以转炉钢渣作为固化稳定化药剂, 采用toxicity characteristic leaching procedure（TCLP）方法与傅里叶红外光谱跟踪检测钢渣微粉对重金属土壤修复效果, 以及混合物（钢渣微粉与重金属污染土壤）的微观结构。 建立基于高斯过程回归的钢渣微粉对重金属污染土壤修复效果软测量模型。 结果表明, 钢渣微粉对重金属污染土壤具有较好的修复效果, 180 d内修复效果均保持在90%以上; 修复过程分为前期、 中期与后期, 其中前期（1~3 d）环境碱性较高, 修复方式以离子交换为主, 中期（7~42 d）离子交换作用减弱和凝胶固化作用增强, 后期（56~180 d）形成大量C—S—H凝胶, 凝胶固化作用进一步增强; 基于高斯过程回归的钢渣微粉对重金属污染土壤修复效果软测量模型的真实值与预测值数据吻合较好, 绝对误差为-1.35~-0.48, 相对误差为-1.448%~-0.497%。

Using converter slag as curing stabilizing agents, applying Toxicity Characteristic Leaching Procedure (TCLP) and Fourier transform infrared spectroscopy method of tracking and detection of heavy metals in soil remediation effect of slag powder, and the micro-structure of mixtures (slag powder and heavy metal contaminated soils). It can establish soft sensor mode which is based on Gaussian process regression slag powder on heavy metal contaminated soil remediation effect. by using Gaussian process regression. The results show that the steel slag powder on heavy metals contaminated soil has good repairing effect, 180 d within its restorative effects are maintained over 90%; the repair process is divided into early, middle and late stages, in which the early (1~3 d) repair mode is given priority to with ion exchange high alkaline environment, medium-term (7~42 d) ion exchange and gel setting weaken the enhanced role of the late (56~180 d) form a large number CSH gel, the gel solidification further strengthened; based on Gaussian process regression Steel slag powder on heavy metal contaminated soil remediation effect soft measurement model of the real and predicted values agree well with the data, the absolute error is -1.35 ~-0.48, relative error of -1.448%～-0.497%.