Scientia Iranica
Volume:29 Issue: 2, Mar & Apr 2022

In the rapidly defiled environment, cancer has emerged out as the most threatening disease to human species. Therefore, a robust classification model is required to diagnose cancer with high accuracy and less computational complexity. Here, random parameters of Extreme Learning Machine (ELM) are optimized by Self Adaptive Multi-Population-based Elite strategy Jaya (SAMPEJ) algorithm. This strategy constructs a robust ELM classifier named as SAMPEJ-ELM model. This model is tested on Breast cancer, Cervical cancer and Lung cancer datasets. Here, a comparative analysis is presented between the proposed model and basic ELM, Jaya optimized ELM (Jaya-ELM), Teaching Learning Based Optimization (TLBO) optimized ELM (TLBO-ELM), SAMPEJ optimized Neural Network (SAMPEJ-NN), SAMPEJ optimized Functional Link Artificial Neural Network (SAMPEJ-FLANN) models. Numerous performance metrices viz. accuracy, specificity, Gmean, sensitivity, F-score with receiver operating characteristic (ROC) curve are used to estimate the proposed model. Moreover, this model is compared with eleven existing models. SAMPEJ-ELM model resulted the highest degree of accuracy, sensitivity and specificity in Breast Cancer (.9895, 1, .9853), Cervical Cancer (.9822, .9948, .9828), Lung cancer (.9787, 1, 1) datasets. The experimental results reveal that SAMPEJ-ELM model classifies both the positive and negative samples of cancer datasets significantly better than others.

A well-planned humanitarian logistics involving rescuing people and providing on-time lifesaving facilities to disaster-affected areas can significantly mitigate the aftermath of disasters. However, damaged bridges and blocked roads can hinder last-mile deliveries in disaster-affected areas by ground vehicles only. So, in this paper, we propose a ground vehicle (GV) and unmanned air vehicle (UAV) collaborative delivery system in such areas. Here, a fleet of homogenous ground vehicles each equipped with a certain number of UAVs is deployed for last-mile deliveries. UAVs make the flight from GVs, deliver to end locations and return to the GV for battery replacement and/or start another flight. The objective of the model is to minimize the total delivery time within UAV flight endurance and payload constraints. Firstly K-means clustering algorithm has been used to cluster the disaster-affected region into different sectors. Then GV_Touring and UAV_Routing have been scheduled using nearest neighbor heuristic to serve ground approachable locations and UAV served locations respectively. Finally, the random walk based ant colony optimization-based (ACS_RW) has been developed to further optimize the overall travel time. Experimentation results show the potential benefits of the proposed algorithm over other available truck-drone collaborative transportation models.

This paper presents sub-transmission and distribution network expansion planning (S&DEP) including distributed generation (DG) and distribution automation (DA) considering reliability indexes. The objective function is to minimize investment, operation, maintenance and reliability costs subjected to AC power flow, system operation and generating units and DG limits, reliability, and distribution automation constraints (including the constraints of protection devices and volt/VAr control mechanism). The proposed model is a mixed integer non-linear programming (MINLP) model which is hard to solve. For this reason, a MINLP problem is transformed to mixed integer linear programming (MILP) model. The validity of the proposed method is investigated in the two synthetic test networks.

This study presents a set of rules for optimal tuning a class of integer-order controllers, known as implementable fractional-order PID controllers, to be applied in control of first-order-plus-dead-time (FOPDT) processes. To this aim, the approach of so-called “tuning based on the implementable form of the controller” is applied instead of the common approach of “tuning based on the ideal form of the controller”. Consequently, no contradiction is found between the behavior of the tuned controller and that of the implemented controller. Also, algebraic relations between the values of cost functions, which are defined based on integral square error (ISE) and integral square time error (ISTE) performance indices, and free parameters of the implementable controller are established. Tuning implementable fractional-order PID controllers via the proposed rules guarantees that the values of performance indices are reduced in comparison with the case of using optimal PID controllers. In addition to numerical results, experimental results are also provided to demonstrate the effectiveness of the proposed tuning rules in practical applications.

