中文摘要：

针对棉针织物前处理过程皂洗难以实现在线连续检测的关键技术问题，以前处理过程中温度、时间、液碱质量分数、双氧水质量分数为变量，白度和吸水性为指标，在分析了棉针织物前处理过程质量指标相关影响因素的基础上，采用最小二乘支持向量机建立了前处理过程工艺技术指标软测量模型，并通过实验验证了软测量模型的有效性。

英文摘要：

Contrapose to the key technical problem of difficult to realize on-line continuous measurement for soap washing of cotton knitted fabrics pretreatment process. The correlation impacting factors of quality index for cotton knitted fabrics pretreatment process are analysised with water absorbance and whiteness as quality index, and temperature, time, liquid caustic soda concentration, hydrogen peroxide concentration etc as variables. According to the relations between the technical indices and such boundary variables, a soft measurement model of technical indices based on Least Square Support Vector Machine （LS-SVM） was proposed innovatively to estimate water absorbance and whiteness. Simulation experiment result proves the soft measurement model of technical indices is effective.