An Early Classification Approach for Improving Structural Rotor Fault Diagnosis

On basis of fault categories detection, the diagnosis of rotor fault causes is proposed, which has great contributions to the field of intelligent operation and maintenance. To improve the diagnostic accuracy and practical efficiency, a hybrid model based on the particle swarm optimization-extreme gradient boosting algorithm, namely, PSO-XGBoost is designed. XGBoost is used as a classifier to diagnose rotor fault causes, having good performance due to the second-order Taylor expansion and the explicit regularization term. PSO is used to automatically optimize the process of adjusting the XGBoost’s parameters, which overcomes the shortcomings when using the empirical method or the trial-and-error method to adjust parameters of the XGBoost model. The hybrid model combines the advantages of the two algorithms and can diagnose nine rotor fault causes accurately. Following diagnostic results, maintenance measures referring to the corresponding knowledge base are provided intelligently. Finally, the proposed PSO-XGBoost model is compared with five state-of-the-art intelligent classification methods. The experimental results demonstrate that the proposed method has higher diagnostic accuracy and practical efficiency in diagnosing rotor fault causes.

This paper presents recent developments in the use of a new technique for the diagnosis of faults in three-phase induction motors. The proposed diagnostic technique, which is based on the multiple reference frames theory (MRFT), has been successfully used in the diagnosis of stator faults, rotor faults and air gap eccentricity in line-connected and direct torque controlled (DTC) induction motors. The use of the MRFT represents a step forward towards the unification of the diagnostic process of faults in induction motors and drives by using the same technique to diagnose any type of faults. Due to the modular structure of the MRFT, virtually it can be used to diagnose any motor fault that has a direct impact in one or more supply currents. In addition, as it was demonstrated in a previous work, the technique is particularly adequate for an easy integration into the control system of a DTC induction motor drive at no additional cost.

In this paper, the induction motor rotor fault diagnosis system using current signals, which are measured using axis-transformation method, and speed, which is estimated using current information, are presented. In inverter-fed motor drives unlike line-driven motor drives the stator currents have numerous harmonics components and therefore fault diagnosis using stator currents is very difficult. The current and speed signal for rotor fault diagnosis needs to be precise. Also, high resolution information, which means the diagnosis system, demands additional hardware such as low pass filter, high resolution ADC, encoder and etc. Therefore, the proposed axis-transformation and speed estimation method are expected to contribute to low cost fault diagnosis systems in inverter-fed motor drives without the need for an encoder and any additional hardware. In order to confirm validity of the developed algorithms, various experiments for rotor faults are tested and the line current spectrum of each faulty situation using Park transformation and speed estimation method are compared with the results obtained from fast Fourier transforms.

This work deals with PQ state space modeling of asynchronous generator for rotor faults detection and diagnosis. Asynchronous generator is one of the mostly used machines as electrical power generator. This is due to its low cost and high reliability compared to other types of renewable energy sources. In order to ensure reliable and profitable operations, asynchronous generator must be carefully surveyed. The main major asynchronous generator faults are the rotor broken bars, the stator shorted turns and the air-gap eccentricities. The proposed method uses stator current to estimate the remaining state variables that are related to active and reactive power of the machine. In this context different models of the machine operating conditions are discussed from the theoretical point of view. Observer state estimation technique is used to reconstruct the non measured state variables of the machine. A simulation example is presented to point out the merits of the proposed double PQ transformation method.

Early rotor faults diagnosis is essential for induction motor life. The reason for which several signal processing diagnosis methods have been proposed. In order to enhance these methods, it is necessary to use a reliable mathematical model of the physical processed signal. Unfortunately, the stator current is usually obtained from an expensive measurement set-up or a complex modelling of the electric motor. The aim of this paper is therefore to use a model of the motor stator current under rotor fault condition. In other words, the searched model must take into account the effects of the fault. Indeed, it has been verified that the effect of a fault is reflected by an amplitude modulation as well as a phase modulation of the stator current corresponding respectively, to a load torque oscillation and an eccentricity in the air-gap. The paper will also investigate the impact of the severity of the rotor fault on these two types of modulation. Finally, a comparison study between both the simulated signal and the experimental signal is carried out to illustrate the merits of the model and its effectiveness.

