multi-mode Tennessee Eastman process (MMTEP) dataset

该数据集名为MMTEP，与传统数据集不同，它包含了多种模态，使得数据更为复杂和多样化。它集成了来自压力、温度和流量传感器等多种传感器的数据流。此外，该数据集适用于多模态数据分析、多模态模式识别以及多模态融合等多个研究方向，并为过程监控、故障诊断和状态估计等应用提供了实验数据集。该数据集包含了41个测量变量、12个操纵变量、28个故障标识以及一个正常状态。其任务是在实时环境中进行多模态故障诊断及参数更新。The dataset named MMTEP differs from conventional datasets in that it incorporates multiple modalities, rendering the data more complex and diverse. It integrates data streams from various sensors including pressure, temperature, and flow sensors. Furthermore, this dataset finds applicability in multiple research directions such as multimodal data analysis, multimodal pattern recognition, and multimodal fusion, and provides experimental datasets for applications including process monitoring, fault diagnosis, and state estimation. This dataset comprises 41 measured variables, 12 manipulated variables, 28 fault labels, and one normal state. Its designated task is to perform multimodal fault diagnosis and parameter updating in a real-time environment.

搜集汇总

背景与挑战

背景概述

该数据集是基于田纳西-伊斯曼过程模型的多模式故障诊断数据集，包含六个不同模式，每个模式模拟了28种故障情况，并提供12个输入变量、41个测量变量和28个扰动变量。它专为算法比较和验证设计，支持直接下载使用或通过参数调整重新生成数据，适用于工业过程监控和故障诊断研究。