Simulated boiler data for fault detection and classification

Citation Author(s):
R.
Shohet
Ryerson University
M.
Kandil
Ryerson University
J.J.
McArthur
Ryerson University
Submitted by:
J.J. McArthur
Last updated:
Sat, 03/30/2019 - 14:58
DOI:
10.21227/awav-bn36
Data Format:
License:
Creative Commons Attribution
1159 Views
Citations:
0
Categories:
Keywords:
5
2 ratings - Please login to submit your rating.

Abstract 

An emulator for the Viessmann Vitorond 200 Gas Fired Boiler VD2 Series 380 dataset was created in Matlab/Simulink based on the Simscape boiler model. After validating with manufacturer test data, a series of faults (Excess air, ranging from 15% - 50%; fouling of the heat exchanger from 1% to 46%; and scaling of the water-side heat exchanger element from 1% to 46%, all in 5% increments) were simulated with this emulator, along with nominal operating conditons for a range of operational parameters (gas fuel rate from 1 kg/s to 4 kg/s, water mass flow rate from 3 kg/s to 12.5 kg/s and combustion air temperature from 283 K to 303 K).. A factorial sampling method was used, resulting in a total of 27,281 simulations. This dataset contains the points visible to a Building Automation System along with the Condition (detailed class) and Class (fault category).

Instructions: 

Either Condition or Class should be used as data labels and the unused label should be excluded for analysis.

Comments

Thank you

Submitted by Yongkwi Lee on Thu, 11/25/2021 - 00:55

thanks

Submitted by M. S. on Fri, 12/03/2021 - 17:57