Improving autoencoder-based unsupervised damage detection in uncontrolled structural health monitoring under noisy conditions

Structural health monitoring is widely utilized in outdoor environments,especially under harsh conditions,which can introduce noise into the monitoring system.Therefore,designing an effective denoising strategy to enhance the performance of guided wave damage detection in noisy environments is crucial.This paper introduces a local temporal principal component analysis(PCA)reconstruction approach for denoising guided waves prior to implementing unsupervised damage detection,achieved through novel autoencoder-based reconstruction.Experimental results demonstrate that the proposed denoising method significantly enhances damage detection performance when guided waves are contaminated by noise,with SNR values ranging from 10 to-5 dB.Following the implementation of the proposed denoising approach,the AUC score can elevate from 0.65 to 0.96 when dealing with guided waves corrputed by noise at a level of-5 dB.Additionally,the paper provides guidance on selecting the appropriate number of components used in the denoising PCA reconstruction,aiding in the optimization of the damage detection in noisy conditions.

Statistical Models for Condition Monitoring and State of Health Estimation of Lithium-Ion Batteries for Ships

Battery systems are increasingly being used for powering ocean going ships,and the number of fully electric or hybrid ships relying on battery power for propulsion is growing.To ensure the safety of such ships,it is important to monitor the available energy that can be stored in the batteries,and classification societies typically require the state of health(SOH)to be verified by independent tests.This paper addresses statistical modeling of SOH for maritime lithium-ion batteries based on operational sensor data.Various methods for sensor-based,data-driven degradation monitoring will be presented,and advantages and challenges with the different approaches will be discussed.The different approaches include cumulative degradation models and snapshot models,models that need to be trained and models that need no prior training,and pure data-driven models and physics-informed models.Some of the methods only rely on measured data,such as current,voltage,and temperature,whereas others rely on derived quantities such as state of charge.Models include simple statistical models and more complicated machine learning techniques.Insight from this exploration will be important in establishing a framework for data-driven diagnostics and prognostics of maritime battery systems within the scope of classification societies.

Temporally Preserving Latent Variable Models:Offline and Online Training for Reconstruction and Interpretation of Fault Data for Gearbox Condition Monitoring

Latent variable models can effectively determine the condition of essential rotating machinery without needing labeled data.These models analyze vibration data via an unsupervised learning strategy.Temporal preservation is necessary to obtain an informative latent manifold for the fault diagnosis task.In a temporalpreserving context,two approaches exist to develop a condition-monitoring methodology:offline and online.For latent variable models,the available training modes are not different.While many traditional methods use offline training,online training can dynamically adjust the latent manifold,possibly leading to better fault signature extraction from the vibration data.This study explores online training using temporal-preserving latent variable models.Within online training,there are two main methods:one focuses on reconstructing data and the other on interpreting the data components.Both are considered to evaluate how they diagnose faults over time.Using two experimental datasets,the study confirms that models from both training modes can detect changes in machinery health and identify faults even under varying conditions.Importantly,the complementarity of offline and online models is emphasized,reassuring their versatility in fault diagnostics.Understanding the implications of the training approach and the available model formulations is crucial for further research in latent variable modelbased fault diagnostics.

AN INTELLIGENT TOOL CONDITION MONITORING SYSTEM USING FUZZY NEURAL NETWORKS

Reliable on line cutting tool conditioning monitoring is an essential feature of automatic machine tool and flexible manufacturing system (FMS) and computer integrated manufacturing system (CIMS). Recently artificial neural networks (ANNs) are used for this purpose in conjunction with suitable sensory systems. The present work in Norwegian University of Science and Technology (NTNU) uses back propagation neural networks (BP) and fuzzy neural networks (FNN) to process the cutting tool state data measured with force and acoustic emission (AE) sensors, and implements a valuable on line tool condition monitoring system using the ANNs. Different ANN structures are designed and investigated to estimate the tool wear state based on the fusion of acoustic emission and force signals. Finally, four case studies are introduced for the sensing and ANN processing of the tool wear states and the failures of the tool with practical experiment examples. The results indicate that a tool wear identification system can be achieved using the sensors integration with ANNs, and that ANNs provide a very effective method of implementing sensor integration for on line monitoring of tool wear states and abnormalities.

