multi-objective genetic algorithm
در نشریات گروه فنی و مهندسی-
Banking, a vital economic pillar worldwide, thrives with effective management, aiding economic growth. Mitigating risks and addressing cost control are key challenges. Prioritizing strategies to enhance performance in both risk management and cost efficiency is crucial for the banking sector's success and economic stability. One approach is to select partners in such a way that the risk of bank insolvency and total costs are reduced, and the capital adequacy of the bank is increased. So, in this work, we first created a mathematical model to achieve the above goals in the field of banking using the approach of selecting partners. In this model, three objective functions are considered for the optimal selection of partners, two of which aim to minimize risk and cost, and the last objective is to maximize capital adequacy. To solve this multi-objective model, we implemented an integrated intelligent system. A combination of a multi-objective genetic algorithm and a neural network was used in this system. A multilayer perceptron neural network is used to calculate the nondeterministic parameters based on the data from different periods. The proposed method was evaluated using a numerical example in MATLAB software. The obtained results and their comparison with one of the classic algorithms show the superiority and reliability of this intelligent system. Using this system, the optimal partners can be selected to achieve the set goals. The most important factors in the field of risk have been identified. Then, a meta-heuristic multi-objective algorithm (NSGA-II) along with an intelligent neural network system has been used to optimally select partners. According to this intelligent system, a suitable methodology is presented along with the optimization algorithm.Keywords: banking, Banks Insolvency Risk, Selecting Partners, multi-objective genetic algorithm, multilayer perceptron neural network
-
امروزه مساله مکانیابی مخازن زباله و مسیریابی خودروهای جمع آوری پسماند در ارتباط نزدیک با سلامت جامعه و همچنین محیط زیست دارد. در این مقاله، مساله مکانیابی- مسیریابی کمان دوره ای همراه با ایستگاه های تخلیه میانی در شرایط عدم قطعیت مورد بررسی قرار می گیرد که هدف آن کاهش تعداد مکان های فعال جهت جمع آوری پسماند در سطح شهر بوشهر، تعیین مسیرهای بهینه جهت سرویس دهی تمامی یال های دارای تقاضای شبکه گراف شهری در طول هفته و تعداد وسایل نقلیه موردنیاز است. یک مدل برنامه ریزی خطی عدد صحیح مختلط همراه با در نظر گرفتن تقاضای فازی برای بهینه سازی مساله توسعه داده می شود و برای حل تقریبی مساله از در ابعاد بزرگ الگوریتم ژنتیک چند هدفه استفاده می شود. برای ارزیابی کارایی الگوریتم پیشنهادی، از حل کننده CPLEX نرم افزار GAMS در حل مسائل با ابعاد کوچک و متوسط بهره برده می شود. در پژوهش حاضر از آنجا که در مساله برنامه ریزی مدیریت پسماند شناسایی دقیق و صحیح توزیع پارامترها خیلی مشکل است و اغلب داده های مورد نیاز از عدم قطعیت برخوردار هستند برای مدلسازی مساله پژوهش از رویکرد بهینه سازی استوار و رویکرد فازی شهودی استفاده شده است. نتایج انجام این پژوهش نشان می دهد که مقدار توابع هدف تعیین شده در روش استوار فازی نسبت به روش قطعی از مقدار کمتری برخوردار است. از دیگر نتایج این پژوهش می توان به کاهش 9 درصدی در میزان استفاده از مخازن مورد نیاز تخصیص داده شده به مکان های جمع آوری پسماند و همچنین کاهش 52 درصدی مکان های فعال جهت جمع آوری پسماند اشاره نمود.
کلید واژگان: مدیریت پسماند، برنامه ریزی استوار فازی، عدم قطعیت، الگوریتم ژنتیک چند هدفه، شهر بوشهرToday, the problem of locating waste tanks and routing waste collection vehicles is closely related to the health of society and the environment. In this article, the positioning-routing problem of the periodic arc along with the intermediate discharge stations is studied under conditions of uncertainty, the purpose of which is to reduce the number of active places for waste collection in Bushehr city, to determine the optimal routes. In order to serve all the areas with the demand of the urban graph network during the week and the number of vehicles is required. A mixed integer linear programming model is developed along with taking into account the fuzzy demand to optimize the problem, and multi-objective genetic algorithm is used for the approximate solution of the problem. To evaluate the efficiency of the proposed algorithm, CPLEX solver of GAMS software is used to solve problems with small and medium dimensions. In the current research, since it is very difficult to accurately and correctly identify the distribution of parameters in the problem of waste management planning, and most of the required data have uncertainty, robust optimization approach and intuitive fuzzy approach were used to model the research problem. Is. The results of this research show that the value of the objective functions determined in the fuzzy robust method is less than the deterministic method. Among the other results of this research, we can mention a 9% reduction in the amount of use of required tanks allocated to waste collection sites, as well as a 52% reduction in active sites for waste collection.
