一种基于知识图谱的液面波动卷渣预测方法
By combining data-driven and knowledge-driven risk assessment with a knowledge graph-based approach, the problems of excessive model simplification and poor adaptability in the prediction of liquid surface fluctuation and slag entrainment were solved, achieving high-precision and interpretable prediction results and improving prediction accuracy and adaptability.
CN122173901BActive Publication Date: 2026-07-17BENGANG STEEL PLATES CO LTD
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- BENGANG STEEL PLATES CO LTD
- Filing Date
- 2026-05-12
- Publication Date
- 2026-07-17
Smart Images
Figure CN122173901B_ABST
Abstract
本发明提供一种基于知识图谱的液面波动卷渣预测方法，属于钢铁连铸过程缺陷诊断技术领域。基于工艺规程和专家经验构建知识图谱及规则；基于知识图谱确定知识图谱推理风险；以当前工况特征向量与知识图谱中所有历史卷渣案例节点特征向量的余弦相似度为依据确定案例匹配风险；通过时序卷积网络‑长短期记忆网络混合模型基于时序特征输出液位数据标准差的预测值；基于液位数据的标准差结合当前时刻拉速以及当前时刻温度的预测值输出数据驱动风险；对数据驱动风险、知识图谱推理风险以及案例匹配风险进行加权融合获得综合风险值；当综合风险值大于预设安全阈值，则判断有卷渣风险；当综合风险值小于等于预设安全阈值，则判断无卷渣风险。
Need to check novelty before this filing date? Find Prior Art