Exploring the Efficiency of Experimental Construction of Sorting Ginned Cotton Seed Machine

In this paper, research has been conducted to increase the quantity of fiber produced in the enterprise by creating a sorting device for spun seeds, dividing them into fractions by geometric dimensions, and by re-ginning, separating those with long fibers. A new model was developed for geometric sorting of cotton seeds in the harvest, and experiments determined its effectiveness and the optimal values of the factors affecting the efficiency using mathematical modeling. Based on the results of the study, graphs of the influence of factors on device performance and on device efficiency were constructed.

Effects of planting patterns on yield, quality, and defoliation in machine-harvested cotton

The aim of this study was to elucidate the effects of different machine-harvested cotton-planting patterns on defoliation,yield,and fiber quality in cotton and to provide support for improving the quality of machine-harvested cotton.In the 2015 and 2016 growing seasons,the Xinluzao 45(XLZ45)and Xinluzao 62(XLZ62)cultivars,which are primarily cultivated in northern Xinjiang,were used as study materials.Conventional wide-narrow row(WNR),wide and ultra-narrow row(UNR),wide-row spacing with high density(HWR),and wide-row spacing with low density(LWR)planting patterns were used to assess the effects of planting patterns on defoliation,yield,and fiber quality.Compared with WNR,the seed cotton yields were significantly decreased by 2.06–5.48%for UNR and by 2.50–6.99%for LWR,respectively.The main cause of reduced yield was a reduction in bolls per unit area.The variation in HWR yield was–1.07–1.07%with reduced bolls per unit area and increased boll weight,thus demonstrating stable production.In terms of fiber quality indicators,the planting patterns only showed significant effects on the micronaire value,with wide-row spacing patterns showing an increase in the micronaire values.The defoliation and boll-opening results showed that the number of leaves and dried leaves in HWR was the lowest among the four planting patterns.Prior to the application of defoliating agent and before machine-harvesting,the numbers of leaves per individual plant in HWR were decreased by 14.45 and 25.00%on average,respectively,compared with WNR,while the number of leaves per unit area was decreased by 27.44 and 36.21%on average,respectively.The rates of boll-opening and defoliation in HWR were the highest.Specifically,the boll-opening rate before defoliation and machine-harvesting in HWR was 44.54 and 5.94%higher on average than in WNR,while the defoliation rate prior to machine-harvesting was 3.45%higher on average than in WNR.The numbers of ineffective defoliated leaves and leaf trash in HWR were the lowest,decreased by 33.40 and 32.43%,respectively,compared with WNR.In conclusion,the HWR planting pattern is associated with a high and stable yield,does not affect fiber quality,promotes early maturation,and can effectively decrease the amount of leaf trash in machine-picked seed cotton,and thus its use is able to improve the quality of machine-harvested cotton.

CREDIT-X1local:A reference dataset for machine learning seismology from ChinArray in Southwest China

High-quality datasets are critical for the development of advanced machine-learning algorithms in seismology.Here,we present an earthquake dataset based on the ChinArray Phase I records(X1).ChinArray Phase I was deployed in the southern north-south seismic zone(20°N-32°N,95°E-110°E)in 2011-2013 using 355 portable broadband seismic stations.CREDIT-X1local,the first release of the ChinArray Reference Earthquake Dataset for Innovative Techniques(CREDIT),includes comprehensive information for the 105,455 local events that occurred in the southern north-south seismic zone during array observation,incorporating them into a single HDF5 file.Original 100-Hz sampled three-component waveforms are organized by event for stations within epicenter distances of 1,000 km,and records of≥200 s are included for each waveform.Two types of phase labels are provided.The first includes manually picked labels for 5,999 events with magnitudes≥2.0,providing 66,507 Pg,42,310 Sg,12,823 Pn,and 546 Sn phases.The second contains automatically labeled phases for 105,442 events with magnitudes of−1.6 to 7.6.These phases were picked using a recurrent neural network phase picker and screened using the corresponding travel time curves,resulting in 1,179,808 Pg,884,281 Sg,176,089 Pn,and 22,986 Sn phases.Additionally,first-motion polarities are included for 31,273 Pg phases.The event and station locations are provided,so that deep learning networks for both conventional phase picking and phase association can be trained and validated.The CREDIT-X1local dataset is the first million-scale dataset constructed from a dense seismic array,which is designed to support various multi-station deep-learning methods,high-precision focal mechanism inversion,and seismic tomography studies.Additionally,owing to the high seismicity in the southern north-south seismic zone in China,this dataset has great potential for future scientific discoveries.

