Fault Diagnosis of Hydraulic Systems Based on Deep Learning Model With Multirate Data Samples

Fault detection and tolerant control with a variable-bandwidth extended state observer for electro-hydraulic servo systems with lumped disturbance and measurement noise

In this study, a novel fault detection and tolerant control method are presented for electro-hydraulic servo systems (EHSS) with actuator partial loss of effectiveness (LOE) fault. Meanwhile, the parameter uncertainties, external disturbances, and output measurement noise are also considered. Firstly, we construct a new mapping function with a simple and flexible structure, and then design a variable-bandwidth extended state observer (VBESO) based on this function to estimate the states and lumped disturbance of the system. The designed VBESO can not only balance estimation accuracy and noise sensitivity, but also decouple the fault component from the lumped disturbance, thus facilitating fault detection. Subsequently, we design an adaptive law with time-varying gain that can accurately approximate the actuator LOE coefficient, and propose a stable fault detection scheme by combining the designed adaptive law, with the detection results that are less affected by changes in system states. The mutual coordination of the mentioned decoupling, adaptive law, and time-varying gain is a key feature of our method, rarely seen in other works. Furthermore, a fault-tolerant controller with the proposed VBESO and adaptive law is developed, which can guarantee that, even in the presence of the considered adverse obstacles, the hydraulic cylinder maintains high tracking accuracy and the control performance can be promptly recovered after an actuator fault occurs. Finally, the proposed method is validated by simulations and experiments, and it is compared with other controllers in different scenarios.

Condition monitoring and multi-fault classification of hydraulic systems using multivariate functional data analysis

Condition monitoring and fault classification in engineering systems is a critical challenge within the scope of Prognostics and Health Management (PHM). The fault diagnosis of complex nonlinear systems, such as hydraulic systems, has become increasingly important due to advancements in big data analytics, machine learning (ML), Industry 4.0, and Internet of Things (IoT) applications. Multi-sensor data provides opportunities to predict component conditions; however, environments characterized by multiple sensors and diverse fault states across various components complicate the fault classification process. To address these challenges, this study introduces a novel multivariate Functional Data Analysis (FDA) framework based on Multivariate Functional Principal Component Analysis (MFPCA) for classifying failure conditions in hydraulic systems. The proposed method systematically tackles condition-based diagnostics and addresses fundamental issues in multi-fault classification. Experimental results demonstrate that this approach achieves high classification accuracy using raw multi-sensor data, establishing multivariate FDA as a powerful tool for fault diagnosis in complex systems.

Bayesian Hierarchical Graph Neural Networks With Uncertainty Feedback for Trustworthy Fault Diagnosis of Industrial Processes

Deep learning (DL) methods have been widely applied to intelligent fault diagnosis of industrial processes and achieved state-of-the-art performance. However, fault diagnosis with point estimate may provide untrustworthy decisions. Recently, Bayesian inference shows to be a promising approach to trustworthy fault diagnosis by quantifying the uncertainty of the decisions with a DL model. The uncertainty information is not involved in the training process, which does not help the learning of highly uncertain samples and has little effect on improving the fault diagnosis performance. To address this challenge, we propose a Bayesian hierarchical graph neural network (BHGNN) with an uncertainty feedback mechanism, which formulates a trustworthy fault diagnosis on the Bayesian DL (BDL) framework. Specifically, BHGNN captures the epistemic uncertainty and aleatoric uncertainty via a variational dropout approach and utilizes the uncertainty information of each sample to adjust the strength of the temporal consistency (TC) constraint for robust feature learning. Meanwhile, the BHGNN method models the process data as a hierarchical graph (HG) by leveraging the interaction-aware module and physical topology knowledge of the industrial process, which integrates data with domain knowledge to learn fault representation. Moreover, the experiments on a three-phase flow facility (TFF) and secure water treatment (SWaT) show superior and competitive performance in fault diagnosis and verify the trustworthiness of the proposed method.

