Winter Wheat Yield Estimation Based on Sparrow Search Algorithm Combined with Random Forest:A Case Study in Henan Province,China

Precise and timely prediction of crop yields is crucial for food security and the development of agricultural policies.However,crop yield is influenced by multiple factors within complex growth environments.Previous research has paid relatively little attention to the interference of environmental factors and drought on the growth of winter wheat.Therefore,there is an urgent need for more effective methods to explore the inherent relationship between these factors and crop yield,making precise yield prediction increasingly important.This study was based on four type of indicators including meteorological,crop growth status,environmental,and drought index,from October 2003 to June 2019 in Henan Province as the basic data for predicting winter wheat yield.Using the sparrow search al-gorithm combined with random forest(SSA-RF)under different input indicators,accuracy of winter wheat yield estimation was calcu-lated.The estimation accuracy of SSA-RF was compared with partial least squares regression(PLSR),extreme gradient boosting(XG-Boost),and random forest(RF)models.Finally,the determined optimal yield estimation method was used to predict winter wheat yield in three typical years.Following are the findings:1)the SSA-RF demonstrates superior performance in estimating winter wheat yield compared to other algorithms.The best yield estimation method is achieved by four types indicators’composition with SSA-RF)(R^(2)=0.805,RRMSE=9.9%.2)Crops growth status and environmental indicators play significant roles in wheat yield estimation,accounting for 46%and 22%of the yield importance among all indicators,respectively.3)Selecting indicators from October to April of the follow-ing year yielded the highest accuracy in winter wheat yield estimation,with an R^(2)of 0.826 and an RMSE of 9.0%.Yield estimates can be completed two months before the winter wheat harvest in June.4)The predicted performance will be slightly affected by severe drought.Compared with severe drought year(2011)(R^(2)=0.680)and normal year(2017)(R^(2)=0.790),the SSA-RF model has higher prediction accuracy for wet year(2018)(R^(2)=0.820).This study could provide an innovative approach for remote sensing estimation of winter wheat yield.yield.

Effects of ginsenoside of stem and leaf combined with choline on learning and memory ability of rat models with Alzheimer diseases

