Data-driven Health Monitoring Research Articles

PHIR: A Platform Solution of Data-Driven Health Monitoring for Industrial Robots

The large-scale application of industrial robots has created a demand for more intelligent and efficient health monitoring, which is more efficiently met by data-driven methods due to the surge in data and the advancement of computing technology. However, applying deep learning methods to industrial robots presents critical challenges such as data collection, application packaging, and the need for customized algorithms. To overcome these difficulties, this paper introduces a Platform of data-driven Health monitoring for IRs (PHIR) that provides a universal framework for manufacturers to utilize deep-learning-based approaches with minimal coding. Real-time data from multiple IRs and sensors is collected through a cloud-edge system and undergoes unified pre-processing to facilitate model training with a large volume of data. To enable code-free development, containerization technology is used to convert algorithms into operators, and users are provided with a process orchestration interface. Furthermore, algorithm research both for sudden fault and long-term aging failure detection is conducted and applied to the platform for industrial robot health monitoring experiments, by which the superiority of the proposed platform, in reality, is proven through positive results.

A Semi-Supervised Matrixized Graph Embedding Machine for Roller Bearing Fault Diagnosis Under Few-Labeled Samples

Exploring historical measurement data-driven health monitoring schemes for roller bearings is a current research hotspot. In engineering practice, the type of fault data obtained is often unknown and requires expensive costs to be annotated. However, most current intelligent diagnostic methods are based on the assumption that the labeled fault data is sufficient, so as to effectively establish the nonlinear mapping relationship between monitoring signals and health status. For this issue, a newly intelligent diagnosis method based on semi-supervised matrixized graph embedding machine (SMGEM) is proposed. In SMGEM, the geometric similarity relationship of unlabeled and labeled samples is obtained, which is subsequently embedded by incorporating a manifold regularization into SMGEM model, so that SMGEM can use the structure information of unlabeled samples to assist modeling. Meanwhile, a weighted nuclear norm (WNN) is used to highlight the importance of large singular values, so that a more accurate weight matrix can be constructed. The proposed method is verified by several roller bearing fault datasets, and experimental results demonstrate that the proposed semi-supervised diagnosis method can use few labeled samples to obtain a better identification accuracy.

Milk yield residuals and their link with the metabolic status of dairy cows in the transition period

The transition period is one of the most challenging periods in the lactation cycle of high-yielding dairy cows. It is commonly known to be associated with diminished animal welfare and economic performance of dairy farms. The development of data-driven health monitoring tools based on on-farm available milk yield development has shown potential in identifying health-perturbing events. As proof of principle, we explored the association of these milk yield residuals with the metabolic status of cows during the transition period. Over 2 yr, 117 transition periods from 99 multiparous Holstein-Friesian cows were monitored intensively. Pre- and postpartum dry matter intake was measured and blood samples were taken at regular intervals to determine β-hydroxybutyrate, nonesterified fatty acids (NEFA), insulin, glucose, fructosamine, and IGF1 concentrations. The expected milk yield in the current transition period was predicted with 2 previously developed models (nextMILK and SLMYP) using low-frequency test-day (TD) data and high-frequency milk meter (MM) data from the animal's previous lactation, respectively. The expected milk yield was subtracted from the actual production to calculate the milk yield residuals in the transition period (MRT) for both TD and MM data, yielding MRTTD and MRTMM. When the MRT is negative, the realized milk yield is lower than the predicted milk yield, in contrast, when positive, the realized milk yield exceeded the predicted milk yield. First, blood plasma analytes, dry matter intake, and MRT were compared between clinically diseased and nonclinically diseased transitions. MRTTD and MRTMM, postpartum dry matter intake and IGF1 were significantly lower for clinically diseased versus nonclinically diseased transitions, whereas β-hydroxybutyrate and NEFA concentrations were significantly higher. Next, linear models were used to link the MRTTD and MRTMM of the nonclinically diseased cows with the dry matter intake measurements and blood plasma analytes. After variable selection, a final model was constructed for MRTTD and MRTMM, resulting in an adjusted R2 of 0.47 and 0.73, respectively. While both final models were not identical the retained variables were similar and yielded comparable importance and direction. In summary, the most informative variables in these linear models were the dry matter intake postpartum and the lactation number. Moreover, in both models, lower and thus also more negative MRT were linked with lower dry matter intake and increasing lactation number. In the case of an increasing dry matter intake, MRTTD was positively associated with NEFA concentrations. Furthermore, IGF1, glucose, and insulin explained a significant part of the MRT. Results of the present study suggest that milk yield residuals at the start of a new lactation are indicative of the health and metabolic status of transitioning dairy cows in support of the development of a health monitoring tool. Future field studies including a higher number of cows from multiple herds are needed to validate these findings.

