| 摘要: |
| 目的 针对高温(温度大于120 ℃)条件下水泥浆稠化时间难以控制、水泥浆稠化实验周期长和水泥浆稠化时间难以预测等问题，研究了高温水泥浆稠化时间的影响因素。方法 采用对稠化时间进行拟合的方法对此进行研究，并且尝试基于灰色理论构建高温下水泥浆稠化时间的预测模型。结果 水泥浆稠化时间主要受缓凝剂加量、养护条件和特殊外加剂的影响，且在不同条件下各因素的影响程度不同；按照养护温度范围划分，并根据影响因素首次使用灰色理论建立了水泥浆稠化时间预测模型，该预测模型具有精度高、适用于小样本的特点。结论 高温水泥浆稠化时间影响因素敏感性分析与预测模型的建立，大大节约了水泥浆稠化实验的成本，并为高温水泥浆稠化时间的调配提供了理论基础。 |
| 关键词: 油井水泥 稠化时间 影响因素敏感性 预测模型 灰色模型 |
| DOI：10.3969/j.issn.1007-3426.2024.04.014 |
| 分类号: |
| 基金项目: |
|
| Influencing factors analysis and prediction model study of high-temperature cement slurry thickening time |
|
Zi CHEN1, Yuhao WEN2,3, Xiujian XIA3, Pu XU3, Huiting LIU3, Zhengrong ZHANG2,3
|
|
1.University of Paris-Saclay, Paris, France;2.China University of Petroleum (Beijing), Beijing, China;3.CNPC Engineering Technology R&D Company Limited, Beijing , China
|
| Abstract: |
| Objective This study addresses challenges associated with controlling the thickening time of cement slurries at elevated temperatures (exceeding 120 ℃), including prolonged experimental periods and difficulties in predicting thickening times. Methods This article explores the factors influencing the thickening time of cement slurries at high temperatures through fitting methods and constructs a predictive model develops a predictive model utilizing grey theory. Results The findings indicate that the thickening time of cement slurries is primarily influenced by the dosage of retarders, curing conditions, and specific additives, with varying degrees of impact under different conditions. The grey model used to predict the thickening time for the first time exhibits high accuracy and is suitable for small samples. Conclusion The sensitivity analysis of the factors affecting the thickening time at high temperatures and the establishment of a predictive model significantly reduce the cost of cement slurry thickening experiments and provide a theoretical basis for the formulation of thickening times in high-temperature cement slurries. |
| Key words: Oil well cement thickening time sensitivity of influencing factors prediction model grey model |