Keywords: Waste Management, Fuzzy Robust Planning, Uncertainty, Multi-Objective Genetic Algorithm, Bushehr City -
در دهه اخیر رایانش ابری مورد توجه بسیاری از ارایه دهندگان و استفاده کنندگان فناوری اطلاعات قرارگرفته است. یکی از مدل های پرکاربرد ارایه خدمات در حوزه رایانش ابری، مدل نرم افزار به عنوان خدمت بوده که معمولا به صورت ترکیبی از مولفه های داده و برنامه ارایه می شوند. یکی از چالش های مهم در این حوزه، یافتن مکان بهینه برای مولفه های نرم افزاری بر روی زیرساخت های ابری است که در آن نرم افزار به عنوان خدمت بتواند بهترین عملکرد ممکن را داشته باشد. مسئله جایابی نرم افزار به عنوان خدمت به چالش تعیین اینکه کدام سرویس دهنده ها در مرکز داده ابر، بدون نقض محدودیت های نرم افزار به عنوان خدمت، می توانند میزبان کدام مولفه ها باشند اشاره دارد. در این مقاله، راهکار بهینه سازی چند هدفه با هدف کاهش هزینه و زمان اجرا جهت جایابی مولفه های در محیط های ابری را ارایه می دهیم. راهکار پیشنهادی خود را با استفاده از کتابخانه Cloudsim شبیه سازی کرده و در نهایت با دو الگوریتم ازدحام ذرات چندهدفه و فاخته مورد ارزیابی و مقایسه قرار دادیم. نتایج شبیه سازی نشان می دهد که راهکار پیشنهادی عملکرد بهتری نسبت به دو الگوریتم پایه داشته و موجب کاهش 9/4 درصدی زمان اجرای جایابی مولفه های نرم افزار به عنوان خدمت، کاهش 9/1 درصدی هزینه و افزایش 7/9 درصدی بهره وری می گردد.
کلید واژگان: رایانش ابری، مولفه های نرم افزاری ترکیبی، مدیریت منابع، جایابی SaaS، الگوریتم ژنتیک چند هدفهJournal of Innovations of Aplied Information and Communication Technology, Volume:1 Issue: 2, 2022, PP 19 -33In the last decade, cloud computing has attracted the attention of many IT providers and users. One of the most widely used models of providing services in the field of cloud computing is the “software as a service” or SaaS model, which is usually provided as a combination of data and application components. One of the major challenges in this area is finding the optimal location for the software components on the cloud infrastructure where the software as a service can perform at its best. The problem of locating software as a service, addresses the challenge of determining which components in the cloud data center can host which components without violating the limitations of the software as a service. In this paper, we have presented a multi-objective optimization solution with the aim of reducing costs and execution time for locating components in cloud environments. We have simulated our proposed solution using the Cloudsim library and finally evaluated and compared it with two multi-objective and cuckoo search algorithms. The simulation results show that the proposed solution performs better than the two basic algorithms, reducing the implementation time of the “software as a service” components and the costs by 9.4% and 9.1% respectively, and increasing the productivity by 7.9%.
Keywords: cloud computing, Hybrid Software Components, Resource Management, SaaS Placement, Multi Objective Genetic Algorithm -
Project scheduling is one of the most important and applicable concepts of project management. Many project-oriented companies and organizations apply variable cost reduction strategies in project implementation. Considering the current business environments, in addition to lowering their costs, many companies seek to prevent project delays. This paper presents a multi-objective fuzzy mathematical model for the problem of project scheduling with the limitation of multi-skilled resources able to change skills levels, optimizing project scheduling policy and skills recruitment. Given the multi objectivity of the model, the goal programming approach was used, and an equivalent single-objective model was obtained. Since the multi-skilled project scheduling is among the NP-Hard problems and the proposed problem is its extended state, so it is also an NP-Hard problem. Therefore, NSGA II and MOCS meta-heuristic algorithms were used to solve the large-sized model proposed using MATLAB software. The results show that the multi-objective genetic algorithm performs better than the multi-objective Cuckoo Search in the criteria of goal solution distance, spacing, and maximum performance enhancement.
Keywords: Project scheduling, multi skilled resources, goal programming, multi-objective genetic algorithm, multi-objective Cuckoo Search -
در پژوهش حاضر عملکرد اجکتور تک فازی مافوق صوت با سیال کاری هوا در نرم افزار انسیس به صورت دوبعدی شبیه سازی شده است. هدف کار بررسی میدان های سرعت، فشار، رژیم جریان خروجی از نازل اولیه و نسبت مکش در شرایط مختلف عملکردی است. سپس برای رسیدن به بازده بیشتر، هندسه ی اجکتور با استفاده از الگوریتم چندهدفه ی ژنتیک بهینه سازی شده است. بهینه سازی با در نظر گرفتن اثر 4 پارامتر هندسی شامل قطر خروجی نازل اولیه، فاصله ی خروجی نازل اولیه تا ورودی قسمت اختلاط سطح مقطع ثابت، قطر و طول قسمت اختلاط مقطع ثابت انجام شده است. نتایج تحلیل حساسیت نشان می دهد که قطر قسمت اختلاط مقطع ثابت بیشترین تاثیر را بر روی دو تابع هدف نسبت مکش و نسبت فشار دارد. همچنین پس از بهینه سازی هندسه ی اجکتور مورد مطالعه، میزان نسبت مکش در حالت خفگی 11٫8 درصد و دامنه ی کارکرد آن 5 درصد افزایش یافته است.