BACKGROUND： Central adrenergic nerve and 5-serotonergic nerve can influence central cholinergic nerve on learning and memory and make easy for study; however, ginsenoside of stem and leaf （GSL） can improve functions of central adrenergic nerve; moreover, 5-serotonergic nerve and the combination with choline can produce synergistic effect and enhance learning and memory ability so as to improve learning and memory disorder of patients with Alzheimer disease （AD）. OBJECTIVE ： To observe the effects of GSL combining with choline on learning and memory of AD model rats DESIGN ： Randomized grouping design and controlled animal study SETIING ： Department of Pharmacology, Taishan Medical College MATERIALS ： The experiment was carried out in the Pharmacological Department of Medical College of Jilin University from October 1996 to January 1997. Forty healthy male Wistar rats of clean grade were randomly divided into 5 groups, including sham-injury group, model group, GSL group, choline group and combination group, with 8 rats in each group. Main medications： GSL with the volume more than 92.8% was provided by Department of Chemistry, Norman Bethune Medical College of Jilin University. Panaxatriol, the main component, was detected with thin layer scanning technique and regarded as the index of GSL quality [（55±1）%, CV= 2%, n = 5]. Choline was provided by the Third Shanghai Laboratory Factory. METHODS ： 150 nmol quinolinic acid was used to damage bilateral Meynert basal nuclei of adult rats so as to establish AD models. Rats in GSL, choline and combination groups were intragastric administrated with 400 mg/kg GSL, 200 mg/kg choline （20 mL/kg）, and both respectively last for 17 days starting from two days before operation. Rats in sham-injury group and model group were perfused with the same volume of distilled water once in each morning for the same days. （1） Passive avoidance step-down test： Five minutes later, rats jumped up safe platform when they were shocked with 36 V alternating current. If rats jumped down from the platform and the feet touched railings, the response was wrong. Numbers of wrong response were recorded within 3 minutes, and then the test was redone after 24 hours. （2） Morris water-maze spatial localization task： Swimming from jumping-off to platform directly was regarded as right response. Additionally, 4 successively right responses were regarded as the standard. Each rat was trained 10 times a day with 120 s per time for 3 successive days. The interval was 30 s. Three days later, numbers of right response were recorded. The training times were increased to 30 for unlearned rats. （3） Measurement of activity of choline acetylase in cerebral cortex： Rats were sacrificed at 17 days after operation to obtain cerebral cortex to measure activity of choline acetylase with radiochemistry technique. （4） Synergistic effect： It was expressed as Q value： Q value = factual incorporative effect/anticipant incorporative effect; Q ≥ 1 was regarded as synergistic effect. Anticipant incorporative effect = （EA＋EB-EA·EB）, EA and EB were single timing effect, respectively in GSL group and choline group. E（step-down test and Morris water maze test） = （x in model group - factual value in medicine groups）/x in model group; E （activity of choline acetylase） = （factual value in medicine groups -xin model group）/xin model group. MAIN OUTCOME MEASURES ： （1） Passive avoidance step-down test and Morris water-maze spatial localization task in the study of learning and memory; （2） activity of choline acetylase. RESULTS ： All 40 rats were involved in the final analysis. （1） Passive avoidance response： At learning phase on first day and retesting phase on the next day, numbers of wrong responses within 3 minutes were more in model group than sham operation group, and there was significant difference [（5.88±1.46）, （2.25±0.87） times; （2.63±1.06）, （0.50±0.53） times; P 〈 0.01]; numbers of wrong responses within 3 minutes were less in combination group than model group, and there was significant difference [learning phase： （1.12±0.83）, （5.88±1.46） times; retesting phase： （0.38±0.74）, （2.63±1.06）times, P 〈 0.01]; moreover, effect was stronger than that in GSL group and choline group. The Q value was 1.07 and 1.59, respectively and it showed synergistic effect. Spatial localization task： Training times were more in model group than sham operation group, and there was significant difference [（2.9±2.5）, （12.6±3.5） times; P 〈 0.01]. Training times were less in combination group than model group, and there was significant difference [（11.8±2.4）, （27.9±2.5） times, P 〈 0.01]; moreover, effect was stronger than that in GSL group and choline group. The Q value was 1.07 and it showed synergistic effect. （3） Activity of choline acetylase： Activity was lower in model group than sham operation group, and there was significant difference [（30.56±8.33）, （61.11 ±8.33） nkat/g; P 〈 0.01]. Activity was higher in combination group than model group and there was significant difference [（50.00±8.33）, （30.56±8.33） nkat/g, P 〈 0.01];moreover, effect was stronger than that in GSL group and choline group. The Q value was 1.5 and it showed synergistic effect. CONCLUSZON： GSL in combination with choline can synergically improve the disorder of learning and memory of AD model rats. Its mechanism may be involved in enhancing the function of central cholinergic system.

Fuzzy Entropy Based Combined Learning Algorithm for Neural Networks

Learning is one of key problems of artificial neural networks. In this paper, we present a kind of combined learning algorithm based on fuzzy entropy criterion for neural networks. The basic idea is to simulate the learning mechanism of human brain and overcome the limitations of monocrifsterion learning. The comparison is made between the given learning algorithm and the typical BP algorithm in order to show the characteristics of the new algorithm.