Fiber damage of machine-harvested cotton before ginning and after lint cleaning

Machine harvesting increases the foreign matter content of seed cotton. Excessive cleaning causes fiber damage and economic loss. Most trading companies in the Xinjiang Uygur Autonomous Region, China have indicated reluctance to use machine-harvested cotton. The first objective was to determine how the fiber quality was affected by the ginning and lint cleaning and how the fiber damage during levels of lint cleaning changed. The second objective was to determine the optimum number of lint cleaners for machine-harvested cotton based on fiber damage. Cotton samples were collected from 13 fields and processed in seven ginneries between 2013 and 2015. The results indicated that ginning and lint cleaning didn＇t have significant effect on fiber strength and significantly affected both fiber length and short fiber index. Fiber length was reduced by more than 1.00 mm from six of 13 fields after lint cleaning, then the damage rate on short fiber index from 11 of 13 fields was more than 20%. The third lint cleaning caused great fiber damage, reducing fiber length by 0.35 mm and increasing short fiber index by 0.65%. So, the lint should be cleaned by one lint cleaner in the Xinjiang, however, the stage of lint cleaning was sometimes omitted when the foreign matter content of lint was little.

Microstructure & performance of laser cladding on pick surface of coal mining machine

Laser cladding of 316 L steel powders on pick substrate of coal mining machine was conducted, and microstructure of laser cladding coating was analyzed. The micro-hardness of laser cladding coating was examined. The results show that microstructure of laser cladding zone is exiguous dentrite, and there are hard spots dispersible distribution in the laser cladding zone. Performances of erode-resistant, surface micro-hardness and wear-resistant are improved obviously.

FA006 Type Compound Cotton Holding Machine

TheFA006TypeCompoundCottonHoldingMachinemadebytheZhengzhouTextileMachineryFactoryinHenanProvinceishighlyautomatic,andcanholdsmallamountsofcottonevenlyandwithoutdamage.Theoutputreached1000kg/h,withtheindexreachingtheworldleveloftheearly1990s.FA006 Typ...

地膜覆盖和灌溉量对棉田土壤水盐分布特征的影响

为探究不同地膜覆盖和灌溉量对南疆棉田土壤水盐分布特征的影响,本研究于2021—2022年在新疆阿克苏地区沙雅县棉花地开展试验,以裸地(M0)为对照,设置高堡膜覆盖(M1)和普通PE地膜覆盖(M2)2种地膜处理,以及3150(W1)、4050(W2)、4950 m3·hm^(-2)(W3)3个灌溉量,分析了不同地膜覆盖和灌溉量对棉田0~60 cm土壤含水率、盐分、水盐分布均匀系数、盐分变化量和脱盐率的影响。结果表明:2021—2022年,随着灌溉量的增加,棉田土壤的含水量均呈上升趋势,盐分显著降低。棉田土壤水分的纵向运移能力均显著提高,膜间上层土壤盐分的淋溶效果显著,但水分分布均匀度和盐分分布均匀度均逐渐下降,盐分变化量和脱盐率随着灌溉量的增加而增大。M1处理0~60 cm土层平均土壤含水率较M0和M2处理分别高5.76%~14.03%和0.86%~2.22%,且M1处理可显著提高土壤水分分布的均匀度。随着时间的推进,各处理上层(0~30 cm)的土壤含盐量呈先降低后升高的变化趋势,下层(30~60 cm)的土壤含盐量逐渐升高;M0处理0~60 cm土层平均含盐量较M1和M2处理分别高24.22%~35.87%和18.88%~30.64%,M1和M2处理的盐分分布均匀度下降。在二者互作条件下,W3M1处理的土壤含水率最高,2021、2022年分别为26.48%、23.99%,水分的纵向运移能力显著提高,上层土壤盐分降低,W3M2(2021)、W3M1(2022)的盐分变化量和脱盐率最高,分别为0.33 g·kg^(-1)和38.66%、0.33 g·kg^(-1)和45.65%。综上所述,在高堡膜覆盖条件下,采用4950 m3·hm^(-2)的灌溉量时,可提高棉田土壤含水率,显著降低膜间土壤含盐量,有利于土壤排盐,降低土壤次生盐渍化,促进棉花根系的生长。

