نشریه محاسبات نرم
سال دوازدهم شماره 1 (پیاپی 23، Spring-Summer 2023)

The increasing reliance on electric power and the environmental pollution caused by fossil fuels has created a need for new energy sources for electric power production. Renewable energy sources such as wind and solar can be employed to produce electric power, however, their output powers are unpredictable due to stochastic environmental situations. These changes lead to frequency deviations in the power grid, potentially making it unstable. This issue can be more challenging in standalone microgrids since they have low inertia. To overcome this challenge, energy storage systems (ESSs) can be used, although they require significant investment and may not always be cost-effective. Electric vehicles (EVs) can help power systems balance generation and consumption and compensate for renewable energy output changes. This is achieved through the EV batteries, which can be charged when the grid frequency is high and discharged when it is low, a concept known as Vehicle to Grid (V2G). In this paper, we present a new method for controlling EVs in a microgrid in order to reduce frequency deviations. To this end, we introduce a fuzzy controller with optimized membership functions and rules. In the proposed method, the state of charge (SOC) of an EV battery can be controlled while regulating frequency. Simulations conducted in the MATLAB environment demonstrate the effectiveness of the proposed method.

Delay fault simulation is the most general method that is used to assess the quality of generated test sets. Path delay fault is one of the most frequently used delay fault models. Path delay fault simulation is a time-consuming operation, especially for today's complex digital circuits. In this work, a novel critical path tracing algorithm is proposed for parallel path delay fault simulation. The obtained outcomes denote 489 times average speedup compared with the traditional path tracing, as well as 186 times average speed-up in comparison with the latest reported results of previous studies.

Due to the high computational cost of the direct numerical simulation methods of the governing equations of some natural phenomena, surrogate models based on machine learning methods such as deep learning algorithms have been commonly interested in modeling these phenomena. This paper proposes a reduced-order model based on a deep-learning algorithm to simulate temperature changes in a two-dimensional field. This model is developed using three different methods, including a framework based on convolutional neural networks, a physics-informed loss function of the phenomenon, and a reduced-order model using the autoencoder method. The model outcomes were compared with the results obtained from a high-resolution finite difference method. The results show that the reduced-order model (with an accuracy of 2.528×10-6 °C) has higher accuracy than the other two models. Meanwhile, the Model-based physics-informed loss is superior to the other two models in terms of steady-state temperature data consumption (only 400 data of size 8×8).

Identifying retinal blood vessels is widely used for diagnosing eye diseases such as diabetic retinopathy, and glaucoma. Currently, doctors manually extract these vessels, which is a challenging and time-consuming process that often leads to errors. In this paper, a new method is proposed for retinal blood vessel extraction, which includes three basic parts. First, the noise in the image is removed. Next, the center lines of the vessel are extracted. Finally, the blood vessels of the retinal images are extracted using the area expansion and noise removal method. The proposed method is applied to the images of the DRIVE test set and its efficiency is evaluated using four different metrics: sensitivity, specificity, accuracy, and precision. The average results for accuracy, specificity, sensitivity, and accuracy in the proposed method are 0.92896, 0.98965, 0.91756, and 0.96578, respectively.

Nowadays, medical intelligence detection systems have evolved significantly due to advancements in artificial intelligence, however, they face some challenges. Breast cancer diagnosis and classification is one of the medical intelligence systems. There are a variety of screening techniques available to detect breast cancer such as mammography, magnetic resonance imaging, and ultrasound. This research uses the MIAS mammography image dataset and tries to diagnose and classify benign and malignant masses based on image processing and machine learning techniques. Initially, we apply pre-processing for noise reduction and image enhancement using Quantum Inverse MFT, and then image segmentation with the Social Spider Algorithm. The type of mass is then diagnosed by the Convolutional neural network. The results show that the proposed approach has better performance in comparison to others based on some evaluation criteria such as accuracy of 99.57%, sensitivity of 91%, and specificity of 86%.

In the rapidly evolving landscape of web services, efficient interaction and optimal selection among services with different quality parameters are essential. This paper addresses the complex challenge of selecting candidate services for abstract services within probabilistic graph structures. We propose a new hybrid method that combines node-based and path-based graph simplification techniques, allowing for the identification of new patterns, such as parallel and nested loops. We use NSGAII to improve scalability and accuracy in service selection. The proposed approach simplifies the composition graph while optimizing the selection process by considering important quality parameters such as execution cost, response time, and availability. Through systematic simplification and a robust fitness function, we ensure a definitive and accurate response to user queries. The results show significant improvements in the proposed approach compared to existing methods, making it a comprehensive solution for effectively composing web services in dynamic environments.

In this paper, an epidemic model with fuzzy parameters for spreading COVID-19 in a population is considered. The sensitivity analysis is used to determine the model robustness to parameter values of the model. The basic reproduction number of the epidemic model denoted by R_0 determines the dynamics of the model. Then, in order to examine the relative importance of different parameters in the COVID-19 spread, we derive an analytical expression for the sensitivity of the basic reproduction number R_0, namely sensitivity index, with respect to each parameter involved in the model. Finally,  sensitivity analysis results and the numerical simulations of the model are given with different parameter values.

In the age of information technology, analyzing data in the cloud through efficient resource management is recognized as an effective solution to meet the quality-of-service needs of users. In this regard, this paper proposes a fuzzy selector-based approach to improve the virtual machine (VM) migration process that provides a balance between servers during the processing of tasks. The proposed approach is a hierarchical resource prediction that includes local and global parts for processing requests. Evaluations demonstrate the superiority of the proposed approach over state-of-the-art methods. The results show that the proposed approach reduces the cost by 10% compared to FAHP.

According to today's statistics, more than half a billion vehicles move around the world, making inspection and monitoring one of the basic needs of any traffic control system. All vehicles have an identification number exhibited as the license plate, their primary ID, a vital element. Deep Learning methods are adopted to detect vehicle license plates. This proposed method consists of two steps: highlighting the license plate and reading the ID stages. In this context, the combination of deep neural networks (DNN) and the competitive generative adversarial network (GAN) is applied in the encoding-coder network/structure for this highlighting. The proposed models are assessed based on the FZU Cars and Stanford Cars datasets, to which the results of this study are compared and discussed. The findings here indicate that the accuracy of this proposed model is almost 98%, subject to the two datasets.

In this paper, an adaptive network fuzzy inference system (ANFIS) based on the Takagi-Sugeno- Kang technique is used for predicting the effective length of vertical rods buried in two-layer soils. The rods are subjected to two typical lightning return stroke currents namely first and subsequent stroke currents. To train the ANFIS approach, a number of input-output pairs are computed from the multi-conductor transmission line method. The inputs are resistivity values of the upper and lower layers, upper layer thickness and the rise time of the lightning current. After the training process is converged, the prediction of effective length is efficiently carried out in such soils. Also, the comparative study with the horizontal electrode buried in two-layer soils shows that the effective length of vertical rods is considerably less than that of the horizontal electrodes which are financially and practically important, whereas in single-layer soil they are different.