基于e-learning平台的“工学结合”教学模式探索

解决身处异地学生的继续学习问题,是"工学结合"教学模式改革成功的保障。通过e-learning教学平台作用的描述,介绍了e-learning教学平台的构建,阐述了通过e-learning教学平台实施教学。

The application of machine learning under supervision in identification of shale lamina combination types——A case study of Chang 7_(3)sub-member organic-rich shales in the Triassic Yanchang Formation,Ordos Basin,NW China

Organic rich laminated shale is one type of favorable reservoirs for exploration and development of continental shale oil in China.However,with limited geological data,it is difficult to predict the spatial distribution of laminated shale with great vertical heterogeneity.To solve this problem,taking Chang 73 sub-member in Yanchang Formation of Ordos Basin as an example,an idea of predicting lamina combinations by combining'conventional log data-mineral composition prediction-lamina combination type identification'has been worked out based on machine learning under supervision on the premise of adequate knowledge of characteristics of lamina mineral components.First,the main mineral components of the work area were figured out by analyzing core data,and the log data sensitive to changes of the mineral components was extracted;then machine learning was used to construct the mapping relationship between the two;based on the variations in mineral composition,the lamina combination types in typical wells of the research area were identified to verify the method.The results show the approach of'conventional log data-mineral composition prediction-lamina combination type identification'works well in identifying the types of shale lamina combinations.The approach was applied to Chang 73 sub-member in Yanchang Formation of Ordos Basin to find out planar distribution characteristics of the laminae.

The Feasibility of Immobilization of Bioflocculant-producing Bacteria Using Mycelial Pellets as Biomass Carriers

Bioflocculant-producing bacteria Agrobacterium tumefaciens F2 and Bacillus sphaeicus F6 were immobilized onto mycelial pellets to investigate the bioflocculant-producing potential of this combined organism and the corresponding flocculating efficiency. The atomic force microscope ( AFM) images of mycelial surface indicate that the mycelia pellet can immobilize bioflocculant-producing bacteria F2 and F6 as a biomass carrier. The flocculating efficiency of bioflocculant produced by this combined organism was studied under the optimum flocculating conditions obtained by Response Surface Methodology ( RSM ) . The fermentation yield of the combined mycelial pellet is about 2. 6 g / L,which is higher than that of the free bacteria ( only 2. 2 g / L) . Flocculating efficiency of the combined mycelial pellet was comparable with that of bioflocculant generated by the free bacteria. The bioflocculant yield is enhanced and the flocculating efficiency of the co-culture is uninfluenced after immobilized with mycelial pellet as a carrier. In conclusion,the mycelial pellet is feasible as a biomass carrier for the immobilization of bioflocculant-producing bacteria.

Face Super-resolution Reconstruction and Recognition Using Non-local Similarity Dictionary Learning Based Algorithm

One of the challenges of face recognition in surveillance is the low resolution of face region. Therefore many superresolution (SR) face reconstruction methods are proposed to produce a high-resolution face image from one or a set of low-resolution face images. However, existing dictionary learning based algorithms are sensitive to noise and very time-consuming. In this paper, we define and prove the multi-scale linear combination consistency. In order to improve the performance of SR, we propose a novel SR face reconstruction method based on nonlocal similarity and multi-scale linear combination consistency (NLS-MLC). We further proposed a new recognition approach for very low resolution face images based on resolution scale invariant feature (RSIF). A series of experiments are conducted on two public face image databases to test feasibility of our proposed methods. Experimental results show that the proposed SR method is more robust and computationally effective in face hallucination, and the recognition accuracy of RSIF is higher than some state-of-art algorithms. © 2014 Chinese Association of Automation.

Novel ensemble learning based on multiple section distribution in distributed environment

Because most ensemble learning algorithms use the centralized model, and the training instances must be centralized on a single station, it is difficult to centralize the training data on a station. A distributed ensemble learning algorithm is proposed which has two kinds of weight genes of instances that denote the global distribution and the local distribution. Instead of the repeated sampling method in the standard ensemble learning, non-balance sampling from each station is used to train the base classifier set of each station. The concept of the effective nearby region for local integration classifier is proposed, and is used for the dynamic integration method of multiple classifiers in distributed environment. The experiments show that the ensemble learning algorithm in distributed environment proposed could reduce the time of training the base classifiers effectively, and ensure the classify performance is as same as the centralized learning method.