Working Condition Real-Time Monitoring Model of Lithium Ion Batteries Based on Distributed Parameter System and Single Particle Model

Lithium ion batteries are complicated distributed parameter systems that can be described preferably by partial differential equations and a field theory. To reduce the solution difficulty and the calculation amount, if a distributed parameter system is described by ordinary differential equations （ODE） during the analysis and the design of distributed parameter system, the reliability of the system description will be reduced, and the systemic errors will be introduced. Studies on working condition real-time monitoring can improve the security because the rechargeable LIBs are widely used in many electronic systems and electromechanical equipment. Single particle model （SPM） is the simplification of LIB under some approximations, and can estimate the working parameters of a LIB at the faster simulation speed. A LIB modelling algorithm based on PDEs and SPM is proposed to monitor the working condition of LIBs in real time. Although the lithium ion concentration is an unmeasurable distributed parameter in the anode of LIB, the working condition monitoring model can track the real time lithium ion concentration in the anode of LIB, and calculate the residual which is the difference between the ideal data and the measured data. A fault alarm can be triggered when the residual is beyond the preset threshold. A simulation example verifies that the effectiveness and the accuracy of the working condition real-time monitoring model of LIB based on PDEs and SPM.

Impact of Typical Steady-state Conditions and Transient Conditions on Flow Ripple and Its Test Accuracy for Axial Piston Pump

The current research about the flow ripple of axial piston pump mainly focuses on the effect of the structure of parts on the flow ripple. Therein, the structure of parts are usually designed and optimized at rated working conditions. However, the pump usually has to work in large-scale and time-variant working conditions. Therefore, the flow ripple characteristics of pump and analysis for its test accuracy with respect to variant steady-state conditions and transient conditions in a wide range of operating parameters are focused in this paper. First, a simulation model has been constructed, which takes the kinematics of oil film within friction pairs into account for higher accuracy. Afterwards, a test bed which adopts Secondary Source Method is built to verify the model. The simulation and tests results show that the angular position of the piston, corresponding to the position where the peak flow ripple is produced, varies with the different pressure. The pulsating amplitude and pulsation rate of flow ripple increase with the rise of pressure and the variation rate of pressure. For the pump working at a constant speed, the flow pulsation rate decreases dramatically with the increasing speed when the speed is less than 27.78% of the maximum speed, subsequently presents a small decrease tendency with the speed further increasing. With the rise of the variation rate of speed, the pulsating amplitude and pulsation rate of flow ripple increase. As the swash plate angle augments, the pulsating amplitude of flow ripple increases, nevertheless the flow pulsation rate decreases. In contrast with the effect of the variation of pressure, the test accuracy of flow ripple is more sensitive to the variation of speed. It makes the test accuracy above 96.20% available for the pulsating amplitude of pressure deviating within a range of ~6% from the mean pressure. However, with a variation of speed deviating within a range of ±2% from the mean speed, the attainable test accuracy of flow ripple is above 93.07%. The model constructed in this research proposes a method to determine the flow ripple characteristics of pump and its attainable test accuracy under the large-scale and time-variant working conditions. Meanwhile, a discussion about the variation of flow ripple and its obtainable test accuracy with the conditions of the pump working in wide operating ranges is given as well.

Turbopump Condition Monitoring Using Incremental Clustering and One-class Support Vector Machine

