A Low-Cost Multi-Sensor Data Acquisition System for Fault Detection in Fused Deposition Modelling

Emerging Sensor Communication Network-Based AI/ML Driven Intelligent IoT

At present, the field of the Internet of Things (IoT) is one of the fastest-growing areas in terms of Artificial Intelligence (AI) and Machine Learning (ML) techniques [...].

EmerSense: A low-cost multiparameter logger to monitor occurrence and duration of emersion events within intertidal zones

The intertidal zone is a harsh environment for marine life as conditions are often both extreme and variable. A wide range of sessile organisms are partially or fully emersed (exposed above the water line) during low tide. In the tropics, when corals are emersed, high light and temperature can be detrimental to their survival. To date, there is no commercially-available logger that can detect periods of emersion, information that is useful for marine research and for coastal resource management. Here, we present a low-cost Arduino-based multi-parameter logger called 'EmerSense' which can detect instances of emersion while simultaneously recording light and temperature profiles. We describe the different steps involved in fabricating EmerSense, including hardware construction and software design, and discuss the results of our field testing at an intertidal coral reef in Singapore.

Fault Detection for High-Speed Trains Using CCA and Just-in-Time Learning

Online monitors of the running gears systems of high-speed trains play critical roles in ensuring operational safety and reliability. Status signals collected from high-speed train running gears are very complex regarding working environments, random noises and many other real-world constraints. This paper proposed fault detection (FD) models using canonical correlation analysis (CCA) and just-in-time learning (JITL) to process scalar signals of high-speed train gears, named as CCA-JITL. After data preprocessing and normalization, CCA transforms covariance matrices of high-dimension historical data into low-dimension subspaces and maximizes correlations between the most important latent dimensions. Then, JITL components formulate local FD models which utilize subsets of testing samples with larger Euclidean distances to training data. A case study introduced a novel system design of an online FD architecture and demonstrated that CCA-JITL FD models significantly outperformed traditional CCA models. The approach is applicable to other dimension reduction FD models such as PCA and PLS.

Wavelet Time-Frequency Analysis on Bridge Resonance in Train-Track-Bridge Interactive System

With the continuous improvement in the operation speed of trains, the impact of train–induced vibration through the track on the bridge is increasingly prominent. In particular, when the loading frequency is the same as or close to the natural frequency of the bridge, the resonant response of the bridge will be activated, which will probably endanger the safety of the operation and the bridge structure. Normally, the traditional method to indicate the appearance of resonant response is to analyze the frequency spectrum of the response through the Fourier transform from its time history. However, it can simply reflect the contribution of different frequency components within a stationary window. Therefore, continuous wavelet transform is adopted on a 2D train–track–bridge interactive system in this article. It illustrates the evolutionary characteristics of different frequencies from the input excitation to the output response during the bridge resonance in the time–frequency domain, compared with the cases when the bridge is nonresonant. Finally, the article demonstrates the feasibility of the method. It concludes that the resonance and quasi–resonance–triggering band accounts for the highly intensified bridge response, while the staggering domination between the steady-state and the transient response is the main phenomenon for the nonresonant bridge. Additionally, within the low–frequency band, the resonant bridge will have a more significant impact on the track subsystem than the train subsystem.

A Few Shot Classification Methods Based on Multiscale Relational Networks

Learning information from a single or a few samples is called few-shot learning. This learning method will solve deep learning’s dependence on a large sample. Deep learning achieves few-shot learning through meta-learning: “how to learn by using previous experience”. Therefore, this paper considers how the deep learning method uses meta-learning to learn and generalize from a small sample size in image classification. The main contents are as follows. Practicing learning in a wide range of tasks enables deep learning methods to use previous empirical knowledge. However, this method is subject to the quality of feature extraction and the selection of measurement methods supports set and the target set. Therefore, this paper designs a multi-scale relational network (MSRN) aiming at the above problems. The experimental results show that the simple design of the MSRN can achieve higher performance. Furthermore, it improves the accuracy of the datasets within fewer samples and alleviates the overfitting situation. However, to ensure that uniform measurement applies to all tasks, the few-shot classification based on metric learning must ensure the task set’s homologous distribution.

Multi-Sensor GA-BP Algorithm Based Gearbox Fault Diagnosis

To address the problem of the low recognition rate of time-frequency domain methods gearbox fault identification, a method featuring decision-level fusion of DS evidence theory and GA-BP algorithm was proposed in the present study. Firstly, the fault data of each state of the gearbox was classified, based on which the time-frequency domain features were extracted and 19 significant features have been selected. Secondly, the accuracy of the traditional BP algorithm was compared with that of the GA-BP algorithm. On this basis, it has been concluded that the GA-BP algorithm is highly accurate, and the local diagnostic results obtained by the GA-BP algorithm have been used as the basic probability. Finally, the DS evidence theory is currently used to fuses with the GA. In addition, the final fault identification of the gearbox can be achieved by using the DS evidence theory and the multi-sensor local diagnosis results obtained by the GA-BP algorithm for decision fusion. The results of the simulations and experiments showed that the method proposed has improved accuracy over a single algorithm for fault identification of gearboxes, respectively.