大中型拖拉机残值模型的研究

折旧是农业生产中的一项重要成本,而科学的折旧要以正确估计农业机器的残值为基础,ASABE的残值函数对估计农业机器残值提供了通用的方法,但这些模型是根据美国的实际情况模拟,并不能适合中国的实际情况.该文以上海农场近15年出售的116台拖拉机历史数据为观察样本,以机龄和年平均工作小时数为自变量,对6种不同的备选残值模型进行了回归分析和比较,在此基础上优选了双平方根模型为最佳残值模型,并用实际观察数据进行了验证.结果表明,用ASABE的模型来预测中国大中型拖拉机残值偏差很大,而本模型的预测精度是满意的,可用来预测中国大中型拖拉机残值;在残值模型变量中,机龄为决定残值的主要因素.     

基于Shapley值的农机装备水平组合预测(简报)

对农机装备水平的定量预测可以为农业机械化发展目标的制定提供依据.该文选用ARIMA时间序列和BP神经网络模型,再基于Shapley值法分配权重,构建了新的组合预测模型,并以1979～2005年山西省农机总动力、大中型拖拉机及配套农机具、小型拖拉机及配套农机具的统计数据为依据进行了颁测.预测结果表明,该组合预测模型的预测精度高于选定的各预测模型,对农机装备水半的预测是可行、有效的.以此模型预测山西省2010年农机总动力、大中型拖拉机、小型拖拉机、大中型拖拉机配套农机具、小型拖拉机配套农机具、大中拖拉机配套机具比、小型拖拉机配套农机具比将达到2619万kW、43479台、297546台、84638套、327743套、1.95、1.10.     

农场农业机器作业成本动态预测模型的研究

农业机器作业成本的预测是优化决策的关键,该文提出了建立农业机器作业成本动态预测模型的一般方法,模型由折旧费、维修保养费、油料费、劳动力成本和管理费5个预测子模型组成.模型结果表明,利用农场相关实测数据建立的模型均达到了较高的精度,其中拖拉机残值系数、累积维修保养费系数回归模型精度的调节平方和分别为0.8367和0.8840;模型的比较分析表明,在2007年机龄为9.16 a的6号JDT654拖拉机组成型耕机组和旋耕机组分别完成180.28 hm2和165.46 hm2作业面积的条件下,二台机组总作业成本的预测值分别为28585.79元和23868.42元,预测偏差分别-2.11%和-5.92%;二台机组作业总成本和的预测偏差为-3.88%.     

基于改进的GM(1,1)模型的辽西地区地下水预测研究

传统灰色动态预测模型GM(1,1)未能考虑各个变量之间的权重,在收敛计算时易出现无收敛解的缺陷,因此引入改进的GM(1,1)模型。该模型引入权重系数对每个变量赋予权重值,从而解决传统模型存在无收敛解的缺陷,并将改进的GM(1,1)模型用于辽西地区地下水预测。结果表明改进的GM(1,1)模型可解决传统模型无收敛解的局限,且在地下水预测中具有较好的精度,可用于辽西地区地下水的预测。研究为辽西地区地下水预测提供参考。     

采用改进长短时记忆神经网络的水产养殖溶解氧预测模型

为了精确预测水产养殖溶解氧变化趋势,该研究提出了基于K-means聚类和改进粒子群优化(Improved Particle Swarm Optimization,IPSO)的长短时记忆(Long Short-term Memory,LSTM)神经网络预测模型。根据环境因子间的相似度,应用改进的K-means聚类算法将环境数据划分为若干类。在此基础上,基于LSTM神经网络算法构建改进的水产养殖溶解氧预测模型,并引入改进粒子群优化算法对模型参数进行优化,以减少经验选取参数的盲目性。在不同天气状况下利用该模型对溶解氧进行预测。试验结果表明,在良好天气情况下,该模型预测误差曲线波动较小,预测精度更高。当天气发生突变时,溶解氧预测模型评价指标平均绝对百分误差、均方根误差、平均绝对误差和纳什系数分别为0.129 5、0.645 3、0.461 3和0.902 2。该模型一定程度改善了天气突变状况下的数据缺失、鲁棒性差等问题。     