Turbopump condition monitoring is a significant approach to ensure the safety of liquid rocket engine （LRE）.Because of lack of fault samples,a monitoring system cannot be trained on all possible condition patterns.Thus it is important to differentiate abnormal or unknown patterns from normal pattern with novelty detection methods.One-class support vector machine （OCSVM） that has been commonly used for novelty detection cannot deal well with large scale samples.In order to model the normal pattern of the turbopump with OCSVM and so as to monitor the condition of the turbopump,a monitoring method that integrates OCSVM with incremental clustering is presented.In this method,the incremental clustering is used for sample reduction by extracting representative vectors from a large training set.The representative vectors are supposed to distribute uniformly in the object region and fulfill the region.And training OCSVM on these representative vectors yields a novelty detector.By applying this method to the analysis of the turbopump＇s historical test data,it shows that the incremental clustering algorithm can extract 91 representative points from more than 36 000 training vectors,and the OCSVM detector trained on these 91 representative points can recognize spikes in vibration signals caused by different abnormal events such as vane shedding,rub-impact and sensor faults.This monitoring method does not need fault samples during training as classical recognition methods.The method resolves the learning problem of large samples and is an alternative method for condition monitoring of the LRE turbopump.

Online Condition Monitoring of Gripper Cylinder in TBM Based on EMD Method

The gripper cylinder that provides braced force for Tunnel Boring Machine （TBM） might fail due to severe vibration when the TBM excavates in the tunnel. Early fault diagnosis of the gripper cylinder is important for the safety and efficiency of the whole tunneling project. In this paper, an online condition monitoring system based on the Empirical Mode Decomposition （EMD） method is estab- lished for fault diagnosis of the gripper cylinder while TBM is working. Firstly, the lumped mass parameter model of the gripper cylinder is established considering the influence of the variable stiffness at the rock interface, the equivalent stiffness of the oil, the seals, and the copper guide sleeve. The dynamic performance of the gripper cylinder is investigated to provide basis for its health condition evaluation. Then, the EMD method is applied to identify the characteristic frequencies of the gripper cylinder for fault diagnosis and a field test is used to verify the accuracy of the EMD method for detection of the characteristic frequencies. Furthermore, the contact stiff- ness at the interface between the barrel and the rod is calculated with Hertz theory and the relationship between the natural frequency and the stiffness varying with the health condition of the cylinder is simulated based on the dynamic model. The simulation shows that the character- istic frequencies decrease with the increasing clearance between the barrel and the rod, thus the defects could be indicated by monitoring the natural frequency. Finally, a health condition management system of the gripper cylin- der based on the vibration signal and the EMD method is established, which could ensure the safety of TBM.

Sparsity-Assisted Intelligent Condition Monitoring Method for Aero-engine Main Shaft Bearing

Weak feature extraction is of great importance for condition monitoring and intelligent diagnosis of aeroengine.Aimed at achieving intelligent diagnosis of aero-engine main shaft bearing,an enhanced sparsity-assisted intelligent condition monitoring method is proposed in this paper.Through analyzing the weakness of convex sparse model,i.e.the tradeoff between noise reduction and feature reconstruction,this paper proposes an enhanced-sparsity nonconvex regularized convex model based on Moreau envelope to achieve weak feature extraction.Accordingly,a sparsity-assisted deep convolutional variational autoencoders network is proposed,which achieves the intelligent identification of fault state through training denoised normal data.Finally,the effectiveness of the proposed method is verified through aero-engine bearing run-to-failure experiment.The comparison results show that the proposed method is good at abnormal pattern recognition,showing a good potential for weak fault intelligent diagnosis of aero-engine main shaft bearings.

Swarm-LSTM: Condition Monitoring of Gearbox Fault Diagnosis Based on Hybrid LSTM Deep Neural Network Optimized by Swarm Intelligence Algorithms

Nowadays,renewable energy has been emerging as the major source of energy and is driven by its aggressive expansion and falling costs.Most of the renewable energy sources involve turbines and their operation and maintenance are vital and a difficult task.Condition monitoring and fault diagnosis have seen remarkable and revolutionary up-gradation in approaches,practices and technology during the last decade.Turbines mostly do use a rotating type of machinery and analysis of those signals has been challenging to localize the defect.This paper proposes a new hybrid model wherein multiple swarm intelligence models have been evaluated to optimize the conventional Long Short-Term Memory(LSTM)model in classifying the faults from the vibration signals data acquired from the gearbox.This helps to analyze the performance and behavioral patterns of the system more effectively and efficiently which helps to suggest for replacement of the unit with higher precision.The results have demonstrated that the proposed hybrid modeling approach is effective in classifying the faults of the gearbox from the time series data and achieve higher diagnostic accuracy in comparison to the conventional LSTM methods.