基于改进YOLOv8算法对被遮挡柑橘的识别与定位优化

针对果园环境中柑橘果实相互重叠和被枝叶遮挡,导致机器视觉识别柑橘果实与定位目标柑橘空间位置难度较大的问题,提出了一种基于YOLOv8-SAM的改进算法。通过增加BAM(Bottlenet Attention Module)注意力机制提高模型对被遮挡柑橘的识别准确率,运用SAM(Segment Anything Model)算法对被遮挡柑橘轮廓形状进行识别,并运用边缘检测法结合双目立体相机三维稠密深度点云得到被遮挡柑橘有效轮廓边,使用最小二乘法拟合出被遮挡柑橘的完整轮廓以确定目标柑橘果实更精确的空间坐标位置。试验结果表明:该算法可以准确识别并分离目标柑橘果实,同时更精确地定位柑橘果实空间坐标。改进的YOLOv8-SAM算法在果园环境中对被遮挡柑橘果实的识别平均精度达到91.1%,对被遮挡柑橘形心空间坐标的平均定位误差相比传统定位方法减少了16.22 mm,平均果径误差降低了7.99%,可为柑橘采摘机器人对重叠与被遮挡果实的准确识别提供参考。

Design and experiment of a picking robot for Agaricus bisporus based on machine vision

Harvesting represents the crucial stage in the cultivation process of Agaricus bisporus mushrooms.An important way for the production process of Agaricus bisporus to reduce costs and increase income is to ensure timely harvest of Agaricus bisporus,reduce harvesting costs,and improve harvesting efficiency.There are many disadvantages in manual picking,such as high labor intensity,time-consuming work and high cost.In this study,a set of mushroom picking platform including climbing mechanism,picking robot,and control system was designed and developed.The picking robot consisted of a truss mechanism,an image acquisition device,a mushroom collection device,and a picking actuator.The profile picking actuator could realize the function of constant force clamping.An online size detection algorithm for Agaricus bisporus based on deep image processing was proposed.The algorithm included removal of abnormal noise points,background segmentation,coordinate conversion,and diameter detection.The precision picking system for Agaricus bisporus with coordinate compensation function controlled by Industrial Personal Computer was designed,and the visual control interface was developed based on Labview.Through the performance test,the reliability of machine vision recognition and the overall operating stability of the picking platform were verified.The test results showed that in the process of machine vision recognition,the recognition accuracy rate was higher than 92.50%,the missed detection rate was lower than 4.95%,the false detection rate was lower than 2.15%,and the diameter measurement error was less than 4.50%.The image processing algorithm had high recognition rate and small diameter measurement error,which could meet the requirements of picking operation.The picking platform’s picking success rate was higher than 95.45%,the picking damage rate was lower than 3.57%,and the picking output rate was higher than 87.09%.Compared with manual picking,the recognition accuracy rate of the picking platform was increased by 6.70%,the picking output rate was increased by 1.51%.The overall performance of the picking platform was stable and practical.

Bin-Picking中无纹理工件的分割

为了解决Bin-Picking系统中无纹理工件的分割问题,提出了一种将颜色域和深度域信息融合的分割方法。该方法首先提取颜色域和深度域的边缘特征,然后基于二者的边缘特征,分别通过形态学预处理获取标记图,利用标记图辅助分水岭分割。融合二者的分割信息获得初分割结果,最后通过区域生长法以颜色域分割结果为边界优化分割区域。实验结果表明该算法适用于Bin-Picking视觉检测环节中无纹理目标的分割,对于复杂场景可获得较好的目标潜在位置区域,具有较高的鲁棒性和实时性。

基于YOLOv3算法的智能采茶机关键技术研究

在复杂背景下精确识别茶叶嫩芽,是实现高端茶叶智能化采摘的关键技术之一。为实现高端茶叶机械化精准采摘,设计一台基于视觉的采茶样机,根据蛛式机械手采摘茶叶的路径规划,将机械手末端的移动坐标问题转换成静平台3个电机转角问题。针对YOLOv3算法进行改进,采用EfficientNet网络替代DarkNet-53网络进行特征提取,并利用目标函数GIOU优化损失函数。试验结果表明:改进的YOLOv3算法在茶叶嫩芽识别方面,其准确率达到86.53%,单张图像平均识别时间为53 ms,相比传统的YOLOv3算法,性能实现明显的提升,可以达到预期目标,满足机器采摘需求。