کلید واژگان: اجکتور تک فاز، شبیه سازی دوبعدی، بهینه سازی دوهدفه، تحلیل حساسیتIn the present study, the performance of the 2D single-phase supersonic ejector with working fluid of air is simulated in ANSYS CFX. The aim is to investigate the velocity field, pressure distribution, primary nozzle flow regime, and entertainment ratio in different operational conditions. The primary pressure inlet with bar and the secondary inlet as an opening with bar at different outlet pressures are simulated. The turbulence model is used. The sidewalls are considered symmetry boundary conditions and the no-slip condition is applied to the ejector walls. Then, the ejector geometry is optimized using Multi-Objective Genetic Algorithm (MOGA) in order to reach greater efficiency. Optimization is performed considering geometric parameters including primary nozzle exit diameter, nozzle exit position, diameter, and length of the constant area section.Sensitivity analysis results show that the diameter of constant area section has major effect on the entertainment and pressure ratios as two objective functions. The nozzle exit position and external diameter of primary nozzle are the second and third dominant parameters that respectively influence the performance of the ejector, while the effect of constant area section length is negligible. Results indicate that by increasing the pressure ratio, the shock train moves upstream and at the design point, the last oblique shock is located in the exit of the constant area section, letting the remaining pressure recovery be done in the subsonic diffuser which reduces the pressure losses and increases the efficiency. Above the critical pressure ratio, due to the movement of the shock train to upstream weakening the shock strength, the suction pressure increases. Then, the pressure difference is reduced, leading to the lower secondary mass flow suction. The optimized ejector in the double choking condition has a % higher entertainment ratio and its operational range is enhanced by percent in comparison to the original geometry.
Keywords: Single phase ejector, 2D simulation, multi-objective Genetic Algorithm, sensitivity analysis -
Recommender Systems are part of our life nowadays. In almost every big business, recommender systems are suggesting items to users automatically. The only factor that traditional recommender systems take into account is accuracy. They try to estimate user-item ratings as accurate as possible to recommend the more preferable items to users. But from the supplier point of view, profit is the most desirable achievement. In this paper we have proposed a profit-aware recommender system based on the multi-objective genetic algorithm. Our proposed method consider two objectives at the same time: Profit and Accuracy. Profit is amount of profit that company will gain of selling items and accuracy measures how much recommendations are close to user preferences. the NSGA-II as the most successful MO-GA has been selected here and its Crossover and Mutation operations have been designed. Our method and traditional collaborative-filtering method have been implemented and tested on three different datasets: Movielens, Netflix and Yahoo. Results confirm that in both accuracy and net-profit our method prevail over CF method.
Keywords: Profit-Aware Recommender System, Multi-Objective Genetic Algorithm, Optimization -
Journal of Applied Dynamic Systems and Control, Volume:3 Issue: 2, Summer and Autumn 2020, PP 31 -38Nowadays, the presence of distributed generation (DG) units in the distribution network is increasing due to their advantages. Due to the increasing need for electricity, the use of distributed generation sources in the power system is expanding rapidly. On the other hand, in order to respond to the growth of load demand, the network becomes wider and more interconnected. These factors increase the level of fault current in the power system. Sometimes this increase causes the fault current level to exceed the ability to disconnect the protective devices, which can cause serious damage to the equipment in the power system. Using fault current limiters (FCLs) in power system is very promising solution in suppressing short circuit current and leads use of protective equipment with low capacities in the network. In this paper, in order to solve the problem of increasing the fault current, first using sensitivity analysis, network candidate lines are selected to install the fault current limiter, which helps to reduce the time and search space to solve the problem. Simultaneously finding the optimal number, location and amount of impedance for the installation of a resistive superconductor limiter is solved using the multi-objective Non-dominated genetic algorithm with non-dominated sorting (NSGA-II). The method presented in a 20 kV ring sample network, simulated in PSCAD software, is evaluated in the presence of distributed generation sources and its efficiency is shown.Keywords: Superconducting fault current limiters, optimal positioning, Multi-objective genetic algorithm, Non-dominated sorting (NSGA-II), Sensitivity analysis, Search space reduction
-
Web Services provides a solution to web application integration. Due to the significance of trust in choosing the proper web service, a novel optimal configuration of neural networks by multi-objective genetic algorithm and ensemble-classifier approach is used to evaluate the trust of single web services. For evaluating trust in the single web services, first, a set of neural networks were trained by the settings their parameters through the multi-objective genetic algorithm. Next, the best combination of neural networks was selected to make an ensemble classifier. This method was evaluated with single WS dataset considered eight criteria. Three measurements, accuracy, time and ROC curve were considered to assess the efficiency. Ultimately, the results show that the proposed approach can achieve a trade-off between time and accuracy by the multi-objective genetic algorithm. Also using ensemble-classifiers approach increases the reliability of the model. Consequently, the proposed method promote the detection accuracy.
