中文摘要：

基于融合隶属函数法、最大隶属度法、滚动均值法、算术平均值法和自助法,提出一种乏信息融合技术,以实现机床加工误差的调整。首先运用乏信息融合技术融合机床试加工时输出的小样本数据,获取机床试加工时工件的估计真值,以调整机床的加工误差,使工件满足质量要求;然后获取机床调整后输出的小样本可靠数据,运用模糊集合理论,在给定的置信水平下,预测可靠数据的估计区间。仿真和试验结果表明：运用乏信息融合技术,可实现对机床加工误差的调整。调整后的机床是可靠的,验证了运用乏信息融合技术调整机床的可行性。

英文摘要：

Based on fusing membership function method, maximum membership degree method, rolling mean method, arithmetic mean value method and bootstrap method, a poor information fusion technology was proposed to realize adjustment for machining errors of machine tool. The output small sample data in machine tool trial processing were fused using the poor information fusion technology to obtain estimated true value of the parts, and the machining errors of machine tool could be adjusted to make the machined parts meet quality requirements. Then the output small sample reliable data was obtained in machine tool adjusted well processing, and the estimated interval of the reliable data could be predicted using the fuzzy set theory under the given confidence levels. The results of simula- tion and experiments show that the poor information fusion technology can realize the error adjust- ment of machine tool, and the adjusted machine tool is reliable, and the feasibility of utilizing the in- formation fusion technology to adjust machine tool was verified.