Fault detection and diagnosis in voltage source inverters using principle component analysis

This study presents a comparative analysis of fault detection and classification techniques for identifying single and multiple open-switch faults in insulated-gate bipolar transistors (IGBTs) within a two-level, three-phase voltage source inverter (VSI) used in solar water pumping systems. Five machine learning algorithms are evaluated: Artificial Neural Network (ANN), Decision Tree (DT), K-Nearest Neighbors (KNN), Naive Bayes (NB), and Support Vector Machine (SVM). The evaluation employs current-based features (module, surface, and angle) extracted via the Clarke transformation. Standard performance metrics, including accuracy, sensitivity, specificity, precision, and F1 score, are employed for benchmarking. The ANN achieved perfect classification with 100 % accuracy, while the DT provided a competitive result of 95.56 %. In comparison, SVM and NB demonstrated lower performance under the same conditions. These results highlight the effectiveness of ANN for robust and reliable inverter fault diagnosis, while also showing DT as a strong, lightweight alternative. The contributions of this work include establishing a unified framework for comparing machine learning classifiers under identical operating conditions, validating Clarke transformation features for fault diagnosis in noisy signals, and providing quantified benchmarks to support engineers in selecting suitable diagnostic models. By linking algorithmic performance to practical benefits such as improved reliability and efficiency of inverter-based renewable energy systems, this study offers actionable insights for both researchers and practitioners.

Permanent magnet synchronous motors (PMSMs) drives using three-phase voltage-source inverters (VSIs) are currently used in many industrial applications. The reliability of VSIs is one of the most important factors to improve the reliability and availability levels of the drive. Accordingly, this paper presents a robust fault diagnostic method for multiple insulated gate bipolar transistors (IGBTs) open-circuit faults and current sensor faults in three-phase PMSM drives. The proposed observer-based algorithm relies on an adaptive threshold for fault diagnosis. Current sensor and open-circuit faults can be distinguished and the faulty sensors and/or power semiconductors are effectively isolated. The proposed technique is robust to machine parameters and load variations. Several simulation and experimental results using a vector-controlled PMSM drive are presented, showing the diagnostic algorithm robustness against false alarms and its effectiveness in both IGBTs and current sensors fault diagnosis.

One of the most used power converters is the three-phase voltage source power inverter. This is one of the reasons why it is so important to assure a continuous and safety operation of this power converter. This paper considers the application of PCA and average value computation to detect and identify faulty power switches in voltage source inverters. The PCA is applied to the inverter output currents in order to get a specific pattern that allows the identification of the fault. This results in a fast and reliable fault detection methodology. The applicability of the proposed technique is confirmed through simulation and experimental results.

A new switching scheme, which combines conventional space vector modulation (SVM) and carrier-based pulse width modulation (CBPWM), is proposed for a three-phase four-leg voltage source inverter (VSI). The proposed algorithm is simpler than conventional three-dimensional space vector modulation (3-D SVM) and easy in implementation in a real-time DSP system, it avoids the selection of a prism and tetrahedron in a 3-D SVM, therefore alleviates the computation burden of the DSP. Also, the concept of near-state pulse width modulation (NSPWM) method, which is used in a three-phase three-leg inverter to reduce the common mode voltage (CMV) noises, can be adopted in this algorithm so as to reduce the common mode voltage for a three-phase four-leg VSI. The feasibility of the proposed modulation technique is verified by both computer simulation and experimental results.

The proportional–integral (PI) control is usually applied in synchronous reference frame to regulate the output voltage of three-phase voltage source inverter (VSI). However, the PI controller is constrained by the capability of suppressing harmonic and negative sequence components; besides, its direct application in single-phase VSI is limited. While the proportional–resonant (PR) control can track ac reference and achieve harmonic suppression in stationary frame for both single-phase and three-phase VSI. Due to the high order of the resonant controller, lots of efforts are needed during the parameter design to get a satisfactory performance, such as trial-and-error. In this article, the way to approximately reduce the order of the resonant controller in the discrete time domain is presented, and then a method of selective pole placement and cancellation for the PR control design used in VSI is proposed. With the proposed method, the closed-loop pole-zero distribution of the PR control structure used in the VSI is well handled, and the parameters of the feedback controller and the reference feedforward controller are directly calculated by placing the selected poles and zeros in desired positions. The procedure of the parameter design is facilitated, and the performance of the VSI with PR control is improved.