Keywords: web service, Trust, Artificial Neural network, Multi-Objective Genetic Algorithm, Ensemble-Classifier -
پیش بینی واقع گرایانه بیشینه شتاب زمین (PGA)، به منظور استفاده در طراحی سازه های مقاوم در برابر زلزله، به خصوص در مناطق لرزه خیز از اهمیت ویژه ای برخوردار است. با استفاده از تحلیل خطر احتمالاتی زلزله، میزان لرزه خیزی یک منطقه هنگام وقوع زلزله مشخص می گردد. بنابراین یکی از مهم ترین بخش های تحلیل خطر، پیش بینی جنبش های نیرومند زمین می باشد که توسط روابطی موسوم به روابط کاهندگی به دست می آیند. مرکز مطالعات مهندسی زلزله (Peer) روابطی را تحت عناوین روابط کاهندگی NGA-West1 و NGA-West2 برای کل جهان ارایه نموده است. ازآنجاکه یک رابطه کاهندگی باید بتواند در برابر آزمون های آماری نظیر آزمون تحلیل باز نمونه گیری از داده ها که اخیرا توسط آذربخت و همکاران [1] ارایه گردیده است، نتایج مطلوبی را در بر داشته باشد، بنابراین در این پژوهش سعی می شود ضرایب برخی روابط کاهندگی نسل جدید نظیر کمپبل و بزرگ نیا [2]، آبراهامسون و سیلوا [3] و رابطه بور و اتکینسون [4] بر اساس مجموعه داده های منتشر شده توسط مرکز مطالعات مهندسی زلزله و با بهره گیری از الگوریتم ژنتیک چند هدفه، برای بیشینه شتاب زمین، بهینه سازی شوند. نتایج بیانگر تطبیق خوب روابط به دست آمده در برابر سایر آزمون های آماری می باشد. انتظار می رود بتوان از نتایج حاصل از این پژوهش در تحلیل خطر احتمالاتی زلزله بهره گرفت.
کلید واژگان: تحلیل خطر لرزه ای، روابط کاهندگی نسل جدید، الگوریتم ژنتیک چند هدفه، تحلیل حساسیت، کاتالوگ لرزه ایThe realistic estimation of Peak Ground Acceleration (PGA) is crucial for the purpose of seismic design in high seismic prone regions. The common practice is using Seismic Hazard Analysis (SHA) to estimate the design spectra in order to be used in the design and rehabilitation of structures. The most influential part of any SHA is the use of Ground Motion Prediction Equation (GMPE), which usually has the highest level of epistemic uncertainty. The Pacific Earthquake Research Centre (PEER) have released two sets of GMPEs, which are known as NGA-WEST1 and NGA-WEST2, and they are introduced as global GMPEs for all regions around the globe. However, the reliability of GMPEs needs to be assessed properly. Therefore, a recent methodology by Azarbakht et al. (2014) has been implemented in this study in order to enhance the given GMPEs only in the case of PGA. The better model in this approach is the one which has less sensitivity due to the small changes in the input catalogue. This effect cannot be captured by the common statistical tests that are widely using in the development of GMPEs. Therefore, three NGA GMPEs are taken into consideration, and the coefficients are optimized by aiming at maximizing the reliability, i.e. Campbell and Bozorgnia, Abrahamson and Silva, and Boore and Atkinson GMPEs. The ground motion database of the Campbell and Bozorgnia (2014) was used throughout this study, which consists of 15493 records of 319 earthquake events. A multi-objective Genetic algorithm was used to optimize four fitness functions, three of them related to resampling of the data and the forth is taken as the LLH. The results show that the employed resampling analysis show better performance when compared to other statistical approaches such as Var-test, Lillifors, and Z-test. However, the optimized coefficients show better GMPE performance with those statistical tests. Error estimation approaches were also considered, i.e. RMSE, MAE, R-square and E methods. In the end, the hazard curves for a hypothetical site are calculated based on the original and optimized GMPEs. The comparison between the obtained hazard curves shows that the hazard curves obtained from the optimized coefficients result in conservative when compared to the hazard curves from the original GMPEs. In conclusion, the optimized GMPEs show better performance when compared to the original GMPEs by means of the common statistical approaches as well as the new resampling algorithm. This proves that the sensitivity of GMPEs to the input catalogue is a key criterion when developing a new GMPE, otherwise the estimated parameters such as PGA will not be accurate enough.
