China
Africa
Explore chapters and articles related to this topic
Swarm Intelligence for Data Mining
Published in Wellington Pinheiro dos Santos, Juliana Carneiro Gomes, Valter Augusto de Freitas Barbosa, Swarm Intelligence Trends and Applications, 2023
Artificial bee colony algorithm is a SI-based algorithm, developed by Karaboga (Karaboga, 2005) for optimization problems, which mimics the intelligent behavior of honey bees searching for food sources around their hives (Alshamiri et al., 2016). The colony members share the information about the quality of food sources using the dance language (Kumar et al., 2017). A possible solution to the optimization problem is represented by a food source evaluated by the amount of nectar in the food source (Karaboga et al., 2008; Zhang et al., 2010). Artificial bee colony consists of three types of bees: employed bees, onlookers and scouts. The first half of the colony contains employed bees and the second one consists of the onlookers. There is only one employed bee for every food source. Therefore, the number of employed bees is equal to the number of food sources (solutions) (Alshamiri et al., 2016; Zhang et al., 2010). The employed bees search for food sources and collect information about them such as location, quality and quantity of nectar, and share this information with the onlooker bees within the hive. Onlooker bees select one of the food sources based on the information obtained by the employed bees to exploit with a probability proportional to its quality. The employed bee whose food source has been abandoned becomes a scout and randomly searches for a new food source. A scout bee turns into an employed bee when it finds a food source (Alshamiri et al., 2016; Karaboga et al., 2008). The flowchart and a simple procedure of ABC algorithm is given in Fig. 3 and Algorithm 3, respectively.
Bio-inspired optimization algorithms for machine learning in agriculture applications
Published in Govind Singh Patel, Amrita Rai, Nripendra Narayan Das, R. P. Singh, Smart Agriculture, 2021
The foraging behavior and mating behaviors of bees are the motivation for artificial bee colony algorithm. A bee chooses a food source by waiting for a decision in dance area and is called onlooker. Some bees visit the food source before calling employed bees. Random search is performed by scout bees to find new sources of food. The food source location and the amount of nectar correspond to optimization problem’s possible solution and the quality (fitness) of the solution. In this two-dimensional search space, a cluster of virtual bees is created which begins to move about randomly. When some target nectar is found by the bees, they interact. The intensity of bee interactions is used to obtain the solution of the problem. This algorithm runs in three phases. First is search process, the next phase is reproduction and the final phase is replacement of bee and selection [23,24].
Collective Intelligence in Networking
Published in Phan Cong Vinh, Nature-Inspired Networking: Theory and Applications, 2018
MANET is a mobile multi-hop wireless self-organized distributed ad-hoc network that does not require the basic internal construction. Routing in MANET is a challenging problem that draws researchers’ vision, due to nodes mobility, dynamic topology, and lack of central point like base station or servers. Clustering of devices in MANET could reduce overhead, flooding, and collision in communication and make the network topology more stable. The ABC (Artificial Bee Colony) algorithm is a new meta-heuristic population based optimization technique inspired by the intelligent foraging behavior of honeybee swarms. Clustering provides scaling and eases in routing with efficient resource management in MANETs.
Forecasting the PV Power Utilizing a Combined Convolutional Neural Network and Long Short-Term Memory Model
Published in Electric Power Components and Systems, 2023
Ramakrishnan Raman, Bhaveshkumar Mewada, R. Meenakshi, G. M. Jayaseelan, K. Soni Sharmila, Syed Noeman Taqui, Essam A. Al-Ammar, Saikh Mohammad Wabaidur, Amjad Iqbal
The proposed training process for solar power generation forecasting can be converged through the hybrid method based on LSTM and CNN models by using an artificial bee colony for the training procedure. The combination of LSTM and CNN models allows for the capture of both long-term dependencies and spatial information in the data. LSTM models are effective in learning long-term dependencies, while CNN models are useful for identifying spatial patterns in the data. The hybrid model can, therefore, improve the accuracy of solar power generation forecasting. In addition to the LSTM and CNN models, the use of an artificial bee colony algorithm in the training procedure can also aid in convergence. The artificial bee colony algorithm is a swarm intelligence-based optimization algorithm that mimics the behavior of honeybees in nature. The algorithm can be used to optimize the parameters of the hybrid model, which can lead to more accurate forecasting results. Overall, the proposed hybrid method of using LSTM and CNN models along with an artificial bee colony for the training procedure can improve the accuracy of solar power generation forecasting by capturing both long-term dependencies and spatial information in the data while also aiding in convergence.
Grey wolf optimization for optimum sizing and controlling of a PV/WT/BM hybrid energy system considering TNPC, LPSP, and LCOE concepts
Published in Energy Sources, Part A: Recovery, Utilization, and Environmental Effects, 2022
Abdulsamed Tabak, Erhan Kayabasi, Muhammet Tahir Guneser, Mehmet Ozkaymak
As the hybrid systems get complicated, it is necessary to use various optimization algorithm methods to obtain the most reliable, cheapest, and sustainable design. Artificial intelligence has been used extensively in the optimization of hybrid systems in the last decade because it provides faster problem solution and more accurate results (Zahraee, Khalaji Assadi, and Saidur 2016). In this context, some optimization algorithms such as artificial bee colony algorithm (Nasiraghdam and Jadid 2012), genetic algorithm (GA) (Kalantar and Mousavi 2010), harmony search (HS) (Rojas, Dufo-López, and Bernal-Agustín 2012), simulated annealing (SA) (Ghazvini, Abbaspour-Tehrani-fard, and Fotuhi-Firuzabad 2015), and particle swarm optimization (PSO) (Mezzai et al. 2014) are used in optimization of hybrid systems including PV, WT, batteries, supercapacitors, diesel generators, and fuel cells. However, there is a wide gap about the studies on hybrid systems and their optimization in the literature including biomass (BM) (Suganthi, Iniyan, and Samuel 2015). Both GA and SA are frequently used in optimization problems but the main disadvantage of them is slow problem solving (Bevilacqua 2002; Yahiaoui et al. 2017). Moreover, although SA is used for different problem types successfully, it is not suitable for complex and scheduling problems (Bevilacqua 2002).
Reconstruction of the Boundary Condition in the Binary Alloy Solidification Problem with the Macrosegregation and the Material Shrinkage Phenomena Taken into Account
Published in Heat Transfer Engineering, 2021
Adam Zielonka, Edyta Hetmaniok, Damian Słota
Artificial Bee Colony algorithm is the heuristic optimiziation algorithm simulating the behavior of bees looking for the nectar around the hive and sharing the information about the location of the most promising sources of nectar with other members of the swarm. At first the bees-scouts explore the neighborhood of the hive, localize the best sources of food, collect the nectar from the chosen flowers and return to the hive to inform the other bees about their discoveries. The act of sharing the information is executed in the form of a special kind of dance, called the waggle dance, taking place in some special spots of the hive. After the dance the other part of the bees, the bees-viewers, leave the hive and fly to these best sources of food, discovered by the bees-scouts, to take the nectar.