This paper presents a diagnosis approach for inter-turn short circuit faults and open circuit switch faults in a brushless direct current (BLDC) motor driven by a three-phase voltage source inverter taking into account their hybrid behaviour. The diagnosis method consists of a discrete faults diagnosis module and a continuous faults diagnosis module combined by means of a hybrid automaton. The discrete diagnoser constitutes the discrete part of the hybrid automaton. It is designed based on a discrete presentation of the system taking into consideration the nominal and faulty states. On the other hand, continuous fault diagnosis is achieved by a hybrid sliding mode observer based on the Super Twisting algorithm. A Lyapunov based approach is used to address the convergence analysis problem of the hybrid observer. The performance of the proposed approach in detection and isolation of discrete and parametric faults is demonstrated by numerical simulations.

SummaryA fault diagnosis and compensation problem for two‐dimensional discrete time systems with time‐varying state delays is studied in this paper. The concerned two‐dimensional systems are described by the Fornasisi–Marchesini second model and are subject to unknown disturbances. First, a fault detection and diagnosis module is designed to obtain the information on sensor faults; a new fault detection and diagnosis integrated design, using the observer based on descriptor system approach, is proposed to detect and estimate the sensor faults. The integrated design can maximize the fault detection rate for a given false alarm rate. Sufficient conditions for the existence of the integrated fault detection and diagnosis design are derived in the context of norm evaluation and provided in terms of matrix inequalities. Second, a fault‐tolerant control module is proposed upon an existing output feedback controller. When the sensor fault occurs, the faulty measurement can be identified and corrected by the proposed fault detection and diagnosis module. In this case, the feedback controller can guarantee the performance of the closed‐loop system even when encountering sensor faults. Finally, the proposed method is applied to a thermal process to illustrate its effectiveness. Copyright © 2017 John Wiley & Sons, Ltd.

Three-phase voltage source inverters (VSI) reliability is critical for industrial and renewable energy applications, where open-circuit faults in switches can lead to performance degradation, energy loss, and equipment damage. This study presents a machine learning-based fault diagnosis framework integrating artificial neural networks (ANN) and ensemble classifiers to enhance fault detection accuracy under varying load conditions. Current signal waveforms were acquired using a Tektronix Digital Storage Oscilloscope and preprocessed using discrete wavelet transform to extract meaningful time-frequency features. Various machine learning algorithms, including support vector machines (SVM), k-nearest neighbors (KNN), decision trees (DT), random forest (RF), and Adaboost, were used for VSI fault classification. The classifiers' performance was compared using standard evaluation metrics, including precision, recall, F1-score, and accuracy, to ensure a comprehensive and fair assessment of fault classification effectiveness. Experimental results indicate that RF achieved the highest accuracy of 99.97%, followed by DT (99.49%), KNN with k = 3 (99.43%), and ANN (99.6%). The ANN model with an 18-10-1 architecture demonstrated strong classification performance, outperforming AdaBoost, which reached a peak accuracy of 96.95% at 500 estimators. SVM achieved an accuracy of 90.87%, making them less effective than ensemble-based classifiers. The study further validates that wavelet-based feature extraction combined with ANN reduces misclassification rates and improves real-time detection efficiency. Additionally, the proposed system reduced fault diagnosis time by 22.5% compared to conventional methods, making real-time implementation feasible. The findings confirm the effectiveness of the proposed ANN-based fault detection system in accurately classifying single and double open switch faults in VSI. Future work will integrate the system with IoT-based predictive maintenance and cloud-based fault detection for real-world applications.

Reservoir history matching by ensemble smoother with principle component and sensitivity analysis for heterogeneous formations

In modern society, civil plane has become more and more important in people's life. How to ensure safety and reliability of civil plane has attracted more and more attention by many research institutes and researchers. As one of the most important part in the civil plane, how to ensure normal operation of aero-engine becomes particularly important. According to statistics, the sensor fault accounts for more than 80% of the total failures in the aero-engine system. In this paper, an aero-engine model is established according to the operational principle of aero-engine using AMESim software. Secondly, the sensor sub-model in above model is modified to simulate drift failure, bias fault, and pulse fault, and to achieve aero-engine sensors' fault data. After that, sensor fault diagnosis model is built based on principle component analysis. Online detection and diagnosis can be realized using this diagnosis model. Simulation results show that the proposed method is feasible and effective.