Keywords: Seismic Hazard Analysis, New-Generation of Ground Motion, Prediction Equation (GMPE), Multi-Objective Genetic Algorithm, Sensitivity Analysis, Seismic Hazard Catalogue -
مقاله حاضر یک الگوریتم ژنتیک چندهدفه را برای طراحی بهینه یک سیستم تعلیق خودرو به کار می برد. مدل خودرو حرکت های سه بعدی بدنه خودرو را در نظر می گیرد. در این مدل کامل خودرو که دارای 8درجه آزادی است، حرکت عمودی صندلی مسافر، بدنه خودرو و چهار تایر و همچنین حرکت های چرخشی بدنه خودرو، درجات آزادی مدل را تشکیل می دهند. در این مقاله پارامترهای کاربردی تعلیق شامل شتاب صندلی مسافر، زاویه کله زنی بدنه خودرو، زاویه غلتش بدنه خودرو، نیروی دینامیکی تایر، سرعت تایر و انحراف تعلیق در نظر گرفته می شوند و در فرآیند بهینه سازی بهینه می شوند. جفت های متفاوتی از این پارامترها به عنوان توابع هدف، انتخاب و در فرآیند بهینه سازی چندهدفه بهینه می شوند و حل های پارتو برای جفت توابع هدف به دست می آیند. در فرآیند بهینه سازی نهایی، حل پارتو مربوط به مجموع پارامترهای بی بعد در یک گروه پارامترهای تعلیق نسبت به گروه دیگر به دست می آید. در این حل های پارتو، نقاط بهینه مهمی وجود دارند و طراحان می توانند هر یک از نقاط بهینه را برای یک هدف خاص انتخاب کنند. بهینه سازی پارتو بهتر از دیگر روش های بهینه سازی چندهدفه است، زیرا تعداد نقاط بهینه بیشتری در جبهه پارتو وجود دارد که هر نقطه معرف یک سطح از بهینه سازی برای جفت توابع هدف است و طراحان هر یک از نقاط را می توانند به دلخواه انتخاب کنند.کلید واژگان: مدل سه بعدی ارتعاشی خودرو، مدل کامل تعلیق خودرو، بهینه سازی پارتو، الگوریتم ژنتیک چندهدفهThe present paper applies a multi-objective genetic algorithm for optimally design of a vehicle suspension. The vehicle model considers three-dimensional movements of vehicle body. In this full vehicle model having 8 degrees of freedom, vertical movement of passenger seat, vehicle body, and 4 tires as well as rotational movements of vehicle body create the degrees of freedom of the model. In this paper, applicable suspension parameters, consisting of passenger seat acceleration, vehicle body pitch angle, vehicle body roll angle, dynamic tire force, tire velocity, and suspension deflections are considered and optimized in optimization process. Different pairs of these parameters are selected as objective functions and optimized in multi-objective optimization processes, and Pareto solutions are obtained for pair of objective functions. In final optimization process, the Pareto solution related to the summation of dimensionless parameters in one suspension parameters group versus other group, is derived. In these Pareto solutions, there are important optimum points and designers can choose any optimum points for a particular purpose. Pareto optimization is better than other multi-objective optimization methods because there are more optimum points on Pareto front, where each point represents a level of optimization for the pairs of objective functions, and designers can choose any of the points to specific purpose.Keywords: Three-Dimensional Vehicle Vibration Model, Full Vehicle Suspension Model, Pareto Optimization, Multi-Objective Genetic Algorithm
-
در سال های اخیر ساختار صنعت برق دچار تغییر و تحول شده و از آبان ماه سال 1382 که بازار برق کشور راه اندازی شد، ساختار انحصاری آن به ساختار رقابتی تبدیل شده است. در این بازار، پیش بینی قیمت برق نه تنها در قیمت دهی ضروری است بلکه در یافتن استراتژی بهینه بهره برداری از سوی بهره برداران نیروگاه نیز نقشی اساسی دارد. در این مقاله، از روش شبکه عصبی مصنوعی برای پیش بینی قیمت روزانه انرژی در ساعت های اوج استفاده شده و نتایج حاصل از آن برای بهینه سازی چند هدفه بهره برداری از مخزن سد کارون 5 به کار گرفته شده است. هدف های در نظر گرفته شده شامل دو هدف بیشینه سازی درآمد سالانه و بیشینه سازی حداقل انرژی تولیدی روزانه است. روش بهینه سازی چندهدفه NSGA II، به منظور حل مساله بهینه سازی چندهدفه انتخاب شد و به منظور اعمال مدل پیشنهادی، سد و نیروگاه برق آبی کارون 5 به عنوان مطالعه موردی انتخاب گردید. ازآنجاکه خروجی مدل بهینه سازی چندهدفه، محدوده وسیعی از جواب های نامغلوب است، به منظور انتخاب جواب بهینه نهایی از بین جواب های به دست آمده، روش های رفع اختلاف نش، نش نامتقارن، Kalai-Smoronvski و Area Monotonic استفاده شد.کلید واژگان: تولیدکننده برق آبی، بازار برق، شبکه عصبی مصنوعی، الگوریتم ژنتیک چندهدفه، رفع اختلاف، NashIn recent years, the structure of the electricity industry has undergone a change and since November 2003, when the electricity market of the country was launched, its monopoly structure has become a competitive structure. In this market, the forecast of electricity prices is not only necessary in pricing but also plays an important role in finding the optimal operation strategy by the power plant's users. In this paper, an artificial neural network approach is used to predict daily energy prices in peak hours and its results have been used to optimize the multi-purpose operation of the reservoir of Karun 5 dam. The intended goals include two goals of maximizing annual income and maximizing the minimum daily energy production. The multi-objective optimization method NSGA II was selected to solve the multi-objective optimization problem. In order to apply the proposed model, Karun 5 dam and hydroelectric power station was selected as case study. Since the output of the multi-objective optimization model is a wide range of non-dominated responses, In order to select the optimal solution among obtained solutions, Nash bargaining, non-symmetric Nash, Kalai-Smoronvski and Area Monotonic was implemented.Keywords: Hydro-power producer, electricity market, neural networks, multi-objective genetic algorithm, conflict resolution, Nash