基于VMD和GA-SVM的矿井地震自适应噪声压制方法

煤矿井下地震信号往往呈现出复杂的波场特性且伴随着大量噪音干扰,导致地震信号的初至拾取精度降低,从而影响地震数据的反演与解释。针对复杂干扰环境下采集的低信噪比地震信号,提出了基于变分模态分解(VMD)和遗传算法优化支持向量机(GA-SVM)的地震噪声压制与初至提取方法,以提高煤矿井下复杂噪声条件下的地震信号质量。采用变分模态分解对含噪地震信号进行自适应分解,得到数个的变分模态分量(IMF);对VMD分解得到的IMF分量进行特征提取,将提取所得的信号特征作为信号有效性判别的依据;利用遗传算法对支持向量机模型进行优化,得到最优的惩罚因子c与核函数参数g;利用优化后的支持向量机模型对IMF分量进行有效性判别并将有效分量重构成高信噪比信号;通过对人工加噪的地震信号应用噪声压制算法,煤矿井下常见的不同类型噪声被有效地压制,验证了算法的可行性;对矿井巷道实采的地震记录进行噪声压制处理,有效地压制了数据中的干扰噪声,极大程度地提高了地震记录的信噪比,使初至拾取得更加准确。结果表明,基于VMD和GA-SVM的地震噪声压制方法可以很好地提取含噪地震记录中的有效信号,提高初至拾取精度,在矿井复杂干扰条件下具有显著的应用潜力,对解决矿井复杂干扰条件下的地震勘探问题有重要意义。

基于改进实数遗传算法的桑叶采摘机结构参数优化

针对遗传算法在求解桑叶采摘机结构优化问题时容易陷入局部最优和求解精度低等问题,提出一种改进实数遗传算法。首先是给出一种基于序的组合适应度函数的轮盘赌选择算子,该算子在轮盘赌的基础上,通过一个自适应变化的参数在两种适应度函数中选择一个,再去计算适应度值;然后设计一种基于方向的改进启发式交叉算子,该算子既保留两个父代个体中较优个体对子代个体的影响,又增加种群中最优个体对子代个体的影响,提高交叉产生有潜力子代的可能性。接着将改进算法应用于摇杆式桑叶采摘机的优化参数设计中,通过与其他算法作仿真对比试验验证算法的优越性,获得采摘机最优参数组合:行走结构速度为24 mm/s、拨动结构角速度为1.2 rad/s、采摘结构速度为440 mm/s,并由运行结果可知整机性能与优化前相比提高13%。最后用优化得到的参数组合进行实地试验,结果显示桑叶采摘机性能提升10.9%,误差较小为2.1%。可见,所提改进实数遗传算法是优化采摘机参数的一种有效算法。

基于极端梯度提升算法的地震同相轴自动识别

在常规地震同相轴识别方法基础上,通过引入极端梯度提升算法(XGBoost)智能化策略,并结合地震数据相邻道相似性特征,发展一种基于极端梯度提升算法的地震同相轴自动识别技术方法。在编程实现方法的基础上,通过简单层状模型和复杂Marmousi模型模拟的记录进行测试,验证方法的正确性。对含噪音数据和实际资料中的同相轴进行识别测试,同时进行单道对比定量分析以及不同信噪比情况下算法预测结果精度对比。结果表明:新方法对含噪数据和实际资料均具有较好的适应性;在低信噪比(-6.98 dB)情况下,同相轴的查准率仍可超过90%。

基于YOLOv5的茶叶嫩芽图像识别算法研究

现有的机器采茶都需要人工辅助进行采茶,且存在老叶、嫩芽一刀切的情况,会损害一部分茶叶,只适用于低端茶叶的采摘。因此,需要研究出一种精准高效的茶叶嫩芽识别方法。针对茶叶嫩芽图像背景复杂的问题,在YOLOv5算法的基础上,对算法进行多角度的改进,实验结果表明,改进算法的模型的mAP提升了4.1%,Recall提升了4.0%,且改进方法减少了漏检情况的发生。

基于多路神经网络的黄花菜关节点检测算法研究

传统的目标检测算法仅能得到目标框,无法确定黄花菜的生长姿态。针对这一问题,在现有目标检测算法的基础上优化神经网络(Neural Network)结构,由检测框预测改为关节点的预测。首先,按照锚点匹配的方式确定黄花菜的生长方向及长度,统计黄花菜目标生长角度和长度,按照统计结果设置多个锚点,实际生长角度和长度与锚点进行比较,获得目标的相对长度和角度,并将其作为模型预测值进行训练;其次,在模型中加入热力图预测分支,对4个关节点进行预测;最后,利用目标长度和角度信息连接关节点得到黄花菜目标的生长姿态。设计针对线段拟合特点的评估模型方法,在计算精度的过程中引入部分亲和度字段。并据此改进非极大值抑制算法(Non-Maximum Suppression)。试验结果表明:采用热力图校准后的模型对采摘目标识别精度达91.02%,定位精度达99.8%。

