It is shown in [1] that the optimal downlink radio resource allocation for non-realtime traffic in cellular CDMA/TDMA networks can be mapped to a Multi-dimensional Multiple-choice Knapsack Problem (MMKP) which is NP-hard. In this correspondence we propose a heuristic algorithm with polynomial time complexity for this problem. Numerical results indicate significant computational performance improvement in comparison to existing heuristic algorithms for MMKP.

We propose a novel algorithm called RCA_MC, in which we use the breadth first search method (BFS) in conjunction with edge contraction and connectivity properties of a given undirected graph to enumerate and scan all its minimal edge cutsets. It is known that the problem of enumerating all minimal edge cutsets of a given graph is #P-complete. In addition, we introduce the concepts of pivot vertex and absorbable clusters, and use them to develop our enhanced recursive contraction for scanning all mimimal edge cutsets, called ERCA_MC, of a given graph. Simulation results provide empirical evidence that the complexity of the ERCA_MC algorithm is linear per cutset.

در این مقاله، مدل جدید برنامه ریزی غیرخطی عددصحیح مختلط برای مساله مکانیابی- موجودی مستقیم/ معکوس با ظرفیت محدود ارایه شده است، که تصمیمات استراتژیک را در کنار تصمیمات تاکتیکی بهینه می کند. مدل پیشنهادی دوهدفه بوده و هزینه های کل را به عنوان تابع هدف اول و کمبود را به عنوان تابع هدف دوم کمینه می کند. با حل یک مثال عددی، برتری مدل یکپارچه بر مدل غیریکپارچه اثبات شده است و تحلیل حساسیت نیز جهت اعتبارسنجی مدل پیشنهادی انجام شده است.  با توجه به ماهیت NP-hard[1] مساله و وجود دو تابع هدف که در تضاد با یکدیگر هستند، دو الگوریتم فراابتکاری به نام های الگوریتم ژنتیک با مرتب سازی نامغلوب نوع دوم[2] و الگوریتم تکاملی پارتوی قوی نوع دوم[3] برای حل مسایل در ابعاد بزرگ استفاده شده است. تحلیل معیارهای ارزیابی مختلف معرفی شده گویای عملکرد بهتر الگوریتم تکاملی پارتوی قوی نوع دوم نسبت به الگوریتم ژنتیک با مرتب سازی نامغلوب نوع دوم برای حل مسایل در ابعاد بزرگ می باشد.
 
[1] Non-deterministic Polynomial-time hard
[2] Non-dominated Sorting Genetic Algorithm-II (NSGA-II)
[3]Strength Pareto Evolutionary Algorithm-II (SPEA-II)

In this paper, a new algorithm based on processing differential current harmonics is proposed for digital differential protection of power transformers. This algorithm has been developed by considering different behaviors of second harmonic components of the differential currents under fault and inrush current conditions. In the new method, a criterion function is defined in terms of the time variation of the second harmonic rising rate during instants after the occurring disturbance. By evaluating the signs of the criterion function for the three phases, the internal faults can be accurately recognized from inrush current conditions in less than a half-cycle after the occurrence of a disturbance; this is one advantage of the method. Another advantage of the proposed method is that the fault detection algorithm does not depend on the selection of thresholds. A suitable performance of this method will be demonstrated by the simulation of different faults and switching conditions on a power transformer. For this purpose, a small part of the Iran power system involving a power transformer and the transmission lines on both sides of the transformer has been considered. To include effective factors on differential current components, the elements of this power system have been precisely modeled in PSCAD/EMTDC.

این مقاله، عملکرد الگوریتم (Simulated Annealing) SA و (Genetic Algorithm) GA را در تعویض پیش گیرانه بهینه قطعات به منظور حداقل کردن زمان خوابیدگی بررسی می کند. به این منظور، تعدادی معیار ارزیابی برای تحلیل عملکرد این الگوریتم ها تشریح شده تا با استفاده از آن ها بتوان تصمیم گرفت که کدام الگوریتم را در تعویض پیش گیرانه قطعات می توان به کار برد.

Genetic Programming (GP) is a powerful machine learning technique derived from genetic algorithms. We used GP to generate a mathematical function for image denoising based on statistical features derived from detail sub-bands of wavelet transform (WT). The function obtained from GP for image denoising is not dependent to any parameters as represented in other image denoising methods based on WT. Results of the proposed image denoising method is compared to the VisuShrink soft threshold image denoising method, both perceptually and in terms of Peak Signal to Noise Ratio (PSNR).