-
Handwritten digit recognition, as an important issue in pattern classification, has received considerable attention of many researchers to develop theoretical and practical aspects of this problem. The goal is to recognize a printed digit text or a scanned handwritten using an automated procedure. In this paper an optical character recognition system with a multi-step procedure is presented for Farsi handwritten digit classification. First, a pre-processing course is performed on the image to enhance and make prepare the image for the main steps. Then multiple features are extracted which are believed to be effective in the classification step, among which a multi-objective genetic algorithm selects those with more discriminative characteristics in order to reduce the computational complexity of the classification steps. Following this, only the selected features are extracted from the digit images to be entered to three classifiers. Once the classifiers are trained, their outputs are fed into a classifier ensemble to make the final decision. The weights of the linear combination of classifiers are adjusted by an evolutionary firefly algorithm which optimizes the F-criterion. Evaluation of the proposed technique on the standard HODA database demonstrates that the algorithm of this paper attains higher performance indices including F-measure of 98.88%, compared to other existing methods.Keywords: Classifiers ensemble, Feature extraction, Feature selection, Firefly algorithm, Multi-objective genetic algorithm, Optical character recognition
-
The use of lasers is being considered as a modern method for forming process in recent years. This method has been used in various industries, such as aerospace, marine and oil industry. Extensive research has been done in the field of modeling and optimization of direct paths parameters with process of laser forming. Although forming in circular paths can be used for producing complex parts, due to some technical reasons, it is considered less. The main purpose of this paper is to detect the proper estimation model and obtain optimal variables conditions for complete circular paths in perforated circular parts by means of genetic algorithms. In this process the outer edges are fixed and the inner edges are being formed by laser. At first, the finite element simulation model is studied then the estimation model has been discussed, after that multi-objective functions have been examined with the least error and energy. Furthermore, the optimization results of the internal hole diameters are reported and analyzed in terms of Pareto charts. In conclusion, optimum forming conditions have been reported in terms of accuracy and energy for different diameters of holes. This study shows with acceptable increasing in the error rate, the required energy could be reduced. Also, increasing in the diameter of inside hole cause to increase energy and decrease of accuracy.Keywords: Laser forming, Circular scanning path, Deflection, Multi objective genetic algorithm, optimization
-
در فرآیندهای شکل دهی قطعات ورقی، یکی از مهم ترین محدودیت ها، برگشت کشسان ورق بعد از برداشتن نیرو است که معمولا منجر به پدیده بازگشت فنری می شود. در صورت کنترل نشدن این پارامتر، تولید محصولات دقیق امکان پذیر نخواهد بود. لذا هدف اصلی این تحقیق، کمینه کردن میزان بازگشت فنری همراه با جلوگیری از جوانه زنی ترک در فرآیند خمکاری یک طرفه ورق آلومینیومی A1050-H14 است. بدین منظور، ابتدا فرآیند خمکاری ورق در نرم افزار المان محدود آباکوس شبیه سازی شد و تاثیر پارامترهای ضریب اصطکاک بین سنبه و ورق و سرعت پایین آمدن سنبه بر میزان بازگشت کشسان ورق و همچنین بر میزان تنش فون میزز ایجادشده در ورق بررسی شد تا ضمن کمینه کردن میزان بازگشت فنری، از جوانه زنی ترک هم جلوگیری شود. در این راستا، از زبان برنامه نویسی پایتون استفاده شد. سپس با ترکیب الگوریتم بهینه سازی ژنتیک چندهدفه با روش المان محدود در نرم افزار مود فرانتیر، مقادیر بهینه پارامترهای ورودی محاسبه شد. شایان ذکر است که به منظور اعتبارسنجی نتایج، نتایج شبیه سازی این تحقیق با نتایج تجربی سایر محققان مورد مقایسه قرار گرفت که درصد خطای نسبی بین نتایج عددی و تجربی برابر با 3/14% بود.کلید واژگان: فرآیند خم کاری، بازگشت فنری، الگوریتم ژنتیک چند هدفه، روش المان محدود جوانه زنی ترکIn sheet metal forming processes, one of the most important limitations relates to the elastic recovery after punch unloading, which usually leads to spring-back phenomenon. Production of precise parts without spring-back controlling is not possible. Hence, the main aim of the present research is to minimize the amount of spring-back as well as to prevent the crack initiation during bending process of Aluminum A1050-H14 sheet. For this purpose, firstly, the sheet metal bending process was numerically simulated in ABAQUS finite element package. Then, the effect of friction coefficient and punch velocity was investigated on elastic recovery and von Mises stress in order to minimize the spring-back as well as to prevent the crack initiation. In this regard, python programming language was utilized. Then, by linking multi-objective genetic algorithm and finite element method in modeFRONTIER software, the optimum values of the process parameters were determined. It should be mentioned that for validation purposes, the simulation results of the present study were compared with the experimental data available in literature, showing a 3.14% relative error between the numerical and experimental results.Keywords: Bending, Spring-back, Multi-objective Genetic Algorithm, Finite Element Method, Crack Initiation
