Anomaly Detection through Transfer Learning in Agriculture and Manufacturing IoT Systems

Project Title

Anomaly Detection through Transfer Learning in Agriculture and Manufacturing IoT Systems

Project Area of Specialization Internet of ThingsProject Summary

IoT systems have been facing increasingly sophisticated technical problems due to the growing complexity of these systems and their fast deployment practices. Consequently, IoT managers have to judiciously detect failures (anomalies) in order to reduce their cyber risk and operational cost. While there is rich literature on anomaly detection in many IoT-based systems, there is no existing work that documents the use of ML models for anomaly detection in digital agriculture and in smart manufacturing systems. These two application domains pose certain salient technical challenges. In agriculture, the data is often sparse, due to the vast areas of farms and the requirement to keep the cost of monitoring low. Second, in both domains, there are multiple types of sensors with varying capabilities and costs. The sensor data characteristics change with the operating point of the environment or machines, such as the RPM of the motor. The inferencing and the anomaly detection processes, therefore, have to be calibrated for the operating point. In this project,, we will analyze data from sensors deployed in an agricultural farm with data from seven different kinds of sensors, and from an advanced manufacturing testbed with vibration sensors. We evaluate the performance of ARIMA and LSTM models for pre-dicting the time series of sensor data. Then, considering the sparse data from one kind of sensor, we perform transfer learning from a high data rate sensor. We then perform anomaly detection using the predicted sensor data. Taken together, we show how in these two application domains, predictive failure classification can be achieved, thus paving the way for predictive maintenance.

Project Objectives

anomaly detection and failure classification for the predictive maintenance problem of IoT-based digital agriculture and smart manufacturing

Project Implementation Method

In offline training, the sensor with large amount of data (let us call it sensor type I) has its data entered to the feature extraction module that performs encoding and normalization of the input signals into numerical features. Second, a deep neural network (DNN) model is trained and tuned using these features and labels of the data (where the sample label is normal, near-failure or failure). We use the DNN as a multi-class classifier due to its discriminative power that is leveraged in different classification applications 

Moreover, DNN is useful for both tasks of learning the level of defect for the same sensor type and for transfer learning across the different sensor types that we consider here. In online mode, any new sensor data under test (here, sensor type II) would have the same feature extraction process where the saved feature encoders are shared. Then, the classifier predicts the defect type (one of the three states mentioned earlier) given the trained model and gives as output the probability of each of the three classes.

Benefits of the Project

fast indication of the transfer learning model and increase the accuracy of failure detection in agriculture

Technical Details of Final Deliverable

anomaly detection and failure classification for the predictive maintenance problem of IoT-based digital agriculture and smart manufacturing. We designed a temporal anomaly detection technique and an efficient defect-type classification technique for these two application domains. We compared the strategies of LSTM and semi-supervised ARIMA for anomaly detection. We observed that LSTM leads to better anomaly detection prediction at the cost of longer training time. We tested our findings on two real-world data-sets. We also studied the effects of several tuning parameters to enhance the failure classification We proposed a transfer learning model for classifying failure on sensors with lower sampling rates (MEMs) using learning from sensors with huge data (piezoelectric). Our findings indicate that the transfer learning model can considerably increase the accuracy of failure detection. Finally, we will use data-augmentation techniques to enhance the prediction of the failure mode. Using such augmentation, the accuracy is enhanced with the enhancement becoming more pronounced in near-failure mode.

Final Deliverable of the Project HW/SW integrated systemCore Industry ITOther IndustriesCore Technology Internet of Things (IoT)Other TechnologiesSustainable Development Goals No PovertyRequired Resources

Item Name	Type	No. of Units	Per Unit Cost (in Rs)	Total (in Rs)
			Total in (Rs)	80000
Iot Camera 8 MP	Equipment	1	12000	12000
Raspberry pi 4	Equipment	1	22000	22000
MEMS-EVAL-BOARD	Equipment	1	36000	36000
Supporting Material	Miscellaneous	1	10000	10000