Early classification of time series has been extensively studied for minimizing class prediction delay in time-sensitive applications such as healthcare and finance. A primary task of an early classification approach is to classify an incomplete time series as soon as possible with some desired level of accuracy. Recent years have witnessed several approaches for early classification of time series. As most of the approaches have solved the early classification problem with different aspects, it becomes very important to make a thorough review of the existing solutions to know the current status of the area. These solutions have demonstrated reasonable performance in a wide range of applications including human activity recognition, gene expression based health diagnostic, industrial monitoring, and so on. In this paper, we present a systematic review of current literature on early classification approaches for both univariate and multivariate time series. We divide various existing approaches into four exclusive categories based on their proposed solution strategies. The four categories include prefix based, shapelet based, model based, and miscellaneous approaches. The authors also discuss the applications of early classification in many areas including industrial monitoring, intelligent transportation, and medical. Finally, we provide a quick summary of the current literature with future research directions.

The rotor system is a core part of rotating machinery equipment. Its safe and reliable operation directly affects the economic benefit of using the equipment and the personal safety of users. To fully explore the complex feature mapping relationship between rotor vibration signals and fault types, rotor vibration signals were studied under different working conditions from the perspective of feature parameter construction and feature information mining. First, a variational mode decomposition algorithm was used to decompose the vibration signals, and quantum behavior particle swarm optimization was used to minimize the mean envelope entropy of intrinsic mode function components to determine the optimal combination of modal number and penalty coefficient. Second, the principal component analysis was used to reduce the dimensionality of IMF components of vibration signals. Finally, a support vector machine was used to mine the feature mapping relationship between vibration data after dimensionality reduction and rotor operation state to accurately identify rotor fault types. The proposed method was used to analyze the measured vibration signals of the rotor system. The experimental results showed that the proposed method effectively extracted characteristic information of the rotor running state from the vibration data, and the accuracies of four types of fault diagnoses were 100%, 88.89%, 100%, and 100%, respectively. In addition, the accuracies of the four fault diagnoses in this study were better than those of the previously reported models.

Reliable induction motor (IM) fault diagnosis is vital for industrial safety and operational continuity, mitigating costly unplanned downtime. Conventional approaches often struggle to capture complex multimodal signal relationships, are constrained to unimodal data or single fault types, and degrade under noise or crossdomain shifts. We propose the Multimodal Hypergraph Contrastive Attention Network (MM-HCAN), a unified framework that integrates multi-head attention with a hypergraph architecture and contrastive learning to jointly model global and localised intra- and inter-modal dependencies beyond Euclidean embedding spaces. MM-HCAN’s gains are driven by three innovations: (i) a structured multimodal fusion scheme with dynamically constructed hyperedges encoding higher-order cross-modal relationships; (ii) modality-specific hypergraph Laplacians for fine-grained, context-aware embedding refinement; and (iii) triplet-based contrastive learning performed directly on the hypergraph manifold, yielding more discriminative, topology-respecting representations. The model enables simultaneous diagnosis of bearing, stator, and rotor faults, addressing the engineering need for consolidated diagnostics. Evaluated on three real-world benchmarks, MM-HCAN achieves up to 99.82% accuracy with strong cross-domain generalisation and robustness to noise. An ablation study confirms the contribution of each component. MM-HCAN offers a scalable and robust solution for comprehensive multi-fault diagnosis, supporting predictive maintenance and extended asset longevity in industrial environments. Code is available at https://github.com/EngrUsmaanAli/MM-HCAN.

Fault diagnosis technology plays a vital role in the variety of critical engineering applications. Fuzzy approach is widely employed to cope with decision-making problems because it is in the simplest and most used form. This paper proposed a new similarity measure of generalized trapezoidal fuzzy numbers used for fault diagnosis. The presented similarity measure combines concepts of the geometric distance, the center of gravity point, the perimeter, and the area of the generalized trapezoidal fuzzy numbers for calculating the degree of similarity between generalized trapezoidal fuzzy numbers. This method is proposed to deal with both standardized and nonstandardized generalized trapezoidal fuzzy numbers. Some properties of the proposed similarity measure have been proved, and 12 sets of generalized fuzzy numbers have been used to compare the calculation results of the proposed similarity measures with the existing similarity measures. Comparison results indicate that the proposed similarity measure can overcome the drawbacks of existing similarity measures. Finally, a fault diagnosis experiment is carried out in laboratory based on multifunctional flexible rotor experiment bench. Experimental results demonstrate that the proposed similarity measure is more effective than other methods in terms of rotor fault diagnosis.

