R&D
❯
Automated QA
❯Anomaly detection in sensor data using deep learning techniques
Predictive testing
Scope:
Automatic detection of inaccurate test outcomes in an application development
process
Goal:
Other
Automated Quality Assurance
For:
Quality Engineers
Predictive data quality
Scope:
A solution for assessing data quality in data collection systems.
Goal:
Other
Anomaly detection in sensor data using deep learning techniques
For:
Maintenance and support functions, monitoring, procurement
Scope:
Temporal data captured from sensors.
Goal:
Other
Anomaly detection in sensor data using deep learning techniques
For:
Maintenance and support functions, monitoring, procurementGoal:
OtherProblem addressed
Identify anomalies and events by learning the temporal patterns of sensor data,
based on deep learning techniques.
Scope of use case
Temporal data captured from sensors.
Description
Mechanical devices such as engines, vehicles, aircraft, etc.,
are typically instrumented with numerous sensors to
capture the behaviour and health of the machine. However,
there are often external factors or variables that are not
captured by sensors, leading to time-series which are
inherently unpredictable. For instance, manual controls
and/or unmonitored environmental conditions or load may
lead to inherently unpredictable time-series. Detecting
anomalies/events in such scenarios becomes challenging
using standard approaches based on mathematical models
that rely on stationarity, or prediction models that utilize
prediction errors to detect anomalies.
LSTM-AD:
Our work started with a stacked LSTM network that was
trained on non-anomalous data and used as a predictor over
a number of time steps. The resulting prediction errors are
modelled as a multivariate Gaussian distribution, which is
used to assess the likelihood of anomalous behaviour. The
efficacy of this approach was demonstrated on four datasets:
ECG, space shuttle, power demand, and multi-sensor engine
dataset.
EncDec-AD:
As an extension to the prior work, we proposed a long short
term memory networks based encoder-decoder scheme for
anomaly detection (EncDec-AD) that learns to reconstruct
normal time-series behaviour, and thereafter uses
reconstruction errors to detect anomalies. We experimented
with three publicly available quasi predictable time-series
datasets (power demand, space shuttle, and ECG) and two
real-world engine datasets with both predictive and
unpredictable behaviour. We had shown that EncDec-AD is
robust and can detect anomalies from predictable,
unpredictable, periodic, aperiodic, and quasi-periodic time-
series. Further, we showed that EncDec-AD is able to detect
anomalies from short time-series (length as small as 30) as
well as long time-series (length as large as 500).
Online-AD:
The common approach of training one model in an offline
manner using historical data is likely to fail under
dynamically changing and non-stationary environments
where the definition of normal behaviour changes over time,
making the model irrelevant and ineffective. We described a
temporal model based on recurrent neural networks (RNNs)
for time series anomaly detection to address challenges
posed by sudden or regular changes in normal behaviour.
The model is trained incrementally as new data becomes
available, and is capable of adapting to the changes in the
data distribution. RNN is used to make multi-step
predictions of the time series, and the prediction errors are
used to update the RNN model as well as detect anomalies
and change points. Large prediction error is used to indicate
anomalous behaviour or a change (drift) in normal
behaviour. Further, the prediction errors are also used to
update the RNN model in such a way that short term
anomalies or outliers do not lead to a drastic change in the
model parameters, whereas high prediction errors over a
period of time lead to significant updates in the model
parameters, such that the model rapidly adapts to the new
norm. We demonstrate the efficacy of the proposed approach
on a diverse set of synthetic, publicly available and
proprietary real-world datasets.