Academic literature on the topic 'Bee Swarm Optimization (BSO)'

This diploma thesis deals with the path planning by swarm intelligence. In the theoretical part it describes the best known methods of swarm intelligence (Ant Colony Optimization, Bee Swarm Optimization, Firefly Swarm Optimization and Particle Swarm Optimization) and their application for path planning. In the practical part particle swarm optimization is selected for the design and implementation of path planning in the C#.

<p> Most engineering fields often encounter challenges in material, performance, and time efficiency. Truss design is a subject many structural engineers confront in their careers. Optimization is an effective approach in solving preliminary designs of truss structures. This paper studies two different multi-objective optimization algorithms, the particle swarm optimization (MOPSO) and the bee algorithm (MOBA), to optimize a simply supported flat-truss bridge designed by California State University, Long Beach&rsquo;s Steel Bridge team for the American Institute of Steel Construction (AISC) Spring 2018 competition. The variables, randomly selected from a continuous domain, are the top chord area, bottom chord area, web member area, and the center-to-center distance between the top and bottom chords. The optimized objectives are the weight and deflections of the bridge for the six load combinations stipulated in AISC&rsquo;s rules. Both algorithms are calibrated using recommended parameter values derived from the parametric studies conducted. To compare their effectiveness, the recommended parameters were selected so that run-times for both optimization codes were similar. Both algorithms generated optimized solutions to the multi-objective truss problem, but MOPSO exhibited more, and better, solutions in a slightly longer run-time than MOBA.</p><p>