Intelligent Fault Diagnosis of Hydraulic System Based on Multiscale One-Dimensional Convolutional Neural Networks with Multiattention Mechanism

Hydraulic systems are critical components of mechanical equipment, and effective fault diagnosis is essential for minimizing maintenance costs and enhancing system reliability. In practical applications, data from hydraulic systems are collected with varying sampling frequencies, coupled with complex interdependencies within the data, which poses challenges for existing fault diagnosis algorithms. To solve the above problems, this paper proposes an intelligent fault diagnosis of a hydraulic system based on a multiscale one-dimensional convolution neural network with a multiattention mechanism (MA-MS1DCNN). The proposed method first extracts features from multirate data samples using a parallel 1DCNN with different receptive fields. Next, a Hybrid Attention Module (HAM) is proposed, consisting of two submodules: the Correlation Attention Module (CAM) and the Importance Attention Module (IAM), which aim to meticulously and comprehensively model the complex relationships between channel features. Subsequently, to effectively utilize the feature information of different frequencies, the HAM is integrated into the 1DCNN to form the MA-MS1DCNN. Finally, the proposed method is evaluated and experimentally compared using the UCI hydraulic system dataset. The results demonstrate that, compared to existing methods such as Shapelet, MCIFM, and CNNs, the proposed method shows superior diagnostic performance.

Intraoperative Hypotension Prediction Based on Features Automatically Generated Within an Interpretable Deep Learning Model

The monitoring of arterial blood pressure (ABP) in anesthetized patients is crucial for preventing hypotension, which can lead to adverse clinical outcomes. Several efforts have been devoted to develop artificial intelligence-based hypotension prediction indices. However, the use of such indices is limited because they may not provide a compelling interpretation of the association between the predictors and hypotension. Herein, an interpretable deep learning model is developed that forecasts hypotension occurrence 10 min before a given 90-s ABP record. Internal and external validations of the model performance show the area under the receiver operating characteristic curves of 0.9145 and 0.9035, respectively. Furthermore, the hypotension prediction mechanism can be physiologically interpreted using the predictors automatically generated from the proposed model for representing ABP trends. Finally, the applicability of a deep learning model with high accuracy is demonstrated, thus providing an interpretation of the association between ABP trends and hypotension in clinical practice.

Hydraulic system fault diagnosis decoupling method based on 2D time-series modeling and self-attention fusion

Hydraulic systems play a pivotal and extensive role in mechanics and energy. However, the performance of intelligent fault diagnosis models for multiple components is often hindered by the complexity, variability, strong hermeticity, intricate structures, and fault concealment in real-world conditions. This study proposes a new approach for hydraulic fault diagnosis that leverages 2D temporal modeling and attention mechanisms for decoupling compound faults and extracting features from multisample rate sensor data. Initially, to address the issue of oversampling in some high-frequency sensors within the dataset, variable frequency data sampling is employed during the data preprocessing stage to resample redundant data. Subsequently, two-dimensional convolution simultaneously captures both the instantaneous and long-term features of the sensor signals for the coupling signals of hydraulic system sensors. Lastly, to address the challenge of feature fusion with multisample rate sensor data, where direct merging of features through maximum or average pooling might dilute crucial information, a feature fusion and decoupling method based on a probabilistic sparse self-attention mechanism is designed, avoiding the issue of long-tail distribution in multisample rate sensor data. Experimental validation showed that the proposed model can effectively utilize samples to achieve accurate fault decoupling and classification for different components, achieving a diagnostic accuracy exceeding 97% and demonstrating robust performance in hydraulic system fault diagnosis under noise conditions.

A multi-rate high-order dynamic twin-latent-variable probabilistic modeling and its process monitoring application

Quality-relevant process monitoring provides important guarantees for the safety of industrial operation, which is based on the assumption that data collection is complete and low-order autocorrelated. However, real industrial processes always exhibit complex characteristics such as multi-rate sampling and high-order dynamic, which pose great challenges for process monitoring. To this end, a multi-rate high-order dynamic twin-latent-variable probabilistic (MHDTVP) model is presented in this paper to extract data correlations among multi-rate measurements from quality-relevant and irrelevant perspectives. Moreover, to reveal the dynamics in the multi-rate sampling process, an autoregressive twin-latent-variable structure is designed to extract both quality-relevant and quality-irrelevant high-order dynamic features. In the MHDTVP model, parameters are trained through an efficient expectation maximization (EM) iteration framework. Finally, the performance conclusions of MHDTVP are validated with the Tennessee Eastman process (TEP) and Thermal Power Plant (TPP). The experimental results demonstrate that the proposed model exhibits superior monitoring efficiency for multi-rate dynamic processes compared to similar approaches.

