作者:
Akin Sherly, L. T.Jaya, T.Anna Univ
Maria Coll Engn & Technol Dept Elect & Commun Engn Chennai Tamil Nadu India Anna Univ
Dept Elect & Commun Engn CSI Inst Technol Chennai Tamil Nadu India
The most common and challenging issues in image recognition are scene character recognition from the street view image, and the scene character consists of both text and number. Recently, the researchers were introduc...
详细信息
The most common and challenging issues in image recognition are scene character recognition from the street view image, and the scene character consists of both text and number. Recently, the researchers were introduced a lot of scene character recognition methods, but the performance of the methods often degraded due to complexity. So, we proposed the improved firefly algorithm for local trapping problem (IFLT) utilizing convolutional neural network (CNN) for the extraction of features from the scene character. The IFLT approach is the improved version of the firefly optimization algorithm to solve local trapping problems. During feature extraction, the hyperparameters on CNN are tuned with the help of the IFLT approach. The alignment and multilayer perceptron layers are used on CNN. Subsequently, the support vector machine approach is used to classify the relevant class of scene characters from the street view image. Experimentally, we use six scene character dataset SVHN, ISN, IIIT5K-words, SVT, ICDAR 2003, and ICDAR 2013 dataset. The performance of the proposed IFLT approach is evaluated with standard deviation, mean, average computational time, and most excellent minimum (MEmin) parameters. The experimental results demonstrate that the proposed IFLT-CNN is well suitable for scene character recognition.
In this paper, a two-control variable controller based on the Proportional-Integral-Derivative Neural Network (PIDNN) is designed to control a certain type of mixed exhaust turbofan engine. According to the working pr...
详细信息
ISBN:
(纸本)9789881563804
In this paper, a two-control variable controller based on the Proportional-Integral-Derivative Neural Network (PIDNN) is designed to control a certain type of mixed exhaust turbofan engine. According to the working principle of the aero-engine, a two-variable small deviation state model of the aero-engine is firstly established. Then PIDNN including an input layer, a hidden layer and an output layer is used to design the controller of aero-engine states model. There are 4 nods in input layer, 6 in hide layer and 2 in output layer. To solve the problems of large steady-state error and long adjustment time of the PIDNN controller, this paper uses the improved firefly algorithm to dynamically adjust the initial connection weights of the PIDNN. The results show that the established aero-engine PIDNN controller based on the improved firefly algorithm has the characteristics of short adjustment time and high accuracy, which meets the requirements of aero-engine controller design.
Reasonable distribution of cooling load between chiller and ice tank is the key to realize the economical and energy-saving operation of ice-storage air-conditioning (ISAC) system. A multi-objective optimization model...
详细信息
Reasonable distribution of cooling load between chiller and ice tank is the key to realize the economical and energy-saving operation of ice-storage air-conditioning (ISAC) system. A multi-objective optimization model based on improved firefly algorithm (IFA) was established in this study to fully exploit the energy-saving potential and economic benefit of the ISAC system. The proposed model took the partial load rate of each chiller and the cooling ratio of the ice tank as optimization variables, and the lowest energy consumption loss rate and the lowest operating cost of the ISAC system were calculated. Chaotic logic self-mapping was used to initialize population to avoid falling into local optimum, and Cauchy mutation was used to increase the population's diversity to improve the algorithm's global search ability. The experimental results show that compared with the operation strategy based on constant proportion, particle swarm optimization (PSO) algorithm, and fireflyalgorithm (FA), the optimal operation strategy based on IFA can achieve more significant energy-saving and economic benefits. Meanwhile, the convergence accuracy and stability of the algorithm are significantly improved. Practical application: The optimized operation strategy of the ice-storage air-conditioning system can reduce energy loss and operating costs. The traditional operation strategies have the problems of low optimization precision and poor optimization effect. Therefore, this study presents an optimal operation strategy based on IFA. The convergence accuracy and stability of the algorithm are increased after the algorithm is improved. The operation strategy can get the maximum energy-saving effect and economic benefit of the ISAC system.