Актуальність теми. Оскільки розмір цифрової інформації зростає в геометричній прогресії, потрібно витягувати великі обсяги необроблених даних. На сьогоднішній день існує кілька методів налаштування та обробки даних відповідно до наших потреб. Найбільш поширеним методом є використання інтелектуального аналізу даних (Data Mining). Data Mining застосовується для вилучення неявних, дійсних та потенційно корисних знань із великих обсягів необроблених даних. Видобуті знання повинні бути точними, читабельними та легкими для розуміння. Крім того, процес видобутку даних також називають процесом виявлення знань, який використовувався в більшості нових міждисциплінарних областей, таких як бази даних, статистика штучного інтелекту, візуалізація, паралельні обчислення та інші галузі. Одним із нових і надзвичайно потужних алгоритмів, що використовуються в Data Mining, є еволюційні алгоритми та підходи, що базуються на рії, такі як мурашиний алгоритм та оптимізація рою частинок. В даній роботі запропоновано використати для інтелектуального аналізу даних досить нову ідею алгоритма бджолиного рою для широко розповсюдженої задачі класифікації. Мета роботи: покращення результатів класифікації даних в сенсі в точності і сталості за допомогою алгоритму інтелектуального аналізу даних на основі алгоритму бджолиного рою. Об’єктом дослідження є процес інтелектуального аналізу даних для задачі класифікації. Предметом дослідження є використання алгоритму бджолиного рою для інтелектуального аналізу даних. Методи дослідження. Використовуються методи параметричного дослідження евристичних алгоритмів, а також методи порівняльного аналізу для алгоритмів інтелектуального аналізу даних. Наукова новизна одержаних результатів роботи полягає в тому, що після проведеного аналізу існуючих рішень, запропоновано використати алгоритм бджолиного рою для задачі класифікації, точність і сталість якого перевищує показники існуючих класифікаторів. Практичне значення одержаних результатів полягає в тому, що розроблений алгоритм показує кращі результати в сенсі точності і сталості в порівнянні з іншими алгоритмами інтелектуального аналізу даних. Тобто адаптація бджолиного алгоритму може розглядатися як корисне та точне рішення для такої важливої проблеми, як задача класифікації даних. Апробація роботи. Основні положення й результати роботи були представлені та обговорювались на науковій конференції магістрантів та аспірантів «Прикладна математика та комп’ютинг» ПМК-2019 (Київ, 2019 р.), а також на науковій конференції магістрантів та аспірантів «Прикладна математика та комп’ютинг» ПМК-2020 (Київ, 2020 р.). Структура та обсяг роботи. Магістерська дисертація складається з вступу, чотирьох розділів, висновків та додатків. У вступі надано загальну характеристику роботи, виконано оцінку сучасного стану проблеми, обґрунтовано актуальність напрямку досліджень, сформульовано мету і задачі досліджень, показано наукову новизну отриманих результатів і практичну цінність роботи, наведено відомості про апробацію результатів і їх впровадження. У першому розділі розглянуто алгоритми інтелектуального аналізу даних, які використовуються для задачі класифікації. Обґрунтовано можливість використання евристичних алгоритмів, а саме алгоритму бджолиного рою для цієї задачі. У другому розділі детально розглянуто алгоритм бджолиного рою та принципи його роботи, також описано запропоновану методику його застосування для інтелектуального аналізу даних, а саме для задачі класифікації. У третьому розділі описано розроблений алгоритм та програмний додаток, в якому він реалізований. У четвертому розділі приведена оцінка ефективності запропонованого алгоритм, на основі тестування алгоритму, а також порівняльного аналізу між розробленим алгоритмом та вже існуючими. У висновках представлені результати магістерської дисертації. Робота виконана на 81 аркуші, містить посилання на список використаних літературних джерел з 18 найменувань. У роботі наведено 38 рисунків та 5 додатків.<br>Actuality of theme. As the size of digital information grows exponentially, large amounts of raw data need to be extracted. To date, there are several methods to customize and process data according to our needs. The most common method is to use Data Mining. Data Mining is used to extract implicit, valid and potentially useful knowledge from large amounts of raw data. The knowledge gained must be accurate, readable and easy to understand. In addition, the data mining process is also called the knowledge discovery process, which has been used in most new interdisciplinary fields, such as databases, artificial intelligence statistics, visualization, parallel computing, and other fields. One of the new and extremely powerful algorithms used in Data Mining is evolutionary algorithms and swarm-based approaches, such as the ant algorithm and particle swarm optimization. In this paper, it is proposed to use a fairly new idea of the swarm of bee swarm algorithm for data mining for a widespread classification problem. Purpose: to develop an algorithm for data mining for the classification problem based on the swarm of bee swarms, which exceeds other common classifiers in terms of accuracy of results and consistency. The object of research is the process of data mining for the classification problem. The subject of the study is the use of a swarm of bee swarms for data mining. Research methods. Methods of parametric research of heuristic algorithms, and also methods of the comparative analysis for algorithms of data mining are used. The scientific novelty of the work is as follows: 1. As a result of the analysis of existing solutions for the classification problem, it is decided to use such metaheuristics as the swarm of bee swarm. 2. The implementation of the bee algorithm for data mining is proposed. The practical value of the results obtained in this work is that the developed algorithm can be used as a classifier for data mining. In addition, the proposed adaptation of the bee algorithm can be considered as a useful and accurate solution to such an important problem as the problem of data classification. Approbation of work. The main provisions and results of the work were presented and discussed at the scientific conference of undergraduates and graduate students "Applied Mathematics and Computing" PMK-2019 (Kyiv, 2019), as well as at the scientific conference of undergraduates and graduate students "Applied Mathematics and Computing" PMK-2020 (Kyiv, 2020). Structure and scope of work. The master's dissertation consists of an introduction, four chapters, conclusions and appendices. The introduction provides a general description of the work, assesses the current state of the problem, substantiates the relevance of research, formulates the purpose and objectives of research, shows the scientific novelty of the results and the practical value of the work, provides information on testing and implementation. The first section discusses the data mining algorithms used for the classification problem. The possibility of using heuristic algorithms, namely the bee swarm algorithm for this problem, is substantiated. The second section discusses in detail the algorithm of the bee swarm and the principles of its operation, also describes the proposed method of its application for data mining, namely for the classification problem. The third section describes the developed algorithm and the software application in which it is implemented. In the fourth section the estimation of efficiency of the offered algorithm, on the basis of testing of algorithm, and also the comparative analysis between the developed algorithm and already different is resulted. The conclusions present the results of the master's dissertation. The work is performed on 89 sheets, contains a link to the list of used literature sources with 18 titles. The paper presents 38 figures and 2 appendices.