基于Gazebo虚拟世界的草莓采摘机器人仿真研究

【目的】提高草莓采摘的自动化水平,解放劳动力并降低采摘成本。【方法】研究小组设计了一种基于曲柄连杆机构的三爪同步夹持式采摘机器人,分析了其主要部件Kinect深度相机、六轴机械臂以及三爪夹持式采摘机械手的结构设计,并在Gezebo虚拟世界中构建了采摘机器人的模型,生成了草莓采摘机器人的末端运动轨迹,完成了该草莓采摘机器人的前期开发试验和算法验证。【结果】该草莓采摘机器人可精准地夹持草莓,减小了采摘过程中对草莓表皮的损伤,同时机械臂末端运动半径最大可达770 mm,完全满足小型辅助采摘设备的基本需求。【结论】Gazebo能真实地还原草莓采摘的自然环境,充分验证了该草莓采摘机器人的可行性,同时还可以在虚拟环境中模拟不同的任务场景,可为六轴机械臂或其他多轴运动机械的运行轨迹设计提供准确的实验数据,并为机器人的实际应用提供指导。

芒果采收车机械臂运动特性分析及试验

为提高芒果采收车机械化采摘的适应性,该研究采用多体动力学理论方法对其机械臂的运动特性进行分析。首先,结合芒果生长和农艺特征及采摘机械臂几何构型与向量特性,运用D-H法建立了机械臂运动学模型以探索其运动特性,并进行末端执行器运动轨迹规划。轨迹规划结果表明末端执行器运动平稳,满足芒果采摘运动要求。进一步地,利用拉格朗日法构建了机械臂的动力学模型,对其进行了正逆动力学仿真验证,以深入了解机械臂的关节运动特性。动力学分析结果表明,在恒力矩工况下,伸缩装置和摆动装置的运动呈现一定的周期性,摆动装置、旋转装置角加速度和伸缩装置加速度均约在1.25、2.30、4.15 s达到阶段峰值;在仅做摆动周期运动的情况下,旋转装置、摆动装置所受驱动力矩均近似呈周期性变化,峰值分别为72.5与52 N·mm,且在一个运动周期内,均有两个极大值点。对机械臂结构进行仿真模态计算和模态试验,结果表明前六阶固有频率误差在5%以内,验证了芒果采收车机械臂多体动力学仿真建模的准确性。研究结果可为保证有效实现机械臂的采摘效果及提高其可靠性与稳定性提供依据。

棉花育苗钵体制钵机设计与实验

针对市场上缺乏适用于个户棉农的小型、实用、经济的棉花钵体制钵农具,对现有制钵机器进行改进设计。在分析制钵机工作原理基础上,结合棉花钵体的物理参数和制钵工序,确定了关键零部件的结构设计参数、整机的动力参数及传动机构的传动比,并利用UG的运动分析模块对制钵机进行运动仿真与分析,研究了冲压机构和钵盘运动机构的运动规律。最后,对制钵机的性能进行试验分析,结果表明:制钵机运动平稳,结构参数设计合理,效率高,制钵完好率达95.54%。研究结果可为棉花育苗钵体制钵机的进一步发展和研究提供参考。

基于多视角三维点云融合的采棉机器人视觉感知方法

针对传统采棉机器人因单一视角和二维图像信息带来的视觉感知局限问题,本文提出了一种多视角三维点云配准方法,以增强采棉机器人实时三维视觉感知能力。采用4台固定位姿的Realsense D435型深度相机,从不同视角获取棉花点云数据。通过AprilTags算法标定出深度相机RGB成像模块与Tag标签的相对位姿,并基于深度相机中RGB成像模块与立体成像模块坐标系间的转换关系,解算出各个相机间点云坐标的对应变换,进而实现点云间的融合配准。结果表明,本文配准方法的全局配准平均距离误差为0.93 cm,平均配准时间为0.025 s,表现出较高的配准精度和效率。同时,为满足采棉机器人感知的实时性要求,本文对算法中点云获取、背景滤波和融合配准等步骤进行了效率分析及优化,最终整体算法运行速度达到29.85 f/s,满足采棉机器人感知系统实时性需求。