【作者】 张帆；

【导师】 邹丽敏；

【作者基本信息】 哈尔滨工业大学 ， 仪器科学与技术， 2023， 硕士

【摘要】 自准直技术是精密小角度测量的重要手段,具有分辨力高、结构简单、使用便捷等优势,因而在航空航天、先进制造、大科学装置、大型结构件装配等领域得到广泛应用。随着上述领域的发展,自准直系统不仅需要具备精密小角度计量能力,还需要在测量环境温度发生变化时维持角度测量精度。然而现有自准直领域对环境温度变化条件下存在角度测量误差这一问题,有待进一步研究。然而现有自准直领域对环境温度变化条件下存在角度测量误差这一问题,有待进一步研究。本课题针对自准直角度测量易受环境温度影响的问题,深入研究环境温度变化对自准直系统参数的作用机理,通过建立自准直几何模型定量分析不同参数变化引起的角度测量误差;在此基础上提出一种基于XGBoost算法的自准直误差补偿方法,补偿环境温度变化引起的自准直系统角度测量误差。经仿真与实验验证,该方法能够有效降低环境温度变化产生的角度测量误差。本课题研究内容如下:1.研究环境温度对自准直系统误差的作用机理。针对现有模型无法准确分析环境温度对自准直系统影响的问题,建立基于光线追迹的自准直系统模型,研究环境温度变化对自准直系统参数的作用机理,定量计算环境温度引起的角度测量误差水平。分析结果表明,当环境温度变化范围(0～40)℃、角度测量范围±3°的条件下,自准直系统角度测量误差值最高为27.900″。2.提出基于回归树算法的自准直环境温度补偿方法。针对大量程自准直系统在环境温度变化范围(0～40)℃条件下缺少有效环境温度补偿方法的问题,使用XGBoost算法补偿自准直系统的角度测量结果。经仿真分析,当环境温度变化范围(0～40)℃、角度测量范围±1000″的条件下:使用传统的质心定位算法计算后角度测量误差最大为0.323″,当环境温度变化范围(0～40)℃、角度测量范围±3°的条件下:使用传统的质心定位算法计算后角度测量误差为2.890″,XGBoost算法补偿后,角度测量误差为0.013″。3.设计并搭建变温环境自准直系统误差补偿算法测试实验系统。实验结果表明,当环境温度变化范围(0～40)℃、角度测量范围±1000″的条件下:使用传统的质心定位算法处理后最大角度测量误差为0.82″;XGBoost算法补偿后,最大角度测量误差为0.14″。实验结果表明,XGBoost算法具有自准直系统误差补偿能力。本课题提出一种能够有效抑制环境温度变化对自准直系统影响的补偿方法,该方法能够有效降低环境温度变化引起的自准直系统的角度测量误差,提升了自准直系统抗环境温度扰动的能力。更多还原

【Abstract】 The autocollimation device is an ultra-precision small-angle measurement instrument based on the principle of optical autocollimation.Due to its advantages of high resolution,simple structure,and convenient use,it has been widely used in aerospace,advanced manufacturing,and other fields.With the development of these areas,there is a growing demand for autocollimation devices to not only possess small-angle measurement capabilities but also maintain angle measurement accuracy under high or low temperature conditions.However,existing autocollimation devices lack effective temperature compensation methods,resulting in significant angle measurement errors in extreme temperature environments,which cannot meet practical requirements.To address the problem of the susceptibility of autocollimation devices to environmental temperature,this paper deeply studies the mechanism of temperature variation’s impact on the parameters of the autocollimation system.It quantitatively analyzes the sources of angle measurement errors.Based on this analysis,a temperature compensation method using the e Xtreme Gradient Boosting algorithm is proposed.Through simulations and experiments,it is demonstrated that this method effectively reduces angle measurement errors caused by temperature variations.The research content of this paper is as follows:1.The influence of temperature on the autocollimation system has been investigated.A ray-tracing-based model of the autocollimation system was established to study the mechanism of temperature’s effect on the system parameters.The variation patterns of parameters under temperature influence were analyzed,and the angle measurement error caused by parameter changes was calculated using the autocollimation system model.The impact of temperature on the autocollimation system was quantitatively analyzed.2.A temperature compensation method for suppressing angle measurement errors in the autocollimation system has been developed.Based on the characteristics of temperature compensation in autocollimation system,the XGBoost algorithm has been selected as the compensation method.The compensation capability of the XGBoost algorithm for a wide range of temperature variations in large-scale autocollimation systems has been verified using simulated training samples generated from the autocollimation system model.3.The compensation effect of the autocollimation temperature compensation method has been experimentally validated.This study has proposed a compensation algorithm that effectively mitigates the influence of temperature variations on the autocollimation system.The algorithm significantly reduces angle measurement errors in autocollimation systems subjected to large temperature variations,thereby enhancing the precision small-angle measurement capability of the autocollimation system in extreme temperature environments.更多还原