Various faults occurred in the heating, ventilation and airconditioning (HVAC) systems usually lead to more energy consumption and worse thermal comfort inevitably. This paper presents a feasible and valid solution of HVAC fault detection and diagnosis (FDD) problem based on statistical machine learning technology. It learns the consistent nature of different types of faults of HVAC operation based on support vector machine (SVM), and then identify types of fault in all subsystems using the statistical relationships between groups of measurements. In order to speed up the learning process, principle component analysis (PCA) has been applied to compress the training data. Our approach models the dynamical sub-systems and sequence data in HVAC system. The learnt models can then be used for automatic fault detection and diagnosis. The approach has been tested on commercial HVAC systems. It had successfully detected and identified a number of typical AHU faults.

The interest in multiphase drives has been steadily growing during the last decade due to the promising potentials offered by multiphase machines over conventional three-phase counterpart. In this context, six-phase induction machines are preferably used in many diversified high-power applications. Generally, thanks to its improved flux distribution, the asymmetrical six-phase winding topology fed from two three-phase voltage source inverters (VSIs) is commonly employed with isolated neutral points to prevent the flow of zero sequence currents and to limit the number of current controllers to four instead of five when neutral points are connected. In this paper, an innovative design of a six-phase dc-ac inverter is proposed for such type of six-phase ac machines based on the single-ended primary-inductance converter (SEPIC) topology. The proposed topology employs only four active-legs with eight switches with the same output voltage magnitudes as in conventional VSIs and without the mandatory dead-time between switches in the same-leg. Also, it naturally delivers a pure sinusoidal waveform at the output stage. The principle of operation of the proposed inverter topology is investigated in details and assessed through a detailed simulation study of an open-loop control system.

Bus-clamping Pulse Width Modulation (PWM) is an effective method to reduce the switching loss in a three-phase voltage source inverter (VSI). In bus-clamping PWM scheme, the phase legs are switched using high frequency PWM signals for two-third of the line cycle, while for the remaining duration of cycle, the pole voltage is clamped to either positive or negative rail of the DC bus. In PWM operation of a half bridge, a dead-time is applied between the gate signals of complementary switches to ensure safe and reliable operation. However, introduction of dead-time leads to poor power quality, increased Total Harmonic Distortion (THD) and variation in actual voltage compared to the intended pole voltage. Moreover, when the bus-clamping technique is used, the PWM has both high frequency switching region and clamped region in a line cycle, and consequently, the undesired effects of dead-time are further aggravated. Therefore, in order to enhance the quality of output voltage, this paper presents a dead-time compensation strategy for a VSI operating with bus-clamping PWM. The proposed method calculates the required compensation term to be added on the modulation signal considering wide range of operating conditions. Additionally, the compensation includes a new strategy for low current conditions near zero-crossing to avoid distortion. The proposed method is verified by simulation and experiments in a three-phase VSI with a switching frequency of 100 kHz and a fundamental frequency of 60Hz.

This paper demonstrates improved performance of a three-phase voltage source inverter (VSI) when digital multisampled space vector modulation is used. The modulator and the inverter controller are implemented entirely using a field-programmable gate array platform, thus achieving increased bandwidth with respect to a typical digital signal processor or microprocessor-based controller. Increased controller bandwidth results in lower output voltage harmonic distortion in the frequency range above the fundamental and below the switching frequency. Experimental validation is presented together with the analysis carried out using a state space model of a VSI with an output LC filter.

The sinusoidal pulse-width modulation (SPWM) technique is typically used for three-phase voltage source inverter (VSI) in low-voltage motor drives due to its simplicity and reliability. However, these VSI face increasing challenge to minimize electromagnetic interference (EMI) and increase power quality. This paper proposes a new SPWM switching based on a Bat-like reference signal that has the advantages of minimizing the output-voltage total harmonics distortion (THDv), lowering the switch rate of change in voltage (dv/dt), reducing the number of switching transitions, and increasing the peak and r.m.s. ac output voltage of the three-phase VSI. The proposed switching technique is verified using Matlab/Simulik on a 2-kW, 300-V three-phase VSI connected to an RL load. This technique shows improvement in reducing the output-voltage THDv by 36% and lowering the number of switching transition by 60% as compared to SPWM switching that is based on singlefrequency signal. Moreover, the dv/dt is low and the peak output voltage is increase by approximately 17%.