International Journal of Industrial Electronics, Control and Optimization
Volume:7 Issue: 4, Autumn 2024

In this paper, we propose the problem of connectivity maintenance for a group of robots that are connected with wireless communication. The communication between the agents is modeled by a fading channel, and the exchange of information is possible at permissible distances. The distributed controller is designed to maintain the global connectivity of the network communication graph. On the other hand, obtaining the best communication quality between agents is the definition of good network connectivity. Therefore, the Laplacian matrix is defined as a weighted graph according to the communication parameters. Initially, the controller uses the supergradation algorithm to remake the network Laplacian matrix for having a bigger second eigenvalue. While each agent in the network only has access to the neighbor's information. The control algorithm uses a multi-agent estimator to find the eigenvector corresponding to the second small eigenvalue. The formation problem has been considered in the second step after the reference topology has been obtained. Because of the nonlinear dynamics of the agents, the sliding mode controller has been used for this purpose. This robust controller could be a suitable choice due to the modeling uncertainty and sensor measurement uncertainty. Finally, an example of multi-agent robots is provided to evaluate the algorithm.

A new framework for controlling load frequency in a complex, interconnected power system with multiple sources has been developed. This framework combines a fuzzy logic controller (FLC) and a tilted integral derivative (TID) controller, creating a self-tuning fuzzy tilted integral derivative (STFTID) controller. The purpose of this controller is to conduct and reduce load frequency perturbations during the operation of a multi-area interconnected multi-source power system. The STFTID controller is optimized using a particle swarm optimization algorithm to minimize the frequency fluctuations effectively. Investigations of the proposed STFTID controller were performed for power systems with generation units of a conventional system and renewable energy sources. In the design process of the STFTID controller, various nonlinearities, uncertainties, and fluctuations are considered to simulate practical challenges. These challenges include generation rate constraints, governor deadband, and communication time delays (as the sources of nonlinearity), as well as fluctuations caused by step load switching and the connection of renewable power plants to the system. The STFTID controller is compared with the proportional integral derivative (PID), titled integral derivative (TID), and integral tilted-derivative (I-TD) controllers. Simulation results show that the developed STFTID controller significantly enhances the system frequency control under various applied conditions, including multi-step load perturbation, renewable power plant integration, communication time delays, and generation rate constraints.

 In this paper, a single switch transformer-less DC-DC converter with continuous input current for photovoltaic applications is proposed. The suggested configuration utilizes a CL1C2D2 structure to achieve a high voltage gain and alleviate voltage pressure on the semiconductor components. This design enables a significant increase in voltage output with a minimal turns ratio, leading to reduced size, weight, and cost of the converter system. The reduction in turns ratio of the coupled inductor results in lowered voltage stress on the semiconductor elements. Additionally, employing just one power switch in the converter simplifies control and reduces expenses. With its continuous input current, this converter is particularly well-suited for integration in photovoltaic systems. Simulation results conducted using PSCAD/EMTDC software validate the efficacy of the proposed power converter. Furthermore, the maximum power output of the photovoltaic module through an MPPT (Maximum Power Point Tracking) controller under varying irradiance levels is determined, and simulations are executed using PSCAD/EMTDC.

Deriving an accurate and simple current-voltage (I-V) characteristic for photovoltaic (PV) module is highly significant for condition monitoring, fault detection and maximizing power production in PV systems. Equivalent circuits consist of one or more diodes are mostly utilized for I-V curve modelling. However, these models are inherently implicit, relatively complex and nonlinear in their parameters. Here, a piecewise quadratic function with four different intervals is generated from the measured I-V data at standard test condition (STC). The intervals are chosen such that the best model performance can be achieved, especially at maximum power point. Each quadratic function is obtained from least square technique according to the experimental data in the corresponding interval. It is easy to obtain the voltage value at MPP from the extracted model, analytically. Also, a suggestion is provided for extending the generated I-V model to the real environmental condition by utilization of artificial neural network. The derived PV module model is highly suitable for maximum power point tracking, monitoring and fault detection due to its simplicity, explicit structure and accuracy.

