法优化极限学习机的船舶运动预测

针对船舶运动预测模型精度不高而造成预测结果误差大的问题,提出一种利用改进非洲秃鹫优化算法(IAVOA)优化模型参数的极限学习机(ELM)预测模型,对船舶运动状况进行预测。在初始化种群时引入Circle混沌映射,增加种群的多样性;加入自适应算子,调整两类秃鹫对其他秃鹫的指引作用,提升算法的收敛速度和解的质量。利用IAVOA优化的ELM模型对船模水池试验运动数据进行预测,并采用均方根误差和平均绝对误差评判该预测模型,与现有其他启发式算法优化ELM模型比较,所提出的IAVOA-ELM具有更优的预测精度和泛化能力。

基于改进非洲秃鹫优化算法的脑MRI图像分割

针对非洲秃鹫优化算法(African vulture optimization algorithm,AVOA)在开发能力与探索能力之间存在的失衡问题,提出了一种多策略改进的非洲秃鹫优化算法(improved African vulture optimization algorithm,IAVOA)。采用佳点集初始化种群以增强多样性,引入混合对立学习以强化开发与探索,实施自适应信任度策略以动态调整搜索过程,应用高斯变异来进一步平衡算法的开发和探索能力。仿真结果显示,在12个典型测试函数上,IAVOA相较于对比算法,在收敛速度、求解精度和稳定性方面均显著提升。提出了IAVOA-FCM算法用于小样本数据集的脑磁共振成像(magnetic resonance imaging,MRI)图像分割,通过IAVOA算法强大的全局寻优能力对FCM算法进行优化。在脑MRI图像分割实验中,与其他5种先进的结合算法相比,IAVOA-FCM在分割精度、稳定性等方面均表现出显著优势。