基于经验模态分解与BP神经网络的农机总动力增长预测

为提高农机总动力增长变化预测结果的准确性和可靠性,根据农机总动力增长变化与其影响因素之间具有在各时间尺度明显的非线性波动特征,提出以1986—2013年农机总动力增长为研究对象,分别对农机总动力增长及其影响因素时间序列数据进行经验模态分解(empirical mode decomposition,EMD),对得到的各时间尺度下的波动分量分别建立BP神经网络预测模型。将EMD-BP网络预测结果与多元线性回归、支持向量机、BP神经网络进行对比分析,结果表明:基于EMD-BP网络建立的农机总动力增长预测模型,拟合和预测平均相对误差分别为0.99%和1.29%,相关决定系数约为0.999,均方根误差为316.35 MW,模型评价等级为"好",各项精度评价指标都优于其他方法,因此该预测模型精度高、可靠性强。研究成果为农业机械化发展规划的制定和出台相关政策提供有效参考。     

改进支持向量机的股票价格预测研究

为了提高股票价格预测精度,提出一种改进支持向量机的股票价格预测模型。该模型利用粒子群算法的全局寻优能力对支持向量机参数进行优化,以提高股票价格的预测精度,采用具体股票价格数据对模型性能进行测试。结果表明,改进支持向量机能够对股票价变化趋势进行预测,是一种有效、高精度的股票价格预测模型。     

改进YOLOv5s和迁移学习对番茄穴盘苗的分级检测

为了提高番茄穴盘苗分级检测精度,该研究提出了改进YOLOv5s目标检测模型,并通过迁移学习对番茄穴盘病苗识别精度进行优化。采用轻量级网络EfficientNetv2的Backbone部分作为特征提取网络,保留YOLOv5s中的SPPF空间金字塔池化模块,压缩模型参数数量以减少计算量;更改模型Neck部分原始上采样模块为CARAFE轻量级上采样模块,在引入很少参数量的情况下提高模型精度;同时将PANet替换为BiFPN,引入特征权重信息,增强不同尺度特征融合能力;引入有效多尺度注意力机制(efficient multi-scale attention,EMA),提高对番茄苗的关注,减少背景干扰;替换CIoU损失函数为SIoU损失函数,考虑真实框与预测框之间的方向匹配,提高模型收敛效果。试验结果表明,改进的YOLOv5s目标检测模型经过迁移学习训练后,平均精度均值达到95.6%,较迁移学习前提高了0.7个百分点;与原YOLOv5s模型相比,改进YOLOv5s模型平均精度均值提升2.6个百分点;改进YOLOv5s模型的参数量、计算量和权重大小分别为原YOLOv5s模型的53.1%、20.0%...     

基于自适应进化相关向量机的耕地面积预测模型

为解决耕地面积预测模型建立过程中的非线性、稀疏化及结果可靠性评价等问题,该文将相关向量机与差分进化优化算法进行融合及改进,提出基于自适应进化相关向量机的耕地面积预测模型。该文以黄石市为例,建立基于自适应进化相关向量机的短期、中期耕地预测模型,并与多元线性回归、BP神经网络、支持向量机算法在精度、计算效率及可靠性方面进行对比分析。试验验数据表明,自适应进化相关向量机的预测精度大约是其余3种方法的2倍以上;模型的计算效率是多元线性回归方法的2倍,比BP神经网络、支持向量机高出2个数量级;测试数据的实际耕地面积均在自适应进化相关向量机估计的95%置信度的置信区间内,并且由后验差比、小误差概率判定模型等级属于"好"。基于以上数据,证实该模型是一种精度高、计算快、可靠性强的耕地预测新方法。     

土壤重力侵蚀灰色系统模型研究

介绍了黄土丘陵沟壑区土壤重力侵蚀灰色系统预测模型的构造原理和方法，并用灰色关联度的方法来反映模型的预测与土壤重力侵蚀观测值之间的关联性，用以分析模型的预测精度，同时为反映土壤重力侵蚀系统的动态变化规律，引入了残差辨识的理论和方法，以提高模型的预报水平，用模型对晋西离石王家沟流域的土壤重力侵蚀进行了试验研究，结果表明模型的预测精度较高，为土壤得力侵蚀研究提供了一条定量化分析的新途径。     