A system for underground road condition monitoring

Poor road conditions in underground mine tunnels can lead to decreased production efficiency and increased wear on production vehicles. A prototype system for road condition monitoring is presented in this paper to counteract this. The system consists of three components i.e. localization, road monitoring, and scheduling. The localization of vehicles is performed using a Rao-Blackwellized extended particle filter, combining vehicle mounted sensors with signal strengths of Wi Fi access points. Two methods for road monitoring are described: a Kalman filter used together with a model of the vehicle suspension system, and a relative condition measure based on the power spectral density. Lastly, a method for taking automatic action on an ill-conditioned road segment is proposed in the form of a rescheduling algorithm.The scheduling algorithm is based on the large neighborhood search and is used to integrate road service activities in the short-term production schedule while minimizing introduced production disturbances.The system is demonstrated on experimental data collected in a Swedish underground mine.

Virtual sensing for gearbox condition monitoring based on kernel factor analysis

Vibration and oil debris analysis are widely used in gearbox condition monitoring as the typical indirect and direct sensing techniques. However, they have their own advantages and disadvantages. To better utilize the sensing information and overcome its shortcomings, this paper presents a virtual sensing technique based on artificial intelligence by fusing low-cost online vibration measurements to derive a gearbox condition indictor, and its performance is comparable to the costly offline oil debris measurements. Firstly, the representative features are extracted from the noisy vibration measurements to characterize the gearbox degradation conditions. However, the extracted features of high dimensionality present nonlinearity and uncertainty in the machinery degradation process. A new nonlinear feature selection and fusion method,named kernel factor analysis, is proposed to mitigate the aforementioned challenge. Then the virtual sensing model is constructed by incorporating the fused vibration features and offline oil debris measurements based on support vector regression. The developed virtual sensing technique is experimentally evaluated in spiral bevel gear wear tests,and the results show that the developed kernel factor analysis method outperforms the state-of-the-art featureselection techniques in terms of virtual sensing model accuracy.

Bio-inspired computational techniques based on advanced condition monitoring

The application of bio-inspired computational techniques to the field of condition monitoring is addressed. First, the bio-inspired computational techniques are briefly addressed; the advantages and disadvantages of these computational methods are made clear. Then, the roles of condition monitoring in the predictive maintenance and failures prediction and the development trends of condition monitoring are discussed. Finally, a case study on the condition monitoring of grinding machine is described, which shows the application of bio-inspired computational technique to a practical condition monitoring system.

Massive Power Device Condition Monitoring Data Feature Extraction and Clustering Analysis using MapReduce and Graph Model

Effective storage,processing and analyzing of power device condition monitoring data faces enormous challenges.A framework is proposed that can support both MapReduce and Graph for massive monitoring data analysis at the same time based on Aliyun DTplus platform.First,power device condition monitoring data storage based on MaxCompute table and parallel permutation entropy feature extraction based on MaxCompute MapReduce are designed and implemented on DTplus platform.Then,Graph based k-means algorithm is implemented and used for massive condition monitoring data clustering analysis.Finally,performance tests are performed to compare the execution time between serial program and parallel program.Performance is analyzed from CPU cores consumption,memory utilization and parallel granularity.Experimental results show that the designed framework and parallel algorithms can efficiently process massive power device condition monitoring data.

Condition Monitoring and Fault Diagnosis Based on Rough Set Theory

In order to raise the efficiency,automatization and intelligentization of condition monitoring and fault diagnosis for complex equipment systems,rough set theory is used to the field. A feature reduction algorithm based on rough set theory is adopted to extract condition information in monitoring and diagnosis for an engine,so that the technology condition monitoring parameters are optimized. The decision tables for each fault source are built and the diagnosis rules rooting in rough set reduction is applied to carry through intelligent fault diagnosis. The cases studied show that rough set method in condition monitoring and fault diagnosis can lighten the work burden in feature selection and afford advantages for autonomic learning and decision during diagnosis.