-
هدف این مقاله طراحی بهینه یک شیار محیطی برای کمپرسور گذرصوتی روتور 37 ناسا است. پارامترهای طراحی شامل عمق، عرض و موقعیت قرارگیری شیار و توابع هدف شامل حاشیه وامانش و راندمان بیشینه می باشند و از نسبت فشار وامانش به عنوان معیاری برای انتخاب طرح بهینه استفاده شده است. مدل سازی ریاضی کمپرسور که شامل برقراری رابطه بین متغیرهای طراحی و توابع هدف می باشد، با استفاده از شبکه های عصبی صورت گرفته است. به منظور آموزش شبکه های عصبی، از نتایج حاصل از دینامیک سیالات محاسباتی (CFD) برای 123 نقطه طراحی استفاده شده است. در ادامه، طراحی بهینه به وسیله الگوریتم ژنتیک چند هدفه انجام شده است که منجر به مجموعه ای از پاسخ های بهینه (پرتو فرانت) گردیده است. پس از مرتب سازی این مجموعه بر اساس بیشترین حاشیه وامانش، اولین پاسخی که نسبت فشار وامانش آن بیشتر از نسبت فشار وامانش حالت پوسته صاف بود به عنوان پاسخ نهایی انتخاب گردید. شبیه سازی این طرح نشان داد که شیار بهینه منجر به افزایش 2/6 درصدی حاشیه وامانش می شود و تاثیر بسیار ناچیزی بر راندمان کمپرسور دارد.کلید واژگان: شیار محیطی، بهینه سازی، الگوریتم ژنتیک چند هدفه، شبکه های عصبی، حاشیه وامانش، CFDThe aim of this paper is the optimum design of a single circumferential groove for NASA Rotor 37, a transonic compressor. The design variables are the width, the depth and the position of the groove. Also the objective functions are the stall margin and the peak efficiency. The total pressure ratio at the near stall condition is used as a criterion to choose the optimum design. Neural networks are used to model the relation between design variables and objective functions. 123 various design points are simulated by CFD and the results are used to train the networks. Afterwards, a multi-objective genetic algorithm is used to optimize the design based on the maximum stall margin and the minimum reduction of the peak efficiency. After sorting the Pareto front according to the stall margin value, the first design point whose total pressure ratio is larger than the smooth wall condition, is selected as the optimum design. Simulation of the compressor with optimum groove shows the increase of the stall margin by 6.2 percent with negligible effect on the efficiency.Keywords: Circumferential groove, Optimization, Multi-objective genetic algorithm, Neural networks, Stall margin, CFD
-
قواعد انجمنی، برای یافتن ارتباط پنهان و وابستگی های میان مجموعه عناصر مختلف در پایگاه داده به کار می روند که در قالب قانون استخراج می شوند؛ اما مشکل این روش، افشاء اطلاعات حساس و تهدید محرمانگی اطلاعات می باشد. فرایند ایمن سازی داده ها باوجود تحقیقات گسترده در این حوزه به عنوان یک مسئله NPHard در نظر گرفته می شود. این مقاله با استفاده از الگوریتم ژنتیک چندهدفه و نیز رویکرد مبتنی بر پشتیبان، سعی در کاهش پشتیبانی مجموعه عناصر حساس موجود در پایگاه داده تراکنشی دارد. روش پیشنهادی با حذف تراکنش هایی که شامل عناصر حساس هستند، باعث کاهش پشتیبانی عناصر حساس به کمتر از حداقل آستانه پشتیبانی شده که ایمن سازی پایگاه داده را به همراه دارد. روش پیشنهادی در هر تکرار، تنها با یک بار پویش تراکنش های حساس به جای پویش کل تراکنش های پایگاه داده، باعث افزایش سرعت و کاهش هزینه های اجرا می گردد. همچنین برای کاهش عوارض ناشی از پنهان سازی، تراکنش ها بر اساس کمترین طول یا بیشترین عنصر حساس و کمترین عنصر غیر حساس مرتب سازی می شوند.کلید واژگان: قواعد انجمنی، پنهان سازی مجموعه عناصر حساس، الگوریتم های ژنتیک چندهدفهAssociation rules are used to find hidden relationships and dependencies among different itemsets in the database that is extracted in the form of the rule, but the problem with this approach is the discovery of sensitive information and the treatment of information privacy. The sanitization process data is considered as a NPHard problem. In this article, we try to reduce support the sensitive itemsets in the transactional database using multi-objective genetic algorithms and the support-based approach. The proposed approach with the transaction deletion that includes sensitive itemsets leads to less support sensitive itemsets than the minimum support threshold and leads to the database sanitization. In each iteration of our method leads to increase the speed and reduce the performance criteria by one time of the scanning of the sensitive transaction instead of scanning the entire database of transactions. In addition, to reduce the effects of hiding the sensitive itemsets, the transactions sort based on the shortest length, the most sensitive itemsets and the least non-sensitive itemsets.Keywords: Association rules, hiding the sensitive itemsets, multi-objective genetic algorithm
-
A mobile ad-hoc network (MANET), a set of wirelessly connected sensor nodes, is a dynamic system that executes hop-by-hop routing independently with no external help of any infrastructure. Proper selection of cluster heads can increase the life time of the Ad-hoc network by decreasing the energy consumption. Although different methods have been successfully proposed by researchers to tackle this problem, nearly all of them have the deficiency of providing a single combination of head clusters as the solution. On the contrary, in our proposed method, using a Multi-Objective Genetic Algorithm, a set of near optimum solutions is provided. In the proposed method, energy consumption, number of cluster heads, coverage and degree difference are considered as objectives. Numerical results reveal that the proposed algorithm can find better solutions when compared to conventional methods in this area namely, weighted clustering algorithm (WCA), comprehensive learning particle swarm optimization (CLPSO) and multi objective particle swarm optimization(MOPSO).Keywords: Mobile Ad-Hoc Network, Multi Objective Genetic Algorithm, Cluster-Head
