混合KPLS與模糊邏輯的雙直接進給軸全行程熱誤差建模及補償
發(fā)布時間:2018-11-25 09:51
【摘要】:為了提高雙直線電動機驅(qū)動的同步直接進給軸的運動精度,對該類直接進給軸的全行程熱誤差在線補償方法進行了研究。分析了雙直接進給軸全行程熱誤差的影響因素,提出一種基于核偏最小二乘法(Kernel partial least squares,KPLS)和模糊邏輯相結(jié)合的雙直接進給軸全行程熱誤差的在線補償方法。應用激光干涉儀測量其熱變形量,使用熱電偶和紅外測溫儀測量進給機構關鍵點的溫度,以時間匹配溫度和變形量數(shù)據(jù)建立統(tǒng)計樣本,在均勻離散點位置建立熱誤差KPLS識別模型,通過在線計算得到離散點熱誤差補償量,再根據(jù)任意位置與離散點的模糊關聯(lián)程度,綜合計算全行程任意位置處熱誤差補償量。以此理論為基礎,建立補償決策函數(shù)和補償系統(tǒng),依據(jù)補償決策函數(shù)智能推斷補償值,通過向數(shù)控系統(tǒng)發(fā)送補償碼實現(xiàn)在線補償。在自構建的龍門雙直線電動機驅(qū)動的直接進給軸平臺上,進行全行程熱誤差在線補償試驗研究,結(jié)果表明:混合KPLS與模糊邏輯可以有效的對雙直接進給軸全行程熱誤差在線補償,經(jīng)過隨機測試驗證,補償后的進給精度提高了50%。
[Abstract]:In order to improve the motion accuracy of synchronous direct feed shaft driven by double linear motor, the on-line compensation method of full stroke thermal error of this kind of direct feed shaft is studied. The factors influencing the full stroke heat error of double direct feed shaft are analyzed, and an online compensation method based on kernel partial least square method (Kernel partial least squares,KPLS) and fuzzy logic is proposed to compensate the full stroke heat error of double direct feed shaft. The laser interferometer is used to measure the thermal deformation, the thermocouple and the infrared thermometer are used to measure the temperature of the key points of the feed mechanism, and the statistical samples are established by time matching temperature and deformation data. The thermal error KPLS identification model is established in the uniform discrete point position, and the thermal error compensation quantity of the discrete point is obtained by on-line calculation. According to the fuzzy correlation degree between the arbitrary position and the discrete point, the thermal error compensation amount at any position of the whole travel is calculated synthetically. On the basis of this theory, the compensation decision function and compensation system are established. According to the compensation decision function, the compensation value is inferred intelligently, and the online compensation is realized by sending the compensation code to the NC system. In this paper, the experimental research on the on-line compensation of the full stroke thermal error is carried out on the direct feed shaft platform driven by the self-constructed double linear motor of the gantry. The results show that the hybrid KPLS and fuzzy logic can effectively compensate the full stroke heat error of the double direct feed shaft on line. After random test, the feed accuracy of the compensation is improved by 50%.
【作者單位】: 上海理工大學機械工程學院;江蘇自動化研究所;
【基金】:國家自然科學基金(51005158) 國家科技重大專項(2013ZX04008-011-02)資助項目
【分類號】:TM359.4
,
本文編號:2355671
[Abstract]:In order to improve the motion accuracy of synchronous direct feed shaft driven by double linear motor, the on-line compensation method of full stroke thermal error of this kind of direct feed shaft is studied. The factors influencing the full stroke heat error of double direct feed shaft are analyzed, and an online compensation method based on kernel partial least square method (Kernel partial least squares,KPLS) and fuzzy logic is proposed to compensate the full stroke heat error of double direct feed shaft. The laser interferometer is used to measure the thermal deformation, the thermocouple and the infrared thermometer are used to measure the temperature of the key points of the feed mechanism, and the statistical samples are established by time matching temperature and deformation data. The thermal error KPLS identification model is established in the uniform discrete point position, and the thermal error compensation quantity of the discrete point is obtained by on-line calculation. According to the fuzzy correlation degree between the arbitrary position and the discrete point, the thermal error compensation amount at any position of the whole travel is calculated synthetically. On the basis of this theory, the compensation decision function and compensation system are established. According to the compensation decision function, the compensation value is inferred intelligently, and the online compensation is realized by sending the compensation code to the NC system. In this paper, the experimental research on the on-line compensation of the full stroke thermal error is carried out on the direct feed shaft platform driven by the self-constructed double linear motor of the gantry. The results show that the hybrid KPLS and fuzzy logic can effectively compensate the full stroke heat error of the double direct feed shaft on line. After random test, the feed accuracy of the compensation is improved by 50%.
【作者單位】: 上海理工大學機械工程學院;江蘇自動化研究所;
【基金】:國家自然科學基金(51005158) 國家科技重大專項(2013ZX04008-011-02)資助項目
【分類號】:TM359.4
,
本文編號:2355671
本文鏈接:http://sikaile.net/kejilunwen/dianlidianqilunwen/2355671.html
最近更新
教材專著