Power systems are presently experiencing a period of rapid change driven by various interrelated issues, e.g., integration of renewables, demand management, power congestion, power quality requirements, and frequency regulation. Although the deployment of Energy Storage Systems (ESSs) has been shown to provide effective solutions to many of these issues, misplacement or non-optimal sizing of these systems can adversely affect network performance. This present research has revealed some novel working strategies for optimal allocation and sizing of utility-scale ESSs to address some important issues of power networks at both distribution and transmission levels. The optimization strategies employed for ESS placement and sizing successfully improved the following aspects of power systems: performance and power quality of the distribution networks investigated, the frequency response of the transmission networks studied, and facilitation of the integration of renewable generation (wind and solar). This present research provides effective solutions to some real power industry problems including minimizationof voltage deviation, power losses, peak demand, flickering, and frequency deviation as well as rate of change of frequency (ROCOF). Detailed simulation results suggest that ESS allocation using both uniform and non-uniform ESS sizing approaches is useful for improving distribution network performance as well as power quality. Regarding performance parameters, voltage profile improvement, real and reactive power losses, and line loading are considered, while voltage deviation and flickers are taken into account as power quality parameters. Further, the study shows that the PQ injection-based ESS placement strategy performs better than the P injection-based approach (in relation to performance improvement), providing more reactive power compensations. The simulation results also demonstrate that obtaining the power size of a battery ESS (MVA) is a sensible approach for frequency support. Hence, an appropriate sizing of grid-scale ESSs including tuning of parameters Kp and Tip (active part of the PQ controller) assist in improving the frequency response by providing necessary active power. Overall, the proposed ESS allocation and sizing approaches can underpin a transition plan from the current power grid to a future one.

碩士<br>國立臺北科技大學<br>工業工程與管理研究所<br>97<br>Most scheduling problems are very complex combinatorial optimization problems and hard to solve. The job-shop scheduling problem (JSP) is one of the problems. In the literature, more and more researchers used different heuristic algorithms to solve combinatorial optimization problems. Common algorithms are simulated annealing, genetic algorithm, tabu search approach, ant colony optimization and particle swarm optimization and so on. By more and more progress technology, the traditional job-shop scheduling is not enough to solve the diversity and a little amount production type. The problem is referred to as the flexible job-shop scheduling problem (FJSP). FJSP is an extension of the classical JSP which allows an operation to be processed by any machine out of a set of machines. It combines all of the complexities of JSP and more elaborate than JSP. Honey-bee mating optimization is a burgeoning heuristic algorithm which included of SA, GA, local search, and some innovations for its self-adaptation. Several studies have been made on efficiency evaluation of HBMO, and HBMO has proven to have good performance and quality in solving NP-hard problems. In this research, we proposed a heuristic algorithm which integrates PSO and HBMO for solving the multi-objective FJSP. Experiment results indicate this method is competitive and efficient.

碩士<br>義守大學<br>資訊管理學系碩士班<br>97<br>Swarm intelligence is one of the most popular derivative-free and population-based optimization algorithm. It has been extensively used for both continuous and discrete optimization problems due to its versatile optimization capabilities. Swarm intelligence is a research limb that models the population of interacting agents or swarms that are able to self-organize. An ant colony, a flock of birds or an immune system is a typical example of a swarm system. Bees’ swarming around their hive is another example of swarm intelligence. Artificial Bee Colony algorithm is an optimization algorithm based on the intelligent behavior of honey bee swarm. This thesis presents Bee Swarm Optimization intended to introduce chaotic sequences and psychology factor of emotion into the algorithm. We define two emotions Bees could have, positive and negative, and correspond to two reaction to perception respectively. For avoiding premature convergence it allows the proposed Emotional Chaotic Bee Swarm Optimization to continue search for better even best optimization in classic optimization problems, reaching better solutions than classic Artificial Bee Colony algorithm with a faster convergence speed.