VAE-Based Interpretable Latent Variable Model for Process Monitoring

Latent variable-based process monitoring (PM) models have been generously developed by shallow learning approaches, such as multivariate statistical analysis and kernel techniques. Owing to their explicit projection objectives, the extracted latent variables are usually meaningful and easily interpretable in mathematical terms. Recently, deep learning (DL) has been introduced to PM and has exhibited excellent performance because of its powerful presentation capability. However, its complex nonlinearity prevents it from being interpreted as human-friendly. It is a mystery how to design a proper network structure to achieve satisfactory PM performance for DL-based latent variable models (LVMs). In this article, a variational autoencoder-based interpretable LVM (VAE-ILVM) is developed for PM. Based on Taylor expansions, two propositions are proposed to guide the design of appropriate activation functions for VAE-ILVM, allowing nondisappearing fault impact terms contained in the generated monitoring metrics (MMs). During threshold learning, the sequence of counting that test statistics exceed the threshold is considered a martingale, a representative of weakly dependent stochastic processes. A de la Peña inequality is then adopted to learn a suitable threshold. Finally, two chemical examples verify the effectiveness of the proposed method. The use of de la Peña inequality significantly reduces the minimum required sample size for modeling.

Physics-Constraint Variational Neural Network for Wear State Assessment of External Gear Pump

Most current data-driven prognosis approaches suffer from their uncontrollable and unexplainable properties. To address this issue, this article proposes a physics-constraint variational neural network (PCVNN) for wear state assessment of the external gear pump. First, a response model of the pressure pulsation of the gear pump is constructed via a spectral method, and a compound neural network is utilized to extract features from the pressure pulsation signal. Then, the response model is formulated into an objective function to softly constrain the learning process of the neural network, forcing the learned features to have explicit physics meaning. Meanwhile, to characterize the system uncertainty, the variational inference is utilized to extend a Kullback-Leibler (KL) divergence into the objective function. Finally, the wear state is evaluated based on the distance of learned physics features. Experimental results on an external gear pump validate the merits of the proposed method in explainable representation learning and system uncertainty estimation. It also offers a controllable and explainable perspective to understand the dynamic behavior of the system.

Integrated Optimal Control for Electrolyte Temperature With Temporal Causal Network and Reinforcement Learning

The electrowinning process is a critical operation in nonferrous hydrometallurgy and consumes large quantities of power consumption. Current efficiency is an important process index related to power consumption, and it is vital to operate the electrolyte temperature close to the optimum point to ensure high current efficiency. However, the optimal control of electrolyte temperature faces the following challenges. First, the temporal causal relationship between process variables and current efficiency makes it difficult to estimate the current efficiency accurately and set the optimal electrolyte temperature. Second, the substantial fluctuation of influencing variables of electrolyte temperature leads to difficulty in maintaining the electrolyte temperature close to the optimum point. Third, due to the complex mechanism, building a dynamic electrowinning process model is intractable. Hence, it is a problem of index optimal control in the multivariable fluctuation scenario without process modeling. To get around this issue, an integrated optimal control method based on temporal causal network and reinforcement learning (RL) is proposed. First, the working conditions are divided and the temporal causal network is used to estimate current efficiency accurately to solve the optimal electrolyte temperature under multiple working conditions. Then, an RL controller is established under each working condition, and the optimal electrolyte temperature is placed into the controller's reward function to assist in control strategy learning. An experiment case study of the zinc electrowinning process is provided to verify the effectiveness of the proposed method and to show that it can stabilize the electrolyte temperature within the optimal range without modeling.

