拖拉机发动机耕作过程扭矩的预测模型研究  

Prediction Model Research on Tractor Engine Tillage Process Torque

作　　者：张捷[1] Zhang Jie(Hunan College of Information,Changsha 410200,China)

机构地区：[1]湖南信息职业技术学院,长沙410200

出　　处：《农机化研究》2024年第7期262-268,共7页Journal of Agricultural Mechanization Research

基　　金：湖南省教育厅科学研究项目(18C1596);长沙市哲学社会科学研究项目(2021csskkt46)。

摘　　要：拖拉机轮轴扭矩通常作为进行拖拉机变速器疲劳寿命分析、优化设计和加速寿命测试时做出各项决策的评价指标之一,而由于现有拖拉机轮轴扭矩测量方法基于扭矩传感器进行检测,投入成本较高,不适宜进行大面积推广应用。针对以上问题,建立了拖拉机在耕作过程中轮轴扭矩的预测模型,选取发动机扭矩、发动机转速、耕作深度、滑移率和行驶速度作为影响因素,对拖拉机耕地作业进行田间试验,并使用测量的校准数据集提出了8个轴扭矩预测回归模型,回归模型的调整决定系数R2为0.259~0.925。其中,预测模型E显示调整后的R2为0.925,不同模型下拖拉机轮轴扭矩预测模型MAPE均小于2.1%。研究结果表明,提出的预测模型均适用于实际的拖拉机作业过程中的轮轴扭矩预测。Tractor axle torque is usually used as one of the evaluation indexes for making various decisions in tractor transmission fatigue life analysis,optimal design and accelerated life testing.As the existing tractor axle torque measure-ment method is based on the torque sensor detection,the investment cost is high,which is not suitable for large area ap-plication.In response to the above problems,this study establishes the prediction model of tractor axle torque during till-age,selects engine torque,engine speed,tillage depth,slip rate and driving speed as the influencing factors,develops a load measurement system to conduct field tests on tractor tillage operations,and proposes eight axle torque prediction regression models using the measured calibration data sets,and the adjusted coefficients of determination R2 of the regres-sion models are 0.259-0.925.Among them,the prediction model E showed an adjusted R2 of 0.925,and the MAPE of tractor axle torque prediction models under different models were less than 2.1%.The results of the study showed that all the prediction models proposed in this study were applicable to the actual tractor operation process of axle torque predic-tion.

关 键 词：农用拖拉机 轴扭矩 预测模型 多重回归 耕作作业 

分 类 号：S219.031[农业科学—农业机械化工程]