基于实际作业工况的拖拉机能效评价

为深入了解拖拉机实际作业的能效情况,该研究提出基于实际作业工况的拖拉机能效评价方法,对拖拉机整体能效进行全面度量。基于距离最短分区原理,通过K-means聚类与成对比较矩阵方法进行拖拉机发动机常用工况点提取;通过对186台162 kW的拖拉机能效分析,提出拖拉机能效等级划分标准,确定能效限值和各级能效比限值,并对不同作业环节的平均能效进行分析。研究结果显示:拖拉机发动机实际8工况点和ISO稳态8工况点区别较大;不同拖拉机能效差异较大,50%的拖拉机能效值分布在3.20～3.65 (kW·h)/kg区间;162 kW拖拉机实际能效合格限值为3.07 (kW·h)/kg;拖拉机不同作业环节的平均能效值差异大,旋耕作业的平均能效值最高,行走模式最低。研究结果可为发动机节能减排技术升级提供数据基础,也可为农机节能考核、与绿色化作业水平挂钩的应用补贴方法研究等提供参考。     

Tractor overturn concerns in Iowa: perspectives from the Keokuk county rural health study

Agriculture remains one of the most hazardous industries in the U.S., with tractor overturns producing the greatest number of agricultural machinery-related fatalities. Rollover protective structures (ROPS) and seatbelts effectively reduce tractor overturn deaths. However, a large proportion of tractors in use in American agriculture are older tractors without ROPS and seatbelts. This article describes the tractor-related responses from participants in a population-based study conducted in Keokuk County, Iowa. This study was designed to measure rural and agricultural adverse health and injury outcomes and their respective risk factors. Questionnaires were partially developed from well-documented national surveys. Questions about agricultural machinery use, presence of safety equipment on the machinery, work practices, and attitudes about farm safety were included. Study participants on farms who owned tractors had an average of 3.1 tractors with an average age of 27 years. Only 39% of the 665 tractors had ROPS. Tractor age was associated with the presence of ROPS; 84% of tractors manufactured after 1984 were ROPS-equipped, whereas only 3% of tractors manufactured before 1960 were ROPS-equipped. ROPS-equipped tractors were significantly more common on larger farms and households with higher income. Only 4% of the farmers reported that their tractors had seatbelts and they wore them when operating their tractors. The results of this study support the findings of other studies, which indicate that many older tractors without ROPS and seatbelts remain in use in American agriculture. Until a dramatic reduction in the number of tractors in the U.S. operated without ROPS and seatbelts is achieved, the annual incidence of 120 to 130 deaths associated with tractor overturns will persist.     

A summary of state laws regulating youth operating farm tractors on highways

Driving farm tractors on highways is hazardous for youth due to the large speed differential between motor vehicles and agricultural vehicles, as well as recent increases in traffic volume on many rural roads. In 1994, the state of Wisconsin enacted legislation requiring youth who operate farm tractors on highways to complete a tractor and machinery certification course. We were interested in finding out whether other states have similar requirements. The purpose of this project was to collect and summarize state laws regulating youth who operate farm tractors on highways in the U.S. A systematic method was used that included an initial Lexis-Nexis database search followed by internet searches in combination with follow-up email and telephone communications when missing or unclear results were obtained. The findings show that 14 states have legislation addressing youth who operate farm tractors on highways. The content of these statutes varies, but includes driver's license or educational requirements, as well as regulations concerning the ages, locations, and/or times of day when youth may drive farm machinery on highways. This compilation of state laws will be useful information for agricultural safety professionals in designing effective outreach programs. A synthesis of the findings may also lead to the development of model legislation or inform future research efforts aimed at preventing youth farm tractor crashes on highways.     