-
Scheduling for a two-stage production system is one of the most common problems in production management. In this production system, a number of products are produced and each product is assembled from a set of parts. The parts are produced in the first stage that is a fabrication stage and then they are assembled in the second stage that usually is an assembly stage. In this article, the first stage assumed as a hybrid flow shop with identical parallel machines and the second stage will be an assembling work station. Two objective functions are considered that are minimizing the makespan and minimizing the sum of earliness and tardiness of products. At first, the problem is defined and its mathematical model is presented. Since the considered problem is NP-hard, the multi-objective genetic algorithm (MOGA) is used to solve this problem in two phases. In the first phase the sequence of the products assembly is determined and in the second phase, the parts of each product are scheduled to be fabricated. In each iteration of the proposed algorithm, the new population is selected based on the non-dominance rule and fitness value. To validate the performance of the proposed algorithm, in terms of solution quality and diversity level, various test problems are designed and the reliability of the proposed algorithm is compared with two prominent multi-objective genetic algorithms, i.e. WBGA, and NSGA-II. The computational results show that the performance of the proposed algorithms is good in both efficiency and effectiveness criteria. In small-sized problems, the number of non-dominance solution come out from the two algorithms N-WBGA (the proposed algorithm) and NSGA-II is approximately equal. Also, more than 90% solution of algorithms N-WBGA and NSGA-II are identical to the Pareto-optimal result. Also in medium problems, two algorithms N-WBGA and NSGA-II have approximately an equal performance and both of them are better than WBGA. But in large-sized problems, N-WBGA presents the best results in all indicators.Keywords: Multi-objective Genetic algorithm, Two-stage production system, Assembly
-
Journal of Advances in Computer Engineering and Technology, Volume:3 Issue: 1, Winter 2017, PP 11 -17
Nowadays, developed and developing countries using smart systems to solve their transportation problems. Parking guidance intelligent systems for finding an available parking space, are considered one of the architectural requirements in transportation. In this paper, we present a parking space reservation method based on adaptive neuro-fuzzy system(ANFIS) and multi-objective genetic algorithm. In modeling of this system, final destination, searching time and cost of parking space have been used. Also, we use the vehicle ad-hoc network (VANET) and time series, for traffic flow predict and choose the best path. The benefits of the proposed system are declining searching time, average the walking and travel time. Evaluations have been performed by the MATLAB and we can see that the proposed method makes a good sum of best cost which is useful and meaningful in a parking space reserved for drivers and facility managers. The simulation results show that the performance and accuracy of the method have been significantly improved compared to previous works.
Keywords: VANET, multi-objective genetic algorithm, ANFIS, reservation -
Different approaches have been proposed for feature selection to obtain suitable features subset among all features. These methods search feature space for feature subsets which satisfies some criteria or optimizes several objective functions. The objective functions are divided into two main groups: filter and wrapper methods. In filter methods, features subsets are selected due to some measures like inter-class distance, features statistical independence or information theoretic measures. Even though, wrapper methods use a classifier to evaluate features subsets by their predictive accuracy (on test data) by statistical resampling or cross-validation. Filter methods usually use only one measure for feature selection that does not necessarily produce the best result. In this paper, we proposed to use the classification error measures besides to filter measures where our classifier is support vector machine (SVM). To this end, we use multi objective genetic algorithm. In this way, one of our feature selection measure is SVM classification error. Another measure is selected between mutual information and Laplacian criteria which indicates informative content and structure preserving property of features, respectively. The evaluation results on the UCI datasets show the efficiency of this method.
Keywords: feature selection, Multi objective genetic algorithm, support vector machine
- نتایج بر اساس تاریخ انتشار مرتب شدهاند.
- کلیدواژه مورد نظر شما تنها در فیلد کلیدواژگان مقالات جستجو شدهاست. به منظور حذف نتایج غیر مرتبط، جستجو تنها در مقالات مجلاتی انجام شده که با مجله ماخذ هم موضوع هستند.
- در صورتی که میخواهید جستجو را در همه موضوعات و با شرایط دیگر تکرار کنید به صفحه جستجوی پیشرفته مجلات مراجعه کنید.