فصلنامه هوش محاسباتی در مهندسی برق
سال چهاردهم شماره 1 (بهار 1402)

In the islanded microgrids, distributed energy sources are responsible for power-sharing, and accurate power-sharing is essential among these microgrids. Due to the mismatch of the physical impedance of power lines, poor reactive power-sharing is observed. Therefore, in this paper, the use of virtual impedance in the form of two methods of virtual positive impedance and virtual negative impedance is proposed to improve the accuracy of reactive power-sharing in a conventional droop controlled islanded microgrid. In this paper, in addition to presenting a controller to properly adjust the virtual impedance, the concept of virtual negative impedance is used to correct the effective impedance of the power supply line and improve the common bus voltage, and increase the output power of energy sources. To evaluate the proposed control methods, a test islanded microgrid is simulated in PSCAD/EMTDC software. The results show that both proposed methods have a good effect on the proper reactive power-sharing in the microgrid. In addition, the results show that the virtual negative impedance method can improve the common bus voltage and increase the output power of the source by reducing the effective line impedance.

In this paper, the Hunger Games Search (HGS) optimization algorithm is used to estimate the electrical parameters of the induction machine steady state model. Induction machine nameplate data is used as input to the proposed algorithm. The performance of the proposed method is confirmed by comparing the output characteristics obtained by estimating the motor parameters including torque, current and power factor in the steady state model of the induction machine with the values provided by the manufacturer. In addition, by evaluating and comparing the results of the proposed method with the results of previous research, it is shown that the proposed algorithm is a very effective and accurate method for the acceptable estimation of induction machine parameters.

Brain tumor segmentation is an important step in the diagnosis and treatment planning of cancer patients. The procedure of manual brain tumor segmentation suffers from a long processing time. In this light, automatic brain tumor segmentation is highly appealing in the clinical routine. This study sets out to segment the tumors from brain MR images and to investigate the effectiveness/usefulness of the different MRI sequences for this purpose. Here, the MR images from the BRATS challenge were utilized. 310 patients with four different MRI sequences, including T1, T1ce, T2, and FLAIR were employed to train a ResNet deep CNN. Four separate models were trained with each of the input MR sequences to identify the best sequence for brain tumor segmentation. To assess the performance of these models, 60 patients (external dataset) were quantitatively evaluated. The quantitative results indicated that the FLAIR sequence is more reliable for automatic brain tumor segmentation than other sequences with an accuracy of 0.77±0.10 in terms of Dice compared to Dice indices of 0.73±0.12, 0.73±0.15, and 0.62±0.17 obtained from T1, T2, and T1ce sequences, respectively. Based on the results of this study, FLAIR is a more reliable sequence than other sequences for brain tumor segmentation.

In this paper, the problem of improving the stability of the power system with DFIG and in the presence of SSSC using nonlinear method is discussed. The nonlinear controller is designed by the multi-input backstepping method with a sliding mode observer. This controller is applied simultaneously to the excitation system of synchronous generators and the rotor side converter in DFIG and SSSC in a way that improves the stability of the power system compared to the linear and nonlinear methods described in this paper. The control coefficient matrices are adjusted using intelligent algorithms in such a way that the stability of the system is more optimized. Practical constraints on the system are considered in selecting the control inputs. The designed controller is robust to changes in the operating point and the location of the disturbance. The performance of the designed controller in a 39-bus, 10 machines NEW ENGLAND network including DFIG generators and in the presence of SSSC is simulated and investigated using MATLAB software.

After extreme natural events like earthquakes, storms, and floods, the electrical distribution network may be isolated from the upstream grid. In such a situation, the electrical power of the upstream grid is lost and the available microgrids (MGs) are the only power sources. Since the power outputs of MGs are restricted, restoring all de-energized loads is impossible and restoring critical loads (CLs) becomes the most important concern of the distribution network operators. Therefore, in this paper, a new hour-by-hour restoration strategy for CLs restoration after a blackout is proposed. The desired objective functions include restored energy and switching operations. The proposed restoration strategy presents a restoration plan for each hour of the outage period. The efficiency of the proposed strategy is carried out on an IEEE 123-bus distribution network and the simulation results confirm the superiority of the proposed hour-by-hour restoration strategy over the other conventional restoration methods.