Fault Diagnosis of Hydraulic Components Based on Multi-Sensor Information Fusion Using Improved TSO-CNN-BiLSTM

In order to realize the accurate and reliable fault diagnosis of hydraulic systems, a diagnostic model based on improved tuna swarm optimization (ITSO), optimized convolutional neural networks (CNNs), and bi-directional long short-term memory (BiLSTM) networks is proposed. Firstly, sensor selection is implemented using the random forest algorithm to select useful signals from six kinds of physical or virtual sensors including pressure, temperature, flow rate, vibration, motor power, and motor efficiency coefficient. After that, fused features are extracted by CNN, and then, BiLSTM is applied to learn the forward and backward information contained in the data. The ITSO algorithm is adopted to adaptively optimize the learning rate, regularization coefficient, and node number to obtain the optimal CNN-BiLSTM network. Improved Chebyshev chaotic mapping and the nonlinear reduction strategy are adopted to improve population initialization and individual position updating, further promoting the optimization effect of TSO. The experimental results show that the proposed method can automatically extract fusion features and effectively utilize multi-sensor information. The diagnostic accuracies of the plunger pump, cooler, throttle valve, and accumulator are 99.07%, 99.4%, 98.81%, and 98.51%, respectively. The diagnostic results of noisy data with 0 dB, 5 dB, and 10 dB signal-to-noise ratios (SNRs) show that the ITSO-CNN-BiLSTM model has good robustness to noise interference.

Real-Time Fault Diagnosis for Hydraulic System Based on Multi-Sensor Convolutional Neural Network

This paper proposed a real-time fault diagnostic method for hydraulic systems using data collected from multiple sensors. The method is based on a proposed multi-sensor convolutional neural network (MS-CNN) that incorporates feature extraction, sensor selection, and fault diagnosis into an end-to-end model. Both the sensor selection process and fault diagnosis process are based on abstract fault-related features learned by a CNN deep learning model. Therefore, compared with the traditional sensor-and-feature selection method, the proposed MS-CNN can find the sensor channels containing higher-level fault-related features, which provides two advantages for diagnosis. First, the sensor selection can reduce the redundant information and improve the diagnostic performance of the model. Secondly, the reduced number of sensors simplifies the model, reducing communication burden and computational complexity. These two advantages make the MS-CNN suitable for real-time hydraulic system fault diagnosis, in which the multi-sensor feature extraction and the computation speed are both significant. The proposed MS-CNN approach is evaluated experimentally on an electric-hydraulic subsea control system test rig and an open-source dataset. The proposed method shows obvious superiority in terms of both diagnosis accuracy and computational speed when compared with traditional CNN models and other state-of-the-art multi-sensor diagnostic methods.

A Transfer-Based Convolutional Neural Network Model with Multi-Signal Fusion and Hyperparameter Optimization for Pump Fault Diagnosis

Pumps are one of the core components of drilling equipment, and their fault diagnosis is of great significance. The data-driven approach has made remarkable achievements in the field of pump fault diagnosis; however, most of them are easily affected by complex background conditions and usually suffer from data scarcity problems in real-industrial scenarios, which limit their application in practical engineering. To overcome the above shortcoming, a novel framework for a model named Hyperparameter Optimization Multiple-Signal Fusion Transfer Convolution Neural Network is proposed in this paper. A convolutional neural network model based on transfer learning is built to promote well-learned knowledge transfer over different background conditions, improve robustness, and generalize the model to cross-domain diagnosis tasks. The multi-signal fusion strategy is involved in capturing system state information for establishing the mapping relationship between the raw signal and fault pattern by integrating the multi-physical signal with the weight allocation protocol. The hyperparameter optimization method is explored in conjunction with the transfer-based model by integrating Grid Search with the Gradient Descent algorithm for further improvement of diagnosis performance. Results show that the proposed model can effectively realize the fault diagnosis of pumps under different background conditions, achieving 95% accuracy.

Incipient Fault Detection in a Hydraulic System Using Canonical Variable Analysis Combined with Adaptive Kernel Density Estimation

Incipient fault detection in a hydraulic system is a challenge in the condition monitoring community. Existing research mainly monitors abnormal working conditions in hydraulic systems by separately detecting the key working parameter, which often causes a high miss warning rate for incipient faults due to the oversight of parameter dependence. A principal component analysis provides an effective method for incipient fault detection by taking the correlation of multiple parameters into consideration, but this technique assumes the systems are Gaussian-distributed, making it invalid for a dynamic non-Gaussian system. In this paper, we combine a canonical variable analysis (CVA) and adaptive kernel density estimation (AKDE) for the early fault detection of nonlinear dynamic hydraulic systems. The collected hydraulic system data set was used to construct the typical variable space, and the state space and residual space are divided to represent the characteristics of different correlations between the two variables, which are quantitatively described using Hotelling's T2 and Q. In order to investigate the proper upper control limits, AKDE was utilised to estimate the underlying probability density functions of T2 and Q by taking the nonlinearity of the hydraulic system variables into consideration. The advantages of the proposed approach for incipient fault detection are illustrated via a marine power plant lubrication system.