Protecting young workers in agriculture: participation in tractor certification training

Tractor-related injuries among youth are an important public health problem. The major objectives of this study were to (1) provide a rigorous estimate of the number of youth operating tractors in Ohio and (2) assess the extent to which these youth are participating in federally mandated tractor safety training. A self-administered questionnaire was completed by approximately 4,000 students ages 14 or 15 years who were enrolled in a stratified cluster sample of 99 Ohio schools. Almost one-third of the youth (30%) reported having operated tractors, with 19% having operated a tractor on a farm not owned by a parent or guardian. More than half of the youth (52%) reported having operated a tractor and/or other hazardous machinery, with 25% having operated the machinery on a farm not owned or operated by a parent or guardian. Extrapolating from 4-H records and Ohio census data, fewer than 1% of the youth who are operating tractors or other hazardous machinery have participated in tractor certification training. Increasing participation in mandated training may be an important step in protecting the health of our nation's young agricultural workers.     

基于DEM的无人驾驶拖拉机前馈稳速控制

针对农田坡度变化影响无人驾驶农机行驶速度稳定性,进而降低播种均匀性和肥药施用精度等问题,该研究设计了一种基于农田数字高程模型（digital elevation model,DEM）和前馈控制策略的拖拉机稳速控制方法。首先建立坡地干扰补偿模型,基于拖拉机实时位置从农田DEM中提取前方作业路径的坡度信息,计算拖拉机前方目标速度补偿量,实现拖拉机行驶的稳速控制。以DF1204无级变速拖拉机为试验平台,在中国农业大学烟台研究院开展3组不同目标速度的上坡、平地和下坡行驶对比试验。试验结果表明,拖拉机以目标速度4.0、6.0和8.0 km/h行驶时,上坡行驶的实测速度均值分别为4.03、5.94和7.85 km/h,平地行驶的实测速度均值分别为4.04、6.02和8.03 km/h,下坡行驶的实测速度均值分别为4.00、6.10和8.19 km/h,与对照组相比,在上坡、平地和下坡行驶时的速度均方根误差分别降低了46.63%、21.92%和37.15%,试验组上坡和下坡行驶的实测速度均值更接近目标速度。所提方法可有效提高无人驾驶拖拉机在起伏农田的稳速控制精度,有助于提高农机作业质量。     

移动式拖拉机性能检测线的研制

拖拉机不仅是田间作业动力,也是我国重要的农村运输工具。长期以来拖拉机运输性能的检测缺乏合适的检测设备。文中介绍了一种YTC型移动式拖拉机性能检测线。该设备移动方便,操作简单。制动力矩精度达0.98N·m,轴重精度达0.1kg,车速精度达0.01km/h,每3min内完成一台车检测,一天可连续检测150台。实际应用表明,该新型检测线可作为农机管理部门开展拖拉机年检年审工作的检测设备     

Assessing farm tractor incidents and awareness levels of operators for tractor safety issues in the Hatay province of Turkey

Studies and statistical data on safety issues related to farm tractors and machinery are very limited in developing countries, including Turkey. This study was carried out to investigate tractor-related incidents in the Hatay province, located in the mid-south of Turkey. A questionnaire was conducted with 107 tractor operators using face-to-face interviews. Data were evaluated according to the incident type, machinery involved, and mechanism of injury or fatality. A total of 101 incidents were reported by 77 of the 107 respondents. Most of the incidents were due to tractor rollovers (65.4%), 14.8% of the incidents were due to entanglement of body parts in moving machinery, and 12.9% involved crashing into other vehicles or obstacles. The leading cause of the incidents was personal mistakes (60.4%). Fatalities resulted from 25.7% of the incidents, while 45.5% of the incidents caused non-fatal injuries. Only 5.6% of the tractors had a ROPS-enclosed cab. The percentage of ROPS-equipped tractors was 19.6%, while 41.3% of the tractors had a shade cover and 33.6% had no protective structure. Only one of the respondents used a seatbelt, although 44.9% of them stated that seatbelts should be used. It was also found that only 13.5% of the operators had training in work safety, while 95.1% stated that incidents might be reduced if people were trained. Development of appropriate policies and training programs are needed for safer operation of agricultural machinery to reduce injuries and fatalities due to farm accidents.