General information:This dataset is meant to serve as a benchmark problem for fault detection and isolation in dynamical systems. It contains pre-processed sensor data from the adaptive high-rise demonstrator building D1244, built in the scope of the CRC1244. Parts of the measurements have been artificially corrupted and labeled accordingly. Please note that although the measurements are stored in Matlab's .mat-format (Version 7.0), they can easily be processed using free software such as the SciPy library in Python.Structure of the dataset: train contains the training data (only nominal), test_easy contains test data (nominal and faulty with high fault amplitude). Faulty samples were obtained by manipulating a single signal in a random nominal sample from the test data. test_hard contains test data (nominal and faulty with low fault amplitude), meta contains textual labels for all signals and fault types. File contents: Each file contains the following data from 16384 timesteps: t: time in seconds u: demanded actuator forces in newtons y: measured outputs (relative elongations measured by strain gauges and actuator displacements in meters measured by position encoders) label: categorical label of the present fault class, where 0 denotes the nominal class and faults in the different signals are encoded according to their index in the list of fault types meta/labels.txt. Faulty samples additionally include the corresponding nominal values for reference u_true: delivered actuator forces y_true: measured outputs without faults. A sample's textual fault label is also contained in its filename (between the first and second underscore).
%0 Generic
%1 stiefelmaier2023fault
%A Stiefelmaier, Jonas
%D 2023
%K darus mult ubs_10007 ubs_10018 ubs_20011 ubs_20023 ubs_30112 ubs_30199 ubs_40170 unibibliografie
%R 10.18419/darus-3332
%T Data for Fault Diagnosis in Adaptive Buildings
%X General information:This dataset is meant to serve as a benchmark problem for fault detection and isolation in dynamical systems. It contains pre-processed sensor data from the adaptive high-rise demonstrator building D1244, built in the scope of the CRC1244. Parts of the measurements have been artificially corrupted and labeled accordingly. Please note that although the measurements are stored in Matlab's .mat-format (Version 7.0), they can easily be processed using free software such as the SciPy library in Python.Structure of the dataset: train contains the training data (only nominal), test_easy contains test data (nominal and faulty with high fault amplitude). Faulty samples were obtained by manipulating a single signal in a random nominal sample from the test data. test_hard contains test data (nominal and faulty with low fault amplitude), meta contains textual labels for all signals and fault types. File contents: Each file contains the following data from 16384 timesteps: t: time in seconds u: demanded actuator forces in newtons y: measured outputs (relative elongations measured by strain gauges and actuator displacements in meters measured by position encoders) label: categorical label of the present fault class, where 0 denotes the nominal class and faults in the different signals are encoded according to their index in the list of fault types meta/labels.txt. Faulty samples additionally include the corresponding nominal values for reference u_true: delivered actuator forces y_true: measured outputs without faults. A sample's textual fault label is also contained in its filename (between the first and second underscore).
@misc{stiefelmaier2023fault,
abstract = {General information:This dataset is meant to serve as a benchmark problem for fault detection and isolation in dynamical systems. It contains pre-processed sensor data from the adaptive high-rise demonstrator building D1244, built in the scope of the CRC1244. Parts of the measurements have been artificially corrupted and labeled accordingly. Please note that although the measurements are stored in Matlab's .mat-format (Version 7.0), they can easily be processed using free software such as the SciPy library in Python.Structure of the dataset: train contains the training data (only nominal), test_easy contains test data (nominal and faulty with high fault amplitude). Faulty samples were obtained by manipulating a single signal in a random nominal sample from the test data. test_hard contains test data (nominal and faulty with low fault amplitude), meta contains textual labels for all signals and fault types. File contents: Each file contains the following data from 16384 timesteps: t: time in seconds u: demanded actuator forces in newtons y: measured outputs (relative elongations measured by strain gauges and actuator displacements in meters measured by position encoders) label: categorical label of the present fault class, where 0 denotes the nominal class and faults in the different signals are encoded according to their index in the list of fault types meta/labels.txt. Faulty samples additionally include the corresponding nominal values for reference u_true: delivered actuator forces y_true: measured outputs without faults. A sample's textual fault label is also contained in its filename (between the first and second underscore). },
added-at = {2023-02-13T14:37:24.000+0100},
affiliation = {Stiefelmaier, Jonas/Universität Stuttgart},
author = {Stiefelmaier, Jonas},
biburl = {https://puma.ub.uni-stuttgart.de/bibtex/2b92fbc7e8400cf98b993558042d1da80/unibiblio},
doi = {10.18419/darus-3332},
howpublished = {Dataset},
interhash = {407c43c65ee880a5edd8fdba3ecd5d84},
intrahash = {b92fbc7e8400cf98b993558042d1da80},
keywords = {darus mult ubs_10007 ubs_10018 ubs_20011 ubs_20023 ubs_30112 ubs_30199 ubs_40170 unibibliografie},
orcid-numbers = {Stiefelmaier, Jonas/0000-0002-1854-9910},
timestamp = {2023-02-13T13:37:24.000+0100},
title = {Data for Fault Diagnosis in Adaptive Buildings},
year = 2023
}