Fault Classification for Cooling System of Hydraulic Machinery Using AI

Hydraulic systems are used in all kinds of industries. Mills, manufacturing, robotics, and Ports require the use of Hydraulic Equipment. Many industries prefer to use hydraulic systems due to their numerous advantages over electrical and mechanical systems. Hence, the growth in demand for hydraulic systems has been increasing over time. Due to its vast variety of applications, the faults in hydraulic systems can cause a breakdown. Using Artificial-Intelligence (AI)-based approaches, faults can be classified and predicted to avoid downtime and ensure sustainable operations. This research work proposes a novel approach for the classification of the cooling behavior of a hydraulic test rig. Three fault conditions for the cooling system of the hydraulic test rig were used. The spectrograms were generated using the time series data for three fault conditions. The CNN variant, the Residual Network, was used for the classification of the fault conditions. Various features were extracted from the data including the F-score, precision, accuracy, and recall using a Confusion Matrix. The data contained 43,680 attributes and 2205 instances. After testing, validating, and training, the model accuracy of the ResNet-18 architecture was found to be close to 95%.

Current Status and Applications for Hydraulic Pump Fault Diagnosis: A Review

To implement Prognostics Health Management (PHM) for hydraulic pumps, it is very important to study the faults of hydraulic pumps to ensure the stability and reliability of the whole life cycle. The research on fault diagnosis has been very active, but there is a lack of systematic analysis and summary of the developed methods. To make up for this gap, this paper systematically summarizes the relevant methods from the two aspects of fault diagnosis and health management. In addition, in order to further facilitate researchers and practitioners, statistical and comparative analysis of the reviewed methods is carried out, and a future development direction is prospected.

Anomaly Detection Based on Time Series Data of Hydraulic Accumulator

Although hydraulic accumulators play a vital role in the hydraulic system, they face the challenges of being broken by continuous abnormal pulsating pressure which occurs due to the malfunction of hydraulic systems. Hence, this study develops anomaly detection algorithms to detect abnormalities of pulsating pressure for hydraulic accumulators. A digital pressure sensor was installed in a hydraulic accumulator to acquire the pulsating pressure data. Six anomaly detection algorithms were developed based on the acquired data. A threshold averaging algorithm over a period based on the averaged maximum/minimum thresholds detected anomalies 2.5 h before the hydraulic accumulator failure. In the support vector machine (SVM) and XGBoost model that distinguish normal and abnormal pulsating pressure data, the SVM model had an accuracy of 0.8571 on the test set and the XGBoost model had an accuracy of 0.8857. In a convolutional neural network (CNN) and CNN autoencoder model trained with normal and abnormal pulsating pressure images, the CNN model had an accuracy of 0.9714, and the CNN autoencoder model correctly detected the 8 abnormal images out of 11 abnormal images. The long short-term memory (LSTM) autoencoder model detected 36 abnormal data points in the test set.

Metric Learning-Based Fault Diagnosis and Anomaly Detection for Industrial Data With Intraclass Variance

Industrial system monitoring includes fault diagnosis and anomaly detection, which have received extensive attention, since they can recognize the fault types and detect unknown anomalies. However, a separate fault diagnosis method or anomaly detection method cannot identify unknown faults and distinguish between different fault types simultaneously; thus, it is difficult to meet the increasing demand for safety and reliability of industrial systems. Besides, the actual system often operates in varying working conditions and is disturbed by the noise, which results in the intraclass variance of the raw data and degrades the performance of industrial system monitoring. To solve these problems, a metric learning-based fault diagnosis and anomaly detection method is proposed. Fault diagnosis and anomaly detection are adaptively fused in the proposed end-to-end model, where anomaly detection can prevent the model from misjudging the unknown anomaly as the known type, while fault diagnosis can identify the specific type of system fault. In addition, a novel multicenter loss is introduced to restrain the intraclass variance. Compared with manual feature extraction that can only extract suboptimal features, it can learn discriminant features automatically for both fault diagnosis and anomaly detection tasks. Experiments on three-phase flow (TPF) facility and Case Western Reserve University (CWRU) bearing have demonstrated that the proposed method can avoid the interference of intraclass variances and learn features that are effective for identifying tasks. Moreover, it achieves the best performance in both fault diagnosis and anomaly detection.