Thermal conductivity is a specific thermal property of soil which controls the exchange of thermal energy. If predicted accurately, the thermal conductivity of soil has a significant effect on geothermal applications....
详细信息
Thermal conductivity is a specific thermal property of soil which controls the exchange of thermal energy. If predicted accurately, the thermal conductivity of soil has a significant effect on geothermal applications. Since the thermal conductivity is influenced by multiple variables including soil type and mineralogy, dry density, and water content, its precise prediction becomes a challenging problem. In this study, novel computational approaches including hybridisation of two metaheuristic optimisation algorithms, i.e. fireflyalgorithm (FF) and improved firefly algorithm (IFF), with conventional machine learning techniques including extreme learning machine (ELM), adaptive neuro-fuzzy interface system (ANFIS) and artificial neural network (ANN), are proposed to predict the thermal conductivity of unsaturated soils. FF and IFF are used to optimise the internal parameters of the ELM, ANFIS and ANN. These six hybrid models are applied to the dataset of 257 soil cases considering six influential variables for predicting the thermal conductivity of unsaturated soils. Several performance parameters are used to verify the predictive performance and generalisation capability of the developed hybrid models. The obtained results from the computational process confirmed that ELM-IFF attained the best predictive performance with a coefficient of determination = 0.9615, variance account for = 96.06%, root mean square error = 0.0428, and mean absolute error = 0.0316 on the testing dataset (validation phase). The results of the models are also visualised and analysed through different approaches using Taylor diagrams, regression error characteristic curves and area under curve scores, rank analysis and a novel method called accuracy matrix. It was found that all the proposed hybrid models have a great ability to be considered as alternatives for empirical relevant models. The developed ELM-IFF model can be employed in the initial stages of any engineering projects for fast det
Coordinate Measurement Machines (CMMs) have been extensively used in inspecting mechanical parts with higher accuracy. In order to enhance the efficiency and precision of the measurement of aviation engine blades, a s...
详细信息
Coordinate Measurement Machines (CMMs) have been extensively used in inspecting mechanical parts with higher accuracy. In order to enhance the efficiency and precision of the measurement of aviation engine blades, a sampling method of profile measurement of aviation engine blade based on the fireflyalgorithm is researched. Then, by comparing with the equal arc-length sampling algorithm (EAS) and the equi-parametric sampling algorithm (EPS) in one simulation, the proposed sampling algorithm shows its better sampling quality than the other two algorithms. Finally, the effectiveness of the algorithm is verified by an experimental example of blade profile. Both simulated and experimental results show that the method proposed in this paper can ensure the measurement accuracy by measuring a smaller number of points.
Complementary multi-energy power generation systems are a promising solution for multi-energy integration and an essential tool for diversifying renewable energy sources. Despite many studies on developing hybrid rene...
详细信息
Complementary multi-energy power generation systems are a promising solution for multi-energy integration and an essential tool for diversifying renewable energy sources. Despite many studies on developing hybrid renewable energy systems, more research is still needed on applicable models or practical methods. Meta-heuristic algorithms such as the fireflyalgorithm are becoming increasingly popular in optimizing hybrid renewable energy systems because they provide fast, accurate, and optimal solutions. Considering the natural complementarity and instability of wind and solar energy, the advantage of pumped storage power plants' "peak adjustment and valley adjustment", as well as the grid's need for a stable and reliable energy supply, the objective of this study is to economically optimize the design of wind-PV pumped storage complementary generation system scheduling with a two-generation fireflyalgorithm based on spatial adaptive and Levy's flight improvement, in comparison with a variety of cutting-edge population intelligence optimization algorithms (GA, PAO, DE, WOA, FA) were compared and analyzed. The impact of pumped storage plants on economic and stabilization objectives is explored. The results show that several meta-heuristics are effective in finding the optimal design. However, the improved firefly algorithm with an objective function value of 7.8331 is superior to several other algorithms by enhancing the wind and PV benefits while suppressing the output fluctuations of the system. After the construction of the additional pumped storage plant, the output fluctuation of the complementary operation system is only 9.7% of that of the wind power and PV in stand-alone operation after the multi-energy coordination and optimal scheduling. This demonstrates the effectiveness of the optimization method used in this paper. The results of this study can provide a reference for the complementary optimization of pumped storage power plants for intermittent renewabl
As an important part of container logistics, quay cranes (QCs) are crucial equipment in multimodal container transportation. The scheduling and allocation of QCs determine the operational efficiency of container termi...
