In this paper, we apply the recently developed generalized parameter estimation-based observer design technique for state-affine systems to the practically important case of linear time-varying descriptor systems with...
详细信息
Fault detection in electric drives is crucial for ensuring operational reliability and minimizing downtime. This paper provides a brief overview of the methods based on machine learning used for fault detection in ele...
详细信息
ISBN:
(数字)9798350389234
ISBN:
(纸本)9798350353051
Fault detection in electric drives is crucial for ensuring operational reliability and minimizing downtime. This paper provides a brief overview of the methods based on machine learning used for fault detection in electric drives. Additional this research presents a machine learning approach that utilizes the Long Short-Term Memory Autoencoder model for fault detection deal with bearing faults. A flowchart illustrating the implementation steps of the Long Short-Term Memory Autoencoder algorithm is included. Additionally, a case study on bearing faults in an electric drive system is presented to demonstrate the advantage of the proposed approach. By training the Long Short-Term Memory Autoencoder on operational data, accurate prediction of bearing faults is achieved, enabling timely maintenance interventions. It is shown that in the 7th epoch of training, the model achieved a notably high accuracy more than $\mathbf{9 9 \%}$. The results highlight the potential of Long Short-Term Memory Autoencoder models as a valuable tool for fault detection in electric drives, thereby improving system reliability and performance.
Hyperexponential stability is investigated for dynamical systems with the use of both, explicit and implicit, Lyapunov function methods. A nonlinear hyperexponential control is designed for stabilizing linear systems....
Hyperexponential stability is investigated for dynamical systems with the use of both, explicit and implicit, Lyapunov function methods. A nonlinear hyperexponential control is designed for stabilizing linear systems. The tuning procedure is formalized in LMI form. Through numeric experiments, it is observed that the proposed hyperexponential control is less sensitive with respect to noises and discretization errors than its finite-time analog. It also demonstrates better performance in the presence of delays as well. Theoretical results are supported by numerical simulations.
The paper analyzes the preparation of software for acoustic signal classification with machine learning techniques for microcontrollers. The design process was tested for three types of devices: Nordic Thingy:53, *** ...
详细信息
Wired or contact power transfer (CPT) technology for electric vehicle (EVs) charging infrastructures has been commercialized by several manufacturers with either AC or DC charging configurations. AC charging system co...
详细信息
ISBN:
(数字)9798331530402
ISBN:
(纸本)9798331530419
Wired or contact power transfer (CPT) technology for electric vehicle (EVs) charging infrastructures has been commercialized by several manufacturers with either AC or DC charging configurations. AC charging system comes as Level-1 and Level-2 while DC charging system is either Level3 or Level-4. For DC charging systems which are known as fast chargers, shorter charging duration is the main advantage over the AC charging systems. However, both AC and DC based CPT require human interventions to plug the cable before charging process to take place. On the other hand, wireless power transfer (WPT) such as inductive, capacitive or laser beam based power transfer are the alternatives for unmanned aerial vehicles (UAVs) autonomous charging techniques. Apart from being resilient to environmental factors like dust and water, WPT eliminates human intervention and the hassles related to battery swapping. However, WPT technologies have some limitations such as coil weight constraints and low charging performance due misalignment which hinders its efficient utilization in UAVs applications. Thus, this paper presents the performance analysis of WPT in terms of power transfer capability while uncovering several constraints for its deployment in UAV applications. Specifically, WPT performance parameters like power transfer efficiency, coefficient of coupling, and quality factors are analyzed in detail. The results of the analysis show that increasing the misalignment reduces the coefficient of coupling and hence overall power transfer efficiency.
Wind turbines based on the Magnus effect open new prospects in terms of energy generation for areas with low average wind speeds. The main feature of these turbines is rotating cylinders which are used to create lift ...
Wind turbines based on the Magnus effect open new prospects in terms of energy generation for areas with low average wind speeds. The main feature of these turbines is rotating cylinders which are used to create lift instead of blades. Due to this unconventional design, different parameters of the cylinder, such as mass and moment of inertia, can greatly affect the dynamic characteristics of such energy systems. The goal of this paper was to present an approach to structural optimization of a cylinder flange of the experimental Magnus effect-based wind turbine. The proposed approach, presented as an algorithm, allows us to determine input data for the preferred structural optimization software and achieve lightweight, low-inertia parts. control simulation has shown that optimized parts ensure a high enough factor of safety while also significantly reducing the moment of inertia.
This article presents the application possibilities of unmanned aerial vehicles (UAVs) for the rapid and efficient determination of carbonyl compounds: formaldehyde and acetaldehyde concentrations in gaseous environme...
详细信息
Algorithms of evaluation battery state of charge (SoC) and power consumption should be implemented to electrical drive systems to control parameters of a battery. In the course of the work, the characteristic of disch...
Algorithms of evaluation battery state of charge (SoC) and power consumption should be implemented to electrical drive systems to control parameters of a battery. In the course of the work, the characteristic of discharge of the battery was obtained, the control system for a semiconductor step-up voltage converter for the battery management system (BMS) was developed, and the accuracy of the model was evaluated. A simulation model of the «PWM converter-DC drive» system with the battery was developed. Simulation results were compared with experimental data and conclusions were drawn about the possibility of determining the SoC of the battery.
This article presents the results of implementing a digital signal processing method for analyzing the acoustic emission (AE) signature and detecting defects such as cracks and pores. The signal processing method is b...
This article presents the results of implementing a digital signal processing method for analyzing the acoustic emission (AE) signature and detecting defects such as cracks and pores. The signal processing method is based on a cascade connection of digital high-pass filters and ensures the selection of informative signals and increases the signal-to-interference ratio. In order to control the formation of defects in the process of selective laser melting, the paper presents the results of experimental testing of the cascade filtration method for detecting defects in the internal structure such as cracks and pores. The signal components recorded during the development of defects during the growth of additive manufacturing products were extracted. AE signatures were analyzed and statistical relationships were assessed. The relationship between the parameters of the AE signal and the values of laser radiation power has been identified, which characterizes the process of defect formation.
This work deals with the problem of finite-time homogeneous observer design for linear systems. Compared to other works, the stability conditions and parameters tuning are formulated using linear matrix inequalities b...
详细信息
This work deals with the problem of finite-time homogeneous observer design for linear systems. Compared to other works, the stability conditions and parameters tuning are formulated using linear matrix inequalities being less conservative. The proposed results are based on the homogeneity property and Implicit Lyapunov Function method. The results are supported with simulation examples.
暂无评论