No abstract

No abstract

The feature extraction technique plays a vital role in obtaining better classification accuracy. In this paper, a novel framework is proposed, which develops two-dimensional (2D) images for convolutional neural network (CNN) to classify four (left hand, right hand, feet, and tongue) MI tasks. 2D image is formed by decomposing each trial using continuous wavelet transform (CWT) filter bank after pre-processing the MI-based EEG data by multi-class common spatial pattern (CSP) method. Obtained images are used to train the CNN model for classification. The proposed framework is evaluated using publicly available BCI competition IV dataset 2a by calculating the classification accuracy for all subjects. Results show that the proposed framework has been giving better classification accuracy than some existing CNN-based and conventional machine learning-based approaches compared in this paper. The average time required to train CNN using the proposed framework is 12.67 s, acceptable for online MI-based BCI applications.

A new approach is proposed to design the sliding mode (SM) controller for the unstable second-order plus dead-time (SOPDT) processes. The sliding mode control consists of two control laws ie continuous control law and discontinuous control law. The continuous control law parameters have been derived in terms of unstable SOPDT process parameters using the root locus technique. On the other hand, the parameters of discontinuous control law are tuned by optimizing a performance index using a recently developed metaheuristic search algorithm, namely the grasshopper optimization technique. The performance index is framed to achieve a good trade-off between performance and control efforts. Finally, simulations are conducted to validate the effectiveness of the proposed approach over the other existing techniques. It is observed that the proposed approach is able to deliver better disturbance rejection, minimal control efforts and good setpoint tracking.

Purpose This paper aims to present an efficient and simplified proportional-integral/proportional-integral and derivative controller design method for the higher-order stable and integrating processes with time delay in the cascade control structure (CCS). Design/methodology/approach Two approaches based on model matching in the frequency domain have been proposed for tuning the controllers of the CCS. The first approach is based on achieving the desired load disturbance rejection performance, whereas the second approach is proposed to achieve the desired setpoint performance. In both the approaches, matching between the desired model and the closed-loop system with the controller is done at a low-frequency point. Model matching at low-frequency points yields a linear algebraic equation and the solution to these equations yields the controller parameters. Findings Simulations have been conducted on several examples covering high order stable, integrating, double integrating processes with time delay and nonlinear continuous stirred tank reactor. The performance of the proposed scheme has been compared with recently reported work having modified cascade control configurations, sliding mode control, model predictive control and fractional order control. The performance of both the proposed schemes is either better or comparable with the recently reported methods. However, the proposed method based on desired load disturbance rejection performance outperforms among all these schemes. Originality/value The main advantages of the proposed approaches are that they are directly applicable to any order processes, as they are free from time delay approximation and plant order reduction. In addition to this, the proposed schemes are capable of handling a wide range of different dynamical processes in a unified way.