فصلنامه هوش محاسباتی در مهندسی برق
سال یازدهم شماره 2 (تابستان 1399)

Expansion of distribution networks to increase the reliability due to change of behavior and type of customers is a requirement. On the other side, the concern about environment and climate changes has led to use of renewable energy sources as a replacement for fossil generation sources. DGs based on RES and control switches can play an important role in improving the reliability of the grid, decreasing the costs and mitigating environmental effects. The main purpose of this article is evaluating the remote control switches and manual control and sizing of RES-based DG units including solar cells and wind turbines and simultaneous placement of these units considering their generation uncertainty. The intended function is a bi-objective function that includes reducing network costs and improving the Side Reliability Index and technical and economic constraints are considered too. The intended function is modeled as a mixed integer nonlinear programming and four evolutionary algorithms are used to solve this problem and their results are compared. Roy Billinton 4-bus network is chosen as the test network in this article.

Today, dual-input DC-DC converters are used for various purposes, including simultaneous use of two sources to solve the challenge of the intermittent nature of the renewable energy sources. In this paper, a novel structure of a semi-isolated dual-input dc-dc converter based on the impedance network along with a controlling approach is proposed. In the proposed converter, quasi-Z source network has been utilized and the voltage gain has improved dramatically with the use of switched capacitors and coupled inductor techniques. Due to the necessity of an appropriate controller to improve the performance indicators of these type of converters, a two-layer intelligent adaptive control structure is proposed. In the first layer of the controller, the proposed converter is investigated based on the analysis of multivariate systems, and then in the second layer, in order to achieve a robust performance for the converter responses against the uncertainties such as abrupt inputs voltage change and disturbances, a self-tuning adaptive controller is used. Also, the proposed controller structure has fast transient convergence in different situations for output voltage stability, which makes it a good candidate for renewable energy sources and fuel cells. Simulation results from MATLABSimulink have examined the performance of the proposed topology. Finally, experimental evaluation validates the claimed advantages.

In this work, a method for order reduction of unstable systems by minimizing the multi-objective weighted sum function ITSE and H∞ norm with bat algorithm is investigated. At first, the unstable model is divided into two parts: stable and unstable. Then, the order of reduced model is determined by considering the stable part and using Hankel norm. In the following, a fixed structure with unknown coefficients is considered for a reduced order model. By minimizing the weighted sum multi-objective included ITSE and H∞ norm with bat algorithm, in which the error is the difference between the step response of original system and reduced order model, unknown parameters of the reduced order model are determined. The Routh-Hurwitz criterion has been used to satisfy the stability condition, which is introduced as a constraint in the optimization problem, and thus the unconstrained optimization problem is converted to constrained optimization problem. Finally, the unstable part is joined to the reduced order part. To confirm and demonstrate the capability and efficiency of the proposed method, three large systems have been reduced and compared with some order reduction methods such as balanced truncation and optimal Hankel approximation. The results show the capability of the proposed method over classical methods.

The increase in power generated by using solar energy increases the uncertainty in the power grid, since the majority of meters measure only the net load of a grid regardless of the output of distributed generations. This paper proposes a framework for probabilistic mid-term net load forecasting in a power grid based on separate forecasts of the load and output power of a solar station using the combination of principal components analysis and the extreme learning machine methods. The data used in this paper is related to the NERL and GEFCom2014 data bases and the matrix of scores is extracted by the use of principal component analysis. The prediction models are trained using ORELM model and evaluated in three sections: training, validation and mid-term prediction. The main objective of the proposed method is to increase the precision of net load forecasting by improving point forecasts. The comparison between the results presented in this thesis with other references shows that the MAPE error of predicted load, and predicted output power of the solar station improved  to 1.1333 and 0.3118, respectively, which will reduce overall forecast error.

Power switches are among the equipment that can help improve the distribution system condition and reduce the cost of failure at times of error. Finding the optimal location and number of these switches in the distribution system can be modeled with a variety of objective functions as a nonlinear optimization problem to improve reliability, reduce energy not supplied and other system costs. In this paper, a Hybrid Particle Swarm Optimization with Sinusoidal and Cosine Acceleration Coefficients (H-PSO-SCAC) is used to solve the problem of optimal placement of switches with the goal of reducing the cost of switches installation and outage simultaneously in a software format for the use of Malayer Distribution Company operators and other sections. The simulation results have been tested using the corresponding algorithm on the Malayer 108-bus distribution feeder and IEEE 118-bus sampling network and compared with the results of other methods. The results show that the algorithm (H-PSO-SCAC) is better than other algorithms in terms of response quality and computational efficiency, also its implementation in a user friendly software has made it easy for operators and Users to use.

Nowadays, linear machines have more developed for high power applications such as long-distance transportation. But, their expensive investment and some technical problems are major challenges of these technologies. Here, a linear flux switching machine (LFSM) with segmented secondary is studied. These fluctuations, in other words, detent force have a negative impact on machine performance. The effect of these issues can be notably reduced with a suitable design. In this paper, a FEM-based Analytical model is presented for a linear flux switching machine. The results of analytical model are verified by FEM Simulations and then a control method is proposed to reduce fluctuations of the thrust force.

In this paper, a bi-level model is proposed for low and medium voltage distribution network expansion planning in the presence of distributed generations (DGs) which are located in the low-voltage network. In the bi-level model, the upper and lower levels are medium and low voltage networks, respectively. The model is decomposed to two sub problems and the final aim is to find the answer with minimum planning and operation costs consisting of the cost of installing and upgrading of new and existing transformers, substations, feeders, distributed generations and cost of losses. The model is a mixed-integer non-linear programming, which is solved by a genetic algorithm with a proposed codification. To show the effectiveness of the proposed model and solution method, the model is implemented in a distribution system in three case studies, including, problem solving with traditional planning, bi-level programming without considering DGs and bi-level programming considering DGs in the low-voltage side. The obtained results show that the importance of incorporating both low and medium voltage networks simultaneously is very effective in reducing costs.

In this paper, a new control strategy based on Fuzzy Cognitive Maps (FCMs) in order to design a supervisory controller for a complex system is presented. This supervisory controller is designed to provide reference signals needed for local controllers used in different sections of complex system. Designing of proposed supervisory controller consists of two part: 1- First, a basic FCM is constructed based on the cause and effect relationships between various components of the complex system, 2- Then, according to the control needs, new concepts are added to this cognitive map, and the relationship between these new concepts and the concepts of the primary fuzzy cognitive map is optimized based on a functional index. The proposed control approach is applied to energy management of a series-parallel plug-in hybrid electric vehicle problem. The simulation results show that the proposed control strategy has a better performance in satisfying the control objectives for a complex system than the conventional fuzzy controller.