Economic Load Dispatch (ELD) is an important part of cost minimization procedure in power system operation. Different derivative and probabilistic methods are used to solve ELD problems. This paper proposes a powerful Salp Swarm Algorithm (SSA) to explain the ELD issue including equality and inequality restrictions. The main aim of ELD is to satisfy the entire electric load at minimum cost. The SSA is a population based probabilistic method which guides its search agents that are randomly placed in the search space, towards an optimal point using their fitness function and also keeps a track of the best solution achieved by each search agent. SSA is being used to solve the ELD problem with their high exploration and local optima escaping technique. This algorithm confirms that the promising areas of the search space are exploited to have a smooth transition from exploration to exploitation using the movement of Salps in the sea. Simulation results prove that the proposed algorithm surpasses other existing optimization techniques in terms quality of solution obtained and computational efficiency. The final results also prove the robustness of the SSA.

The main contribution of this paper is to present the systematic and low-complexity translation techniques betweena class of hybrid systems referred to as automaton-based DHA and piecewise affine (PWA) systems. As an startingpoint the general modeling framework of the automaton-based DHA is represented which models the controlled anduncontrolled switching phenomena between linear continuous dynamics including discrete and continuous states,inputs and outputs. The basic theoretical definitions on the state trajectories of the proposed DHA with forwardand backward evolutions which yield forward and backward piecewise affine (FPWA and BPWA) systems are given.Next, the well-posedness and equivalency properties are proposed and the sufficient conditions under which the wellposedness property is achieved with the automaton-based DHA and PWA systems are given. It is shown that thegraphical structure of the proposed automaton-based DHA makes it possible to obtain analytically the equivalent PWAsystem with a polynomial complexity in contrast to the existing numerical translation techniques via decomposedstructure of the DHA with an exponential complexity. Examples are presented to confirm the effectiveness of theproposed translation techniques.

This paper focuses on a dynamic equilibrium considering the flexible ramp market and demand response resources. With ever-swelling installation of variable renewable energies, demand response programs can play an important role in mitigating the system ramping deficiency. Hence in this paper, the ramping capability of demand response resources in procuring system ramp requirement is considered. The strategic behavior of different players is modeled through a multi-leader-common-follower game, in which suppliers and demand response aggregators are laid as the leaders and market operator is considered as the single follower of the game. In addition, a dynamic forward rolling process to find equilibria at the real-time market is proposed. The effect of considering demand response resources and flexible ramp penalty price on the strategic behavior of players in equilibrium is evaluated. Finally, the effectiveness of the proposed approach is verified on a three-firm system. While revealing demand response resources roles in mitigating ramping deficiency, the results show that how penalty price on flexible ramp violation can lead uplift payments to be formed.

In this paper, three active-C synthetic grounded inductance simulator circuits are presented, which realize tunable lossless and lossy series and parallel RL-type inductances. Each of which employs two voltage differencing buffered amplifiers (VDBAs) as active components, and a single grounded capacitor as a passive component. In all the proposed circuits, the simulated equivalent resistance and inductance values can be adjusted electronically through the transconductance gains of the VDBAs. They also do not require any critical component matching conditions and cancellation constraints. Detail non-ideal analysis including transfer errors of the VDBA has been analyzed. For circuit performance verification and comparison, some application examples are given together with computer simulation results by PSPICE program.

This paper presents solution of short-term hydrothermal scheduling problem using an algorithm called Grasshopper optimization. The objective of hydrothermal scheduling is to reduce the total cost of generation by optimizing the output of power generation of different thermal and hydro plants for a certain time interval. A non-linear relationship between hydropower generation, net head and rate of water discharge is considered here. The complicated head-sensitive water-to-power conversion and piecewise output limit is also considered. To investigate the performance of this new technique, two test systems have been considered. The simulation results are compared with some well-known optimization methods such as Genetic algorithm, Biogeography based optimization, hybrid Differential evolution with Biogeography based optimization and Grey wolf optimizer algorithm. The simulation results show the superiority of this technique as compared to other well-known optimization methods.