A combined forecasting method for intermittent demand using the automotive aftermarket data

Intermittent demand forecasting is an important challenge in the process of smart supply chain transformation,and accurate demand forecasting can reduce costs and increase efficiency for enterprises.This study proposes an intermittent demand combination forecasting method based on internal and external data,builds intermittent demand feature engineering from the perspective of machine learning,predicts the occurrence of demand by classification model,and predicts non-zero demand quantity by regression model.Based on the strategy selection on the inventory side and the stocking needs on the replenishment side,this study focuses on the optimization of the classification problem,incorporates the internal and external data of the enterprise,and proposes two combination forecasting optimization methods on the basis of the best classification threshold searching and transfer learning,respectively.Based on the real data of auto after-sales business,these methods are evaluated and validated in multiple dimensions.Compared with other intermittent forecasting methods,the models proposed in this study have been improved significantly in terms of classification accuracy and forecasting precision,which validates the potential of combined forecasting framework for intermittent demand and provides an empirical study of the framework in industry practice.The results show that this research can further provide accurate upstream inputs for smart inventory and guarantee intelligent supply chain decision-making in terms of accuracy and efficiency.

Estimation of Potato Biomass and Yield Based on Machine Learning from Hyperspectral Remote Sensing Data

The estimation of potato biomass and yield can optimize the planting pattern and tap the production potential.Based on partial least square(PLSR),multiple linear regression(MLR),support vector machine(SVM),random forest(RF),BP neural network and other machine learning algorithms,the biomass estimation model of potato in different growth stages is constructed by using single variables such as original spectrum,first-order differential spectrum,combined spectrum index and vegetation index(VI)and their coupled combination variables.The accuracy of the models is compared and analyzed,and the best modeling method of biomass in different growth stages is selected.Based on the optimized modeling method,the biomass of each growth stage is estimated,and the yield estimation model of different growth stages is constructed based on the estimation results and the linear regression analysis method,and the accuracy of the model is verified.The results showed that in tuber formation stage,starch accumulation stage and maturity stage,the biomass estimation accuracy based on combination variable was the highest,the best modeling method was MLR and SVM,in tuber growth stage,the best modeling method was MLR,the effect of yield estimation is good.It provides a reference for the algorithm selection of crop biomass and yield models based on machine learning.

Research on Dynamic Forecast of Flowering Period Based on Multivariable LSTM and Ensemble Learning Classification Task

The flowering forecast provides recommendations for orchard cleaning, pest control, field management and fertilization, which can help increase tree vigor and resistance. Flowering forecast is not only an important part of the construction of agro-meteorological index system, but also an important part of the meteorological service system. In this paper, by analyzing local meteorological data and phenological data of “Red Fuji” apples in Fen County, Linfen City, Shanxi Province, with the help of machine learning and neural networks, we proposed a method based on the combination of time series forecasting and classification forecasting is proposed to complete the dynamic forecasting model of local flowering in Ji County. Then, we evaluated the effectiveness of the model based on the number of error days and the number of days in advance. The implementation shows that the proposed multivariable LSTM network has a good effect on the prediction of meteorological factors. The model loss is less than 0.2. In the two-category task of flowering judgment, the idea of combining strategies in ensemble learning improves the effect of flowering judgment, and its AUC value increases from 0.81 and 0.80 of single model RF and AdaBoost to 0.82. The proposed model has high applicability and accuracy for flowering forecast. At the same time, the model solves the problem of rounding decimals in the prediction of flowering dates by the regression method.

Producing Podcasts in a Chinese Classroom: A Digital Humanities Project

This action research which situates itself in the context of Digital Humanities focuses on engaging students of Introductory Chinese course with an assignment to create podcasts in the form of an interview.Requiring students to integrate the aural,oral,and textural information via technology so as to carry a conversation with an interlocutor challenges them to convey their thoughts from the language they think in into meaningful ideas using the target language.In this way,students are“pushed”to convert their listening,speaking,and writing from being isolated components into a more dynamic and cohesive mode of communicating.As they explore expressing their ideas in the language they are learning,they develop the skills and sensibility needed to critically analyze their own work.Although limited in scale,the podcast format urges students to transform what they need to learn into a familiar product.