The screw rotor was a main part of the single screw compressor. The crack of the screw rotor would lead to fault of the single screw compressor. Therefore it was necessary to have fault diagnosis for the screw rotor. The wavelet finite element method was used to analyze the fault of the screw rotor. The changes of the natural frequency reflected the location and sizes of the cracks. The effect of main sizes including the length, the width and the depth on the natural frequency was studied, which offered the theoretical guidance for the fault diagnosis of the screw rotor in the further researches.

A fault diagnosis system based on LabVIEW is researched to analyze the rotating machinery fault. By developing data acquisition, signal analysis, processing and simulation, pattern recognition and fault diagnosis modules, this system can automatically measure and identify the shaft orbit of a rotor and get fault results. A pattern recognition method on the basis of the invariant moment algorithm is studied with LabVIEW and used to recognize shaft orbit shapes. The experimental results show that the shaft orbit of a rotor can be identified through invariant moment calculation. Because the shape of the shaft orbit of the rotor is close to the ellipse the main fault is imbalance. The fault result is proved to be correct by spectrum analysis. This study is helpful to develop a online fault diagnosis system based on LabVIEW.

This paper deals with different procedures that can be applied to diagnose the severity degree of rotor faults in closed loop controlled induction machines. A solution based on field oriented control (FOC) was investigated in order to validate the robustness of the diagnostic procedures towards the regulator parameters. Different diagnostic procedures were investigated, that rely on different diagnostic indexes. Specifically a few electric quantities are considered as possible fault alert signals: the spectral analysis of manipulated and controlled variables, the voltage homopolar components (voltage zero sequence), and the spectral analysis of motor input current due to a high frequency voltage injection. It turns out that the last quantity is a simple and effective index able to detect and quantify the fault severity degree. The procedures based on the above mentioned quantities are validated by experiments made with a DSP based current controlled voltage source inverter (CCVSI) supplying an induction machine with three rotors in different health conditions.

Fault diagnosis methods based on deep learning typically require a substantial amount of data. However, in practical engineering applications, it is often challenging to collect sufficient vibration signals due to various constraints, affecting the performance of these diagnostic methods. This paper proposes a Wavelet Scattering Transform-based Generative Adversarial Network model (WST-GAN). In this model, the wavelet scattering transform is embedded into the discriminator to enhance its discriminative capability. The performance of the generator is improved through adversarial training, which also increases the volume of training samples and enhances the model’s generalisation ability. The method includes offset resampling of limited samples to augment the initial sample size and converts signals into matrix images through pixel stacking, which are then fed into the generative adversarial network for training. The approach is applied to a case study using vibration signals from a simulated gas turbine rotor system. The fault detection accuracy of the dataset that underwent data augmentation exhibited a significant improvement compared to conditions with a small sample size, achieving a diagnostic accuracy of at least 97.03%. Experiments are conducted using a dual-disk rotor, and the identification accuracy for nine different fault conditions approached an accuracy of at least 99.47%.

This paper presents a spectrum analysis of the terminal voltage after supply disconnection. This voltage is induced only by the rotor flux and does not depend on the mechanical load, the voltage supply time harmonics, etc... The diagnosis technique is based on monitoring the amplitude of harmonics components in the frequency domain using both FFT and MUSIC methods. The experimental results show the presence of harmonics of order 6mplusmn1, m=1,2,3... for healthy and faulty machine. When there is a rotor bar fault, the amplitude of these harmonics grows. The spectral analysis shows that MUSIC method is more reliable than the FFT technique, and the induced voltage technique is independent on the load level.

Established the unsteady oil-film force with rub-impact fault of rotor-stator-bearing system mechanical models by using Lagrange equation, Application of numerical analysis to study it, concluded that the rotor system in frequency as the only incentive system control parameters of orbit and bifurcation diagrams; analysis of the nonlinear behavior of the response of the system and the failure mechanism, provide a theoretical basis for the safe operation of the rotor system and fault diagnosis. The results show that, when the excitation frequency as the only control parameters, the system exist periodic motion, quasi periodic motion and chaotic motion etc abundant nonlinear phenomena.