This paper presents a new high step-up DC-DC switched-mode converter in which not only the input and output ports can be connected to a common ground, but also the service time of the input source is extended due to continuous current injection to the proposed power converter. Furthermore, the fast dynamic response is offered by minimum phase characteristics. This key feature is achieved by making the proposed converter’s control to output transfer function, free from the right half plane zero (RHPZ). Full description of the operation principles, theoretical analysis related to steady-state operation, and also the small-signal modelling derivation of the proposed converter, are presented in this paper. In the end, to confirm all the merits of the proposed converter and accuracy of theoretical analysis, a sample 25 V - 100 V laboratory prototype DC-DC converter with 100W output power has been implemented, and the main experimental results have also been outlined.

Concerning the increasing application of plug-in electric vehicles (PEVs), planning PEV fast-charging stations (PEVF-CS) has become an important research topic. Regarding the reactive power compensation capability, the optimal planning of PEVF-CS reduces voltage deviation and power loss in the distribution network. Also, one of the basic requirements for expanding electric transportation is the optimal placement of accessible PEVF-CSs, considering the geographic information data. Therefore, the optimal placement of PEVF-CS requires attention to different geographical criteria and power distribution network constraints. In this sense, this paper aims to propose an approach that integrates the Geographic Information System (GIS) technique, Multi-Criteria Decision-Making (MCDM) method, and Mixed-Integer Nonlinear Programming to find the optimal location of a PEVF-CS in Kabul city. The first stage is decision analysis based on the GIS technique and the MCDM approach. The second stage is suitability analysis of the power distribution network constraints to improve power quality. This paper considers ten different suitability criteria, and the Technique for Order Preference Similarity to Ideal Solution (TOPSIS) is applied to rank the different candidate locations. The analysis identified Junction 4 as the optimal choice and demonstrated a significant 3.6% reduction in power loss during peak hours, decreasing from 1071 kW to 1032 kW. These results demonstrate the effectiveness of our approach in optimizing PEVF-CS placement to enhance power quality and reduce the power loss.

The Micro-Expression (ME), which automatically reveals genuine human emotions, has gained significant attention. Recognizing the ME is crucial for many real-time applications. However, there are significant challenges to overcome. For instance, the number of ME frames are limited due to their short duration, and the subtle facial movements can be hard to detect due to their low intensity. These challenges need to be addressed to improve ME recognition. We propose a novel method for the ME recognition in real-time. In this method, first, the apex frame is spotted using the rotated local binary pattern from six planes (RLBPS) and correlation coefficient (CC). Next, three hand-crafted methods such as the multi-color rotated local binary pattern from six planes (MRLBPS), the histograms of directed gradients from six planes (HDGS), and the histogram of image gradient direction from six planes (HIGDS) extract the features from the apex frame and its surrounding frames. Finally, the stacks of features as matrixes are fed into a three-dimensional convolutional neural network (3D-CNN), and the output is the maximum recognition rate by voting three results. The proposed method has shown promising results when compared to most state-of-the-art methods. According to the results, an average precision of 99% has been obtained using our proposed method. The combination of the RLBPS and the CC creates a strong method for spotting the apex frame. Also, feeding the stacks of spatiotemporal features into the 3D-ResNet increases the ME recognition rate in real-time.

The delay-and-sum (DS) beamforming and delay-weight-and-sum (DWS) beamforming are primary methods in ultrasonic imaging with phased arrays. Total focusing method (TFM) and Minimum Variance (MV) based adaptive beamforming are well-known methods within DS and DWS beamforming, respectively. The MV-based adaptive beamforming significantly reduces interferences and provides high-resolution image compared to TFM beamforming, at the cost of high computational complexity, and sensitivity to input statistics for matrix inversion. To address these challenges, recently, iterative MV (IMV) has been proposed to alleviate computational burdens without the need for matrix inversion. The delay-multiply-and-sum (DMAS) beamforming enhances TFM beamforming performance by employing spatial information of the array signals. However, the resulting images remain susceptible to speckle and background noises in all of these beamformers. In this paper, we aim to improve these beamforming methods so that speckle and Gaussian background noise are reduced while preserving the quality of the reflective echoes in ultrasonic images. In the proposed method, a wavelet transform with a novel threshold function is applied to the received signals to initially reduce the noise, followed by the application of the beamformer. Subsequently, the coherence weighting using the denoised signals is derived, and the obtained coherence weighting is then integrated into the beamforming process. Simulation results demonstrate that the proposed method achieves a significant reduction in background noise and speckles of the above beamformer, and particularly reduces background noise and speckles of beamformer up to approximately -27dB while preserving the detection capability of reflective points.