Tennessee Eastman simulation dataset

Citation Author(s):
Xiaolu
Chen
Submitted by:
xiaolu chen
Last updated:
Sun, 06/09/2019 - 21:55
DOI:
10.21227/4519-z502
License:
0
0 ratings - Please login to submit your rating.

Abstract 

Tenessee Eastman (TE) process simulates actual chemical processes and is widely used as a benchmark in test fault diagnosis and process control. The overall process consists of five operating units: reactor, condenser, vapor-liquid separator, recycle compressor and product stripper. It has standard training and test data sets for fault detection and diagnosis, classification, etc. Each data set is under different operating conditions。

Instructions: 

Based on the actual chemical reaction process, Eastman Chemical Company of the United States developed the open and challenging chemical model simulation platform - Tennessee Eastman (TE) simulation platform, which produces data with time-varying, strong coupling and nonlinear characteristics, widely used in Test control and fault diagnosis models for complex industrial processes. The entire TE data set consists of a training set and a test set. The data in the TE set consists of 22 different simulation run data, and each sample in the TE set has 52 observed variables. d00.dat to d21.dat are training set samples, and d00_te.dat to d21_te.dat are test set samples.

Comments

Easy to use, this is a great standard dataset.

Submitted by xiaolu chen on Sun, 06/09/2019 - 21:57

Dear Xiaolu Chen,

I am Morteza Soleimani, a Ph.D. student from the University of Bradford, UK.
I am interested in Tennessee Eastman chemical process data. Could you please share this data with me? Thanks
m.soleimani@bradford.ac.uk

Regards
Morteza

Submitted by Morteza Soleimani on Mon, 03/09/2020 - 17:18

Dear Xiaolu Chen,

I am Mostafa Ammar, a Ph.D. student from the University of Cairo, Egypt.I am interested in Tennessee Eastman chemical process data. Could you please share this data with me?
mostaphaammar@hotmail.com
Thanks in advance
Best Regards
Mostafa Ammar

Submitted by mostafa Ammar on Fri, 11/26/2021 - 04:12

nice

Submitted by zhang zy on Sat, 02/25/2023 - 11:28

Hi,

Could you please supply the PFD with stream labels and match these to the columns in the data sets?

Thanks,

Joanne

Submitted by Joanne Tanner on Wed, 04/29/2020 - 22:07

Dear Xiaolu Chen,

I am Fabian Hartung, a Ph.D. student from the Technische Universität in Kaiserslautern (Germany) and doing researches in Time-series anomaly detection for chemical processes. I wanted to ask if there is an option, to get access to the dataset, could you share it with me, please?
My mail is: fhartung@rhrk.uni-kl.de

Thanks in advance
Fabian

Submitted by Fabian Hartung on Tue, 08/02/2022 - 04:28

Dear Xiaolu Chen,

I am Zhang Zhiyan, a Ph.D. student from the University of Science and Technology Beijing and doing researches in Time-series for chemical processes. I wanted to ask if there is an option, to get access to the dataset, could you share it with me, please?
My mail is:b20200569@xs.ustb.edu.cn

Thanks in advance
Fabian

Submitted by zhang zy on Sat, 02/25/2023 - 11:31

Dear Xiaolu Chen,

I am Mohammad Qalaji, a master student from the University of Freiburg (Germany) and doing researches in Time-series failure detection and Root Causes Analysis. I wanted to ask if there is an option, to get access to the dataset, could you share it with me, please?
My mail is: mohammad.qalaji@gmail.com

Thanks in advance
Mohammad

Submitted by Mohammad Qalaji on Fri, 07/07/2023 - 12:33

Dear Xiaolu Chen,

I am Hamza Gorgulu, a master student from the Koc University (Turkey) and doing researches in predictive process monitoring. I wanted to ask if there is an option, to get access to the dataset, could you share it with me, please?
My mail is: hamzagorgulu7@gmail.com

Thanks in advance
Hamza

Submitted by Hamza Gorgulu on Wed, 01/10/2024 - 06:54