煤矿井下掘进机器人路径规划方法研究

针对煤矿非全断面巷道条件下掘进机器人移机难度大、效率低下等问题,分析了煤矿井下非结构化环境特征及掘进机器人运动特性,提出了基于深度强化学习的掘进机器人机身路径规划方法。利用深度相机将巷道环境实时重建,在虚拟环境中建立掘进机器人与巷道环境的碰撞检测模型,并使用层次包围盒法进行虚拟环境碰撞检测,形成巷道边界受限下的避障策略。考虑到掘进机器人形体大小且路径规划过程目标单一,在传统SAC算法的基础上引入后见经验回放技术,提出HER-SAC算法,该算法通过环境初始目标得到的轨迹扩展目标子集,以增加训练样本、提高训练速度。在此基础上,基于奖惩机制建立智能体,根据掘进机器人运动特性定义其状态空间与动作空间,在同一场景下分别使用3种算法对智能体进行训练,综合平均奖励值、最高奖励值、达到最高奖励值的步数以及鲁棒性4项性能指标进行对比分析。为进一步验证所提方法的可靠性,采用虚实结合的方式,通过调整目标位置设置2种实验场景进行掘进机器人的路径规划,并将传统SAC算法和HER-SAC算法的路径结果进行对比。结果表明:相较于PPO算法和SAC算法,HER-SAC算法收敛速度更快、综合性能达到最优;在2种实验场景下,HER-SAC算法相比传统SAC算法规划出的路径更加平滑、路径长度更短、路径终点与目标位置的误差在3.53 cm以内,能够有效地完成移机路径规划任务。该方法为煤矿掘进机器人的自主移机控制奠定了理论基础,为煤矿掘进设备自动化提供了新方法。

基于组合深度学习的轨道交通短时进站客流预测模型

针对轨道交通短时进站客流考虑不充分和特征学习不全面而导致预测精度不高的问题,选取客流特征、天气、空气质量和道路交通拥堵指数等多个因素,提出了一种基于组合深度学习的轨道交通短时进站客流预测模型(CNN-ResNet-BiLSTM)。基于卷积神经网络(CNN)对多因素客流时间序列进行自动提取,在CNN网络中插入多个残差神经网络(ResNet)来加深网络深度,利用双向长短时记忆神经网络(BiLSTM)捕捉前后两个方向的客流时间序列特征并得到预测结果;以杭州市全网80个站点工作日的进站客流为例,验证了该模型的有效性。研究结果表明:与常用的几种模型相比,多因素CNN-ResNet-BiLSTM组合模型的均方根误差(E RMS)至少降低了8.50%,平均绝对误差(E MA)至少降低了6.74%,平均绝对百分比误差(E MPA)至少降低了6.52%。

基于CEEMDAN-GRU组合模型的碳排放交易价格预测研究

准确的碳价格预测有助于监管部门观测碳交易市场运行状况及投资者进行科学决策,对实现碳达峰和碳中和具有重要作用。但碳价序列具有非线性、非平稳性和高噪声的特性,很难对其进行准确预测。将完全自适应噪声集合经验模态分解(CEEMDAN)方法与门控循环单元(GRU)相结合,构建一个碳排放交易价格预测模型。该模型基于分解、集成思想,利用CEEMDAN将原始碳价序列分解,获得不同频率的本征模函数(IMF)和残差序列,使用GRU神经网络分别为各子序列建立预测模型,最后集成预测结果得到碳价预测值。以湖北省碳交易市场的日度成交价为例进行实证分析,结果表明:相较于其他5种基准模型,CEEMDAN-GRU模型具有更小的预测误差和更高的拟合优度,在碳价格预测上具有一定的优势。