碩士<br>國立中山大學<br>電機工程學系研究所<br>103<br>Under the guidance of international energy event occurred and international agreements, so Energy Saving and Carbon Reduction have already become an important issue in every county. However, the advances in green power not only provided alternative programs, but also reduced environmental pollution when using traditional way to produce energy. As increasing those unstable supply of green power. It must do some impact on traditional power grid. Such as power quality, system reliability, cost of power, etc. Therefore a microgrid which can quick react and dispatch the power demand is taken seriously gradually. How to build a microgrid with quick reaction and enhance power efficiency is an important issue currently. This thesis combined microturbines, wind power, solar power, power storage system, and combined cooling, heating and power(CCHP) to form a microgrid system. Then applying this design into Penghu power system, and reach the function of demand response by power storage system. For minimum cost of generating power this objective. Using combine fuzzy rule into Bee Swarm Optimization (BSO) to solve the problem of generation unit commitment (UC) and economic dispatch(ED). The UC and ED problem must satisfy the constraints of load demand, generating limits, ramp rate limits, and also the minimum up/down time of generators, and capacity of power storage system, etc. For avoid the local optimality problem, this thesis proposed the utilization of combined Probability Selection Fuzzy Rule into Self-Adaption Enhanced Bee Swarm Optimization (SAEBSO) method, which can quickly reach the optimal solution with better performance and accuracy.

碩士<br>國立臺北科技大學<br>工業工程與管理研究所<br>98<br>In the recently years, the stock market is very unstable in Taiwan; especially after attacked by the worldwide financial crisis, the market trend becomes quite unclear. Under today’s unpredictable stock market, how to use financial instruments to make asset allocation and gain the best profit return is the most important topic to address. Therefore, through the help of information technology and heuristic algorithms, this research aimed to establish an information system to identify the best portfolio out of stocks in Taiwan 50 Index and Taiwan Med-Cap 100 Index so that financial risks could be diversified; furthermore, through investment method of this research, investors with little financial knowledge could obtain an investment portfolio with low risks and stable profits. This research was divided into two stages. First, we adopted principle component analysis to create main components with high explanatory power from hundreds of indicators in financial statements. Through these main components, the best portfolio in each quarter could be identified from Taiwan 50 Index and Med-Cap 100. Second, Multi-objective Particle Swarm Optimization combined (MOPSO) with Honey-Bee Mating Optimization (HBMO) was used to develop appropriate weights of asset allocation and weights of technology indicators. Then, investors with little financial knowledge could apply the derived information from our model to make investment, estimate the return rate of the next quarter, and predict the rise and fall of the purchased stocks The component stocks of Taiwan 50 Index and Mid-Cap 100 Index were our study target, and the time span of this study was from the fourth quarter of 2006 to the third quarter of 2009. Empirical results showed the investment portfolio built in this study would generate a return rate of 23.86% and a prediction rate of 70% over this three-year period, which were superior to those of other indexes such as Taiwan 50 Index and Med-Cap 100. Therefore, we can conclude that this method could not only predict the future trend of the stock market but also obtain stable returns in each quarter while reducing investment risks.

碩士<br>國立臺灣科技大學<br>營建工程系<br>100<br>This study hybridizes particle swarm optimization (PSO) and artificial bee colony (ABC) to develop approaches which is more applicable than ABC and PSO. To hybride of ABC and PSO(HBP) approaches, agents including PSO particles and ABC bees are categorized into two sub-swarms by their species. Sequentially, agents in a sub-swarm are allowed to migrate to the other sub-swarm based on the sub-swarm fitness. And then, the PSO sub-swarm is permitted to learn from the global information, which involves the best position of the ABCsub-swarm. Twenty-three benchmark functions are employed to compare performance of HBP approaches against single ABC and PSO approaches. A practical hospital facility layout problem is investigated to validate the practicality of the HBP approaches. Results reveal the designed HBP approaches have dynamical sub-population sizes,superior performance to single ABC and PSO approaches, improvidences on a referenced hospital layout, and high practicality without judging performance of ABC and PSO on problems.