Today, the new coronavirus (Covid-19) has become a major global epidemic. Every day, a large proportion of the world's population is infected with the Covid-19 virus, and a significant proportion of those infected dies as a result of this virus. Because of the virus's infectious nature, prompt diagnosis, treatment, and quarantine are considered critical. In this paper, an automated method for detecting Covid-19 from chest X-ray images based on deep learning networks is presented. For the proposed deep learning network, a combination of convolutional neural networks with type-2 fuzzy activation function is used to deal with noise and uncertainty. In this study, Generative Adversarial Networks (GANs) were also used for data augmentation. Furthermore, the proposed network is resistant to Gaussian noise up to 10 dB. The final accuracy for the classification of the first scenario (healthy and Covid-19) and the second scenario (healthy, Pneumonia and Covid-19) is about 99% and 95%, respectively. In addition, the results of the proposed method in terms of accuracy, precision, sensitivity, and specificity in comparison with recent research are promising. For example, the proposed method for classifying the first scenario has 100% and 99% sensitivity and specificity, respectively. In the field of medical application, the proposed method can be used as a physician's assistant during patient treatment.

These days, one of the biggest challenges in the world is dealing with the outbreak of the Covid 19 virus. Recently, new variants of this virus have been identified that have a much higher rate of transmission. To effectively control and manage the spread of the disease, a clear understanding of its transmission dynamics and effective control techniques to reduce or inhibit the spread of the virus is necessary. Although vaccine production and distribution are currently underway, Non-Pharmacological Interventions (NPI) continue to be an important and fundamental strategy for controlling the spread of the virus in various countries around the world. In this paper, Covid 19 dynamics is modeled using four well-known categories (SEIR): Susceptible-Exposed-Infected-Recovered. Since the parameters of the model have uncertainty, a robust control method should be designed. In this paper, using fractional calculus and fuzzy logic, a robust fuzzy fractional-order sliding mode controller (FOFSMC) for Covid 19 dynamics is proposed, which aims to control the prevalence of the disease using NPI. The proposed method is implemented both on the integer and fractional order model. Finally, the performance of the proposed controller on the new variant of the Covid 19 virus with a faster disease transmission rate will be evaluated.

Prediction with high reliability is very important in solving real-world problems, especially those that affect public health. The statistical properties of complex problems such as Covid-19 disease constantly change over time which makes modeling of such problems associated with high-level uncertainty. It has been proven that the type-2 fuzzy logic has the potential for modeling uncertainty to solve complex problems. In this research, for the first time, an intelligent method based on the capability of type-2 fuzzy logic was presented to manage uncertainty in predicting short-term and long-term time series in environmental crises such as the Covid-19 pandemic. The performance of the proposed model was evaluated using a real dataset collected from official sources. The results confirm the high efficiency of the proposed method on Covid-19 based on a ROC curve analysis. The obtained results showed an efficiency of 93.81% for short and 91.33% for long-term time series. This indicates the high efficiency and capability of the proposed model for managing uncertainty in predicting patterns of Covid-19 in comparison with similar methods. The proposed model can be useful to take strategic decisions and prevent the consequences of the Covid-19 epidemic in the short and long terms.

Anomaly detection has been in researchers’ scope of study for a long time. The wide variety of anomaly detection use cases ranges from quality control in production lines to providing security in public places. One of the most attractive topics in anomaly detection is in video surveillance systems. In this paper, we propose a method that works based on frame prediction and optical flow to improve anomaly detection in videos. The use of optical flows in normal frames helps the system to better detect the entrance of people or objects to forbidden areas by its information about the amount of movement in different regions of the frames. Based on the optical flow of normal videos and that of current video, the threshold for anomaly decision is adaptively adjusted. This could ultimately lead to a better overall performance of the anomaly detection system compared to the recent similar works. The presented method is general and can be simply incorporated to other video anomaly detection systems to improve the detection accuracy.

Cognitive neuroscience is useful for understanding human behaviors related to marketing and adapting to consumer preferences. By analyzing consumers' brain responses to marketing stimuli, researchers seek to discover the reasons for decision-making. This study proposes a framework for participants' decision-making processes in terms of liking and disliking when viewing and selecting the products of an online store. To this end, the participants' brain signal (EEG) is used when displaying different products. Estimation of power spectrum density by Welch method, detrended fluctuation analysis (DFA), and recurrence quantification analysis (RQA) were used to extract the feature vector. The results show that the two categories of liking or disliking a product can be classified with 73.5% accuracy using a support vector machine (SVM), which compared to the previous study, there is a 3.5% improvement in results. By better understanding consumer behavior and mastery of consumer demands, market strategies can be determined in a way that in addition to customer satisfaction, increase sales and profits. The results are promising and the proposed method can be used for a better electronic commerce model.