详细信息
As an important part of container logistics, quay cranes (QCs) are crucial equipment in multimodal container transportation. The scheduling and allocation of QCs determine the operational efficiency of container terminals. By analyzing the way quay cranes are operated, this paper establishes a mixed-integer dynamic rolling-horizon programming model for the scheduling and allocation of QCs and proposes use of a genetic algorithm and two improved firefly algorithms based on segment encoding technology to formulate an optimum QC scheduling scheme. In doing so, the improved approach has made QC control more efficient and balanced.
When photovoltaic (PV) arrays operate under the partial shadow, the output characteristic curve has multiple peaks, and it is difficult for traditional algorithms to track the global maximum power point (GMPP). To sol...
详细信息
ISBN:
(纸本)9781665434980
When photovoltaic (PV) arrays operate under the partial shadow, the output characteristic curve has multiple peaks, and it is difficult for traditional algorithms to track the global maximum power point (GMPP). To solve this problem, this paper proposes an improved firefly algorithm (IFA), which introduces an adaptive step size based on the traditional fireflyalgorithm (FA). In the early stage of algorithm operation, the larger step size is adopted, in the later stage of the algorithm operation, the smaller step size is taken, so that the algorithm can quickly and accurately realize the PV global maximum power point tracking (GMPPT). A simulation model is built in Matlab/Simulink, the results under static and dynamic conditions prove the effectiveness of the improvedalgorithm.
Ships oscillate periodically while navigating at sea. To study the impact of the ocean environment on ship transportation, it is necessary to build accurate mathematical models of ship motion. Based on the theory of r...
详细信息
ISBN:
(纸本)9798350366907;9789887581581
Ships oscillate periodically while navigating at sea. To study the impact of the ocean environment on ship transportation, it is necessary to build accurate mathematical models of ship motion. Based on the theory of rigid body motion, build the mathematical models for ship heave and pitch, and discretize them;By introducing sigmoid function, roulette wheel strategy, and chaos theory, an improved Chaotic fireflyalgorithm Based on Roulette Wheel Selection Strategy (CRSFA) is proposed, which utilizes representative unimodal and multimodal test functions to optimize the algorithm through different dimensions. Compared with the original algorithm, the improved firefly algorithm based on Roulette Wheel Selection Strategy (CRSFA) proposed effectively avoids algorithm oscillations in this paper, accelerates algorithm convergence speed, and improves the algorithm's global searching ability. Based on the CFD numerical simulation data of the KCS ship model in calm water and 2 to 3 level sea conditions, parameter identification are carried out for the heave and pitch motion of the KCS ship model. The identified parameters are used to simulate and predict the heave and pitch motion of the ship using CRSFA and original FA separately. The simulation results show that fitted with the real time motion trend, verifying the feasibility of the identification scheme proposed in this paper.
In this paper, a two-control variable controller based on the Proportional-Integral-Derivative Neural Network(PIDNN)is designed to control a certain type of mixed exhaust turbofan engine. According to the working pr...
详细信息
In this paper, a two-control variable controller based on the Proportional-Integral-Derivative Neural Network(PIDNN)is designed to control a certain type of mixed exhaust turbofan engine. According to the working principle of the aero-engine, a two-variable small deviation state model of the aero-engine is firstly established. Then PIDNN including an input layer, a hidden layer and an output layer is used to design the controller of aero-engine states model. There are 4 nods in input layer, 6 in hide layer and 2 in output layer. To solve the problems of large steady-state error and long adjustment time of the PIDNN controller,this paper uses the improved firefly algorithm to dynamically adjust the initial connection weights of the PIDNN. The results show that the established aero-engine PIDNN controller based on the improved firefly algorithm has the characteristics of short adjustment time and high accuracy, which meets the requirements of aero-engine controller design.
暂无评论