基于可解释机器学习的黄河源区径流分期组合预报

黄河源区是黄河流域重要的产流区和我国重要的清洁能源基地,提高黄河源区径流预报准确率可为流域水资源科学调配和水风光清洁能源高效利用提供重要支撑。以黄河源区唐乃亥和玛曲水文站为研究对象,基于不同月份径流组分的差异,考虑积雪覆盖率及融雪水当量变化,构建了中长期径流分期组合机器学习预报模型及其可解释性分析框架。研究结果表明:1)年内的径流预报时段可划分为融雪影响期(3—6月)和非融雪主导(以降雨和地下水补给为主)期(7月—次年2月);2)与传统不分期模型相比,唐乃亥站和玛曲站分期组合预报模型的纳什效率系数分别达0.897、0.835,确定系数(R2)分别达0.897、0.839,均方根误差分别降低了10%、17%,提高了径流预报准确率,通过分位数映射校正,唐乃亥站和玛曲站预报模型的R2分别进一步提升至0.926和0.850;3)基于SHAP机器学习可解释性分析框架,辨识了预报因子对径流预报结果的贡献程度,由高到低依次为降水、前一个月流量、蒸发、气温、相对湿度、融雪水当量等,发现了不同预报因子之间交互作用散点分布具有拖尾式或阶跃式的特征。

基于集成学习与深度学习的洪水径流预报研究

深度学习模型凭借其对水文因素间复杂作用的优秀处理能力,在水文预报领域得到了一定的应用,然而,针对集成学习与深度学习耦合模型的研究仍有所缺失。通过融合集成学习AdaBoost算法与深度学习Informer模型,提出了一种组合模型,称为AdaBoost-Informer模型,以提高洪水径流预报的精度。该模型以历史雨量和径流数据作为数据输入,将具备长时序依赖捕获能力的Informer作为集成学习的弱预测器,使用网格搜索法进行超参数调优,使用AdaBoost集成学习算法对弱预测器进行加权组合得到强预测器。在浙江省椒江流域的应用分析表明:对比Random Forest、AdaBoost、Transformer、Informer等模型,AdaBoost-Informer模型表现最佳,RMSE为62.08 m^(3)/s,MAE为23.83 m^(3)/s,NSE为0.980,预报合格率为100%。所提模型可有效提高洪水预报精度,为防汛抢险和防洪系统调度提供决策依据。

线下线上混合式教学模式在机械原理教学中的应用

为适应工程教育认证,切实提高机械原理课程的学习效果,机械原理课程教学方法和教学内容等需要进行相应教学改革。课题组以成果导向为指导理念,优化教学内容;以“学习通”、腾讯会议、QQ群为平台,开展线下线上混合式教学模式,以机械原理“自由度计算基本公式”专题为例开展教学设计,探索线下线上混合教学模式的应用。结果表明,混合式教学模式能有效增加教学内容,激发学生自主学习的兴趣和能力,达到“童蒙求我”的学习效果。

基于改进U-Net的车间场景分割

为了更好地对车间环境进行把控,同时针对车间目标像素级分割任务样本少、类别多、尺度变化大导致的语义分割精度低的问题,设计了一种改进的U-Net车间场景分割模型。改进的模型采用Rep-VGG轻量级主干网络,并在U-Net上采样阶段引入金字塔拆分注意力机制,以增加模型的特征表达能力及推理速度。模型训练则采用Dice-Cross Entropy组合损失函数以解决样本不均衡导致的难以训练的问题。实验数据表明该模型在自建的小样本车间数据集上可达到快速、轻量化及高精度的分割效果。

基于排序蒸馏的序列化推荐算法

为解决当前基于知识蒸馏的推荐算法排名有效性和效率低,以及现有知识蒸馏模型更强调的是静态和单一知识迁移的问题,提出一种基于排序蒸馏的序列化推荐算法。训练一个性能优越、规模大的教师模型,训练一个符合移动终端设备的小模型即学生模型,使学生模型在教师模型的指导下学习排序。学生模型实现了与教师模型相似的排名性能,且学生模型规模较小提高了在线推荐效率。通过在数据集MovieLens和Gowalla上的实验,验证了该模型增强了学生模型的学习效果,缓解了学生模型学习不充分导致排名不佳的问题。模型可以自然地运用于序列化推荐的模型中,具有很好的通用性。