×
The submission system is temporarily under maintenance. Please send your manuscripts to
Go to Editorial ManagerElectrical motors have been engaged in many residential, industrial and commercial applications. The speed of an electric motor is an essential output quantity which is needed in many processing systems. Therefore, estimating the speed of an electrical motor is an integral part in the hierarchy of operational and control process. In this work, a new speed estimation method is proposed which is based on a naturally occurring signal; the mechanical vibrations the body of the motor endure during operation. These vibration signals are measured in multi-axial dimension through accelerometer and gyroscope. Furthermore, the collected data is trained in a machine learning model. The model is used subsequently to estimate the speed of a self-excited direct current (DC) motor. Two approaches (offline and onboard) are followed to evaluate the fitness and the performance of the proposed method. The offline approach is performed using regression learner MATLAB toolbox and many algorithms are tested and results with different performance metrics are presented. The algorithm that yields best performance in terms of minimum Root Mean Square and maximum regression factor (R2) is selected as candidate for offline revolutions per minute (rpm) estimation. Results documents that with Gaussian process regression algorithm, estimations are obtained with a mean square error of 7 rpm and an R2 value of 1 which is considered a very satisfactory performance. The second approach is motor speed estimation in real time using vibration signals with deep learning model implemented on limited resources electronic board which is proposed for the first time to the best of our knowledge. The proposed method has been successfully implemented by low consumption resources from the selected board with 6.5 kb of ram and 91ms latency. Even with the limited resources, a rated speed estimate percentage error of 0.18% was recorded from real time results. Moreover, the proposed method is characterized by its simplicity, low technical requirements and eventually low cost of implementation. The aforementioned features make this method an attractive platform for speed estimation in many industrial applications.
Federated learning (FL) is one of the newest and most significant fields for developing artificial intelligence applications. This technology trains its models in a distributed way, using data from different clients who work together in the system without sharing their data. The training process is kept local to protect the privacy of the data. Among the many difficulties that have arisen due to the novelty of this technology is the issue of heterogeneous data between typical clients. Client’s data may differ from each other in different respects, for example non identically and independent distribution (non-IID) between clients and the difference in the type of data used in each client. This can lead to inconsistencies in the model’s predictions and other undesirable outcomes. This paper discussed ways to solve this problem where clients with heterogeneous data were dealt with in terms of number and type. Because there are different types of image data through which doctors can diagnose coronavirus, such as x-rays, CT-scan. A hybrid convolution neural network (CNN ) and long short-term memory model (LSTM) has been proposed in a federated learning system to predict the incidence of this disease by using two clients, each with one of these different data. Good results were obtained with an accuracy of more than 99% in one customer and more than 95% in the second client while maintaining the privacy of this data.