Drilling signals analysis for tricone bit condition monitoring

This paper presents a novel approach to investigate the relations between drilling signals and bit wear condition in real world full-scale mining operations.This research addresses the increasing demand for automation in mining to increase the efficiency,safety,and ability to work in harsh environments.A crucial issue in fully autonomous unmanned drilling is to have a system to detect the bit wear condition through the drilling signals analysis in real time.In this work,based on extensive field studies,a novel qualitative method for tricone bit wear state classification is developed and introduced.The relations between drilling vibration as well as electric motor current signals and bit wear are investigated and bit failure vibration frequencies,regardless of the geological conditions,are introduced.Bit failure frequencies are experimentally investigated and analytically calculated.Finally,the effect of bit design parameters on the failure frequencies is presented for the application of bit wear condition monitoring and bit failure prediction.

Performance evaluation of near real-time condition monitoring in haul trucks

Strategic maintenance plays a key role in ensuring high availability and utilization of the haul trucks,and as equipment began to grow more complex towards the end of the 20th century,there was a need for a proactive maintenance strategy,which led to the development of condition-based maintenance.Realtime condition monitoring(RTCM)is the ability to perform condition monitoring in real-time and has the ability to alert maintenance and operations of abnormal conditions.These alarms can be used as an indication leading to a problem,and if a suitable corrective action is initiated in time,it could result in significant savings of equipment downtime and repair costs.This study aims to compare some maintenance performance indicators prior to and after implementation of RTCM strategy at a mine site using some tests of statistical significance.The study also indicated the presence of seasonality in the data,and thus the data was deseasonalized and detrended prior to being subjected to the statistical tests.Finally,the results indicated that RTCM strategy has proven to be successful in improving the availability for some of the failure categories chosen in this study.

The integrated monitoring system for running parameters of key mining equipment based on condition monitoring technology

An integrated monitoring system for running parameters of key mining equipmenton the basis of condition monitoring technology and modern communication networktechnology was developed.The system consists of a client computer with functions ofsignal acquisition and processing, and a host computer in the central control room.Thesignal acquisition module of the client computer can collect the running parameters fromvarious monitoring terminals in real-time.The DSP high-speed data processing system ofthe main control module can quickly achieve the numerical calculation for the collectedsignal.The signal modulation and signal demodulation are completed by the frequencyshift keying circuit and phase-locked loop frequency circuit, respectively.Finally, the signalis sent to the host computer for logic estimation and diagnostic analysis using the networkcommunication technology, which is helpful for technicians and managers to control therunning state of equipment.

Vibration Signals and Condition Monitoring for Wind Turbines

Rolling element bearings are critical parts of modern wind turbines as they carry the loads of the turning structure and the wind force. The stochastic nature of the wind loads makes it difficult to estimate the useful operational life of the bearings. Condition monitoring of these bearings in a real time environment could be very helpful in estimating their performance and in scheduling maintenance actions when a condition-based maintenance strategy is followed. This procedure can be successfully implemented by using vibration analysis in the time domain or in the frequency domain, giving useful results about the current condition of bearings and the location of potential faults. Permanently located transducers on proper positions on the bearings’ housings can be used in order to collect, process and evaluate real time measurements and provide information about the bearing’s performance. In this work, a test rig is utilized in order to evaluate the performance of rolling bearings. The results of the experimentation are satisfactory and the progress of fatigue failures can be predicted through vibration analysis techniques showing that implementation in real scale may be useful.

Study of Fuzzy Neural Networks Model for System Condition Monitoring of AUV

A structure equivalent model of fuzzy-neural networks for system condition monitoring is proposed, whose outputs are the condition or the degree of fault occurring in some parts of the system. This network is composed of six layers of neurons,which represent the membership functions, fuzzy rules and outputs respectively. The structure parameters and weights are obtained by processing off-line learning, and the fuzzy rules are derived from the experience. The results of the computer simulation for the autonomous underwater vehicle condition monitoring based on this fuzzy-neural networks show that the network is efficient and feasible in gaining the condition information or the degree of fault of the two main propellers.