Multi-Filter Clustering Fusion for Feature Selection in Rotating Machinery Fault Classification

In the fault classification process, filter methods that sequentially remove unnecessary features have long been studied. However, the existing filter methods do not have guidelines on which, and how many, features are needed. This study developed a multi-filter clustering fusion (MFCF) technique, to effectively and efficiently select features. In the MFCF process, a multi-filter method combining existing filter methods is first applied for feature clustering; then, key features are automatically selected. The union of key features is utilized to find all potentially important features, and an exhaustive search is used to obtain the best combination of selected features to maximize the accuracy of the classification model. In the rotating machinery examples, fault classification models using MFCF were generated to classify normal and abnormal conditions of rotational machinery. The obtained results demonstrated that classification models using MFCF provide good accuracy, efficiency, and robustness in the fault classification of rotational machinery.

The reconstruction on the game networks with binary-state and multi-state dynamics

Reconstruction of network is to infer the relationship among nodes using observation data, which is helpful to reveal properties and functions of complex systems. In view of the low reconstruction accuracy based on small data and the subjectivity of threshold to infer adjacency matrix, the paper proposes two models: the quadratic compressive sensing (QCS) and integer compressive sensing (ICS). Then a combined method (CCS) is given based on QCS and ICS, which can be used on binary-state and multi-state dynamics. It is found that CCS is usually a superior method comparing with compressive sensing, LASSO on several networks with different structures and scales. And it can infer larger node correctly than the other two methods. The paper is conducive to reveal the hidden relationship with small data so that to understand, predicate and control a vast intricate system.

Impact of Sensor Data Characterization with Directional Nature of Fault and Statistical Feature Combination for Defect Detection on Roll-to-Roll Printed Electronics

Gravure printing, which is a roll-to-roll printed electronics system suitable for high-speed patterning of functional layers have advantages of being applied to flexible webs in large areas. As each of the printing procedure from inking to doctoring followed by ink transferring and setting influences the quality of the pattern geometry, it is necessary to detect and diagnose factors causing the printing defects beforehand. Data acquisition with three triaxial acceleration sensors for fault diagnosis of four major defects such as doctor blade tilting fault was obtained. To improve the diagnosis performances, optimal sensor selection with Sensor Data Efficiency Evaluation, sensitivity evaluation for axis selection with Directional Nature of Fault and feature variable optimization with Feature Combination Matrix method was applied on the raw data to form a Smart Data. Each phase carried out on the raw data progressively enhanced the diagnosis results in contents of accuracy, positive predictive value, diagnosis processing time, and data capacity. In the case of doctor blade tilting fault, the diagnosis accuracy increased from 48% to 97% with decreasing processing time of 3640 s to 16 s and the data capacity of 100 Mb to 5 Mb depending on the input data between raw data and Smart Data.

A Novel Approach for Network Intrusion Detection Using Multistage Deep Learning Image Recognition

The current rise in hacking and computer network attacks throughout the world has heightened the demand for improved intrusion detection and prevention solutions. The intrusion detection system (IDS) is critical in identifying abnormalities and assaults on the network, which have grown in size and pervasiveness. The paper proposes a novel approach for network intrusion detection using multistage deep learning image recognition. The network features are transformed into four-channel (Red, Green, Blue, and Alpha) images. The images then are used for classification to train and test the pre-trained deep learning model ResNet50. The proposed approach is evaluated using two publicly available benchmark datasets, UNSW-NB15 and BOUN Ddos. On the UNSW-NB15 dataset, the proposed approach achieves 99.8% accuracy in the detection of the generic attack. On the BOUN DDos dataset, the suggested approach achieves 99.7% accuracy in the detection of the DDos attack and 99.7% accuracy in the detection of the normal traffic.