A Data-Driven Health Monitoring Method Using Multiobjective Optimization and Stacked Autoencoder Based Health Indicator

This article proposes a new data-driven health monitoring method, which uses multiobjective optimization and stacked autoencoder based health indicator. Specifically, the proposed method proposes an improved nondominated sorting genetic algorithm-II (NSGA-II) to perform multiobjective optimization on a large number of candidate features extracted from the sensor measurements. Then, a stacked autoencoder model is used to construct health indicators from the selected features. In the improved NSGA-II algorithm, the optimization goals of feature selection are defined as the minimum gap of health indicators between different states and the number of features. Comparisons between the proposed method and the state-of-the-art methods on simulation experiments show that the proposed method can accurately identify the status of the equipment and effectively limit the complexity of the diagnostic model.

A Health Monitoring Method for Wind Power Generators with Hidden Markov and Probabilistic Principal Components Analysis Models

In this work, we propose a data-driven health monitoring method for wind power generators, which learns an empirical model from the time-series sensor data and detects irregularities or faults in the turbines and blades. Our main objective is to predict any symptoms of faults as early as possible before the generators fall into malfunction. The data obtained from the wind power generators are strongly correlated multidimensional time-series with multiple states. In this study, we take such features into account and develop the probabilistic model for them, namely, hidden Markov and probabilistic principal component analysis. Once the model is learned with the data that contain no faulty events, it can be used to detect faults in new data by comparing the original sensor values and reconstructed ones. In this research, we apply this method to synthetic data and real-world wind turbine data and show the results of experiments to confirm the availability of the proposed method.

A Data-Driven Health Monitoring Method for Satellite Housekeeping Data Based on Probabilistic Clustering and Dimensionality Reduction

In the operation of artificial satellites, it is very important to monitor the health status of the systems and detect any symptoms of anomalies in the housekeeping data as soon as possible. Recently, the data-driven approach to the system monitoring problem, in which statistical machine learning techniques are applied to the large amount of measurement data collected in the past, has attracted considerable attention. In this paper, we propose a new data-driven health monitoring and anomaly detection method for artificial satellites based on probabilistic dimensionality reduction and clustering, taking into consideration the miscellaneous characteristics of the spacecraft housekeeping data. We applied our method to the telemetry data of the small demonstration satellite 4 (SDS-4) of the Japan Aerospace Exploration Agency (JAXA) and evaluated its effectiveness. The results show that the proposed system provides satellite operators with valuable information for understanding the health status of the system and inferring the causes of anomalies.

Sparse and low-rank methods in structural system identification and monitoring

This paper presents sparse and low-rank methods for explicit modeling and harnessing the data structure to address the inverse problems in structural dynamics, identification, and data-driven health monitoring. In particular, it is shown that the structural dynamic features and damage information, intrinsic within the structural vibration response measurement data, possesses sparse and low-rank structure, which can be effectively modeled and processed by emerging mathematical tools such as sparse representation (SR), and low-rank matrix decomposition. It is also discussed that explicitly modeling and harnessing the sparse and low-rank data structure could benefit future work in developing data-driven approaches towards rapid, unsupervised, and effective system identification, damage detection, as well as massive SHM data sensing and management.

Modeling and harnessing sparse and low-rank data structure: a new paradigm for structural dynamics, identification, damage detection, and health monitoring

This paper presents a new paradigm of explicitly modeling and harnessing the data structure to address the inverse problems in structural dynamics, identification, and data-driven health monitoring. In particular, it is shown that the structural dynamic features and damage information, intrinsic within the structural vibration response measurement data, possesses sparse and low-rank structure, which can be effectively modeled and processed by emerging mathematical tools such as sparse representation and compressed sensing, low-rank matrix decomposition and completion, as well as the unsupervised multivariate blind source separation. It is also discussed that explicitly modeling and harnessing the sparse and low-rank data structure could benefit future work in developing data-driven approaches toward rapid, unsupervised, and effective system identification, damage detection, as well as massive SHM data sensing and management. Copyright © 2016 John Wiley & Sons, Ltd.