In this paperan algorithm based on particle swarm optimization algorithm for RBF neural network is propose. With particle swarm optimization algorithm, neural network weights are optimized. Also through the dynamic re...
详细信息
ISBN:
(纸本)9781424446421
In this paperan algorithm based on particle swarm optimization algorithm for RBF neural network is propose. With particle swarm optimization algorithm, neural network weights are optimized. Also through the dynamic regulation of the number of radial basis function in neural network hidden layer, neural network structure is optimized. The algorithm is applied to gearbox fault diagnosis. Experimental results show the effectiveness and great performance. Classification effect of neural network based on particle swarm optimization algorithm is better than that of the RBF neural network for identifying effectively the different status of gearbox and monitoring timely the status changes of gearbox. Also it can reduce the time for fault diagnosis and improve accuracy of fault diagnosis.
A nonlinear ensemble prediction model for typhoon rainstorm has been developed based on particleswarmoptimization-neural network (PSO-NN). In this model, PSO algorithm is employed for optimizing the network structur...
详细信息
A nonlinear ensemble prediction model for typhoon rainstorm has been developed based on particleswarmoptimization-neural network (PSO-NN). In this model, PSO algorithm is employed for optimizing the network structure and initial weight of the NN with creating multiple ensemble members. The model input of the ensemble member is the high correlated grid point factors selected from the rainfall forecast field of Japan Meteorological Agency numerical prediction products using the stepwise regression method, and the model output is the future 24 h rainfall forecast of the 89 stations. Results show that the objective prediction model is more accurate than the numerical prediction model which is directly interpolated into the stations, so it can better been implemented for the interpretation and application of numerical prediction products, indicating a potentially better operational weather prediction.
A new fuzzy identification approach using support vector regression (SVR) and particle swarm optimization algorithm (PSOA) is presented in this paper. Firstly positive definite reference function is utilized to constr...
详细信息
ISBN:
(纸本)9781424442461
A new fuzzy identification approach using support vector regression (SVR) and particle swarm optimization algorithm (PSOA) is presented in this paper. Firstly positive definite reference function is utilized to construct a qualified Mercer kernel for SVR. Then an improved PSOA is developed for parameters selection of SVR, in which the number of support vectors and regression accuracy are regarded simultaneously to guarantee the conciseness of the constructed fuzzy model. Finally, a set of TS fuzzy rules can be extracted from the SVR directly. Simulation results show that the resulting fuzzy model not only costs less fuzzy rules, but also possesses good generalization ability.
To realize the adaptive adjustment of the air-based pseudo-lite (PL) navigation augmentation network, a dynamic configuration method for the air-based PL network deployment is proposed. By surveying and defining the i...
详细信息
ISBN:
(纸本)9783662466353;9783662466346
To realize the adaptive adjustment of the air-based pseudo-lite (PL) navigation augmentation network, a dynamic configuration method for the air-based PL network deployment is proposed. By surveying and defining the indicators describing the performance of navigation augmentation network, based on the basic needs of navigation enhancement task, an objective function is designed for the optimization deployment of the air-based PLs. Using particle swarm optimization algorithm (PSO) for the fine search of the objective function of the air-based PL optimization deployment in multi-dimension, the dynamic optimal deployment is obtained. The simulation results show that, the proposed method could keep the navigation performance optimum, in the case of navigation enhancement service region changing or the local PL being interfered.
The traditional least squares support vector regression (LSSVR) node localization algorithm for wireless sensor networks(WSNs) uses the average hop distance to calculate the actual distance, which may result larger lo...
详细信息
ISBN:
(纸本)9781467371438
The traditional least squares support vector regression (LSSVR) node localization algorithm for wireless sensor networks(WSNs) uses the average hop distance to calculate the actual distance, which may result larger localization error in the obstacle conditions. An improved LSSVR WSNs three-dimensional mobile node localization method in an obstacle conditions was proposed in this paper. The average per hop distance of four anchor nodes closest was used to replace the average distance per hop of traditional LSSVR algorithm in the proposed method, and the new average per hop distance was used to calculate the measurement distance of each unknown node to anchor nodes. The LSSVR localization model was built through sampling of the grid and constructing the training sets. According to mean square deviation of predicted location of virtual nodes and their actual location, fitness function was constructed, and LSSVR kernel function and regularization parameters were optimized by the PSO algorithm. The simulation results show that, compared with the conventional LSSVR localization algorithm, the proposed localization algorithm has a higher localization accuracy, smaller localization errors and lower localization cost in the obstacle conditions.
A novel structure of dynamic model is proposed in this paper and applied to construct a dynamic model to correct the dynamic errors of the infrared thermometer, because of which the dynamic performance of the thermome...
详细信息
ISBN:
(纸本)9781424445189
A novel structure of dynamic model is proposed in this paper and applied to construct a dynamic model to correct the dynamic errors of the infrared thermometer, because of which the dynamic performance of the thermometer is effectively improved. The dynamic compensator is established and the compensation is described and explicated by the Wiener model. According to Wiener model, the novel structure is devised. The identification of thermometer non-linear dynamic compensator is achieved by particle swarm optimization algorithm. The results show that the stabilizing time of the thermometer is reduced less than 7 ms from 26 ms and the dynamic performance is obviously improved after compensation.
As a necessary supporting infrastructure in development of electric vehicles, electric vehicle charging stations can provide electric vehicles charging service. Their locations are reasonable or not directly related t...
详细信息
ISBN:
(纸本)9781479918911
As a necessary supporting infrastructure in development of electric vehicles, electric vehicle charging stations can provide electric vehicles charging service. Their locations are reasonable or not directly related to the development of the electric vehicle industry and their service quality, efficiency, convenience, etc. On the basis of flow capturing location model, this paper treats intercepting the largest demand as the goal and uses particle swarm optimization algorithm to simulate the model to prove its effectiveness and practicability. The simulation result shows that the model and algorithm used in this paper can establish an optimal selection of location, and be able to provide some decisions to the real facility location.
Recent developments in the stock market have created an urgent need for efficient methods to help stockholders take appropriate decisions about their stocks. Since large fluctuations occur in the stock market over tim...
详细信息
ISBN:
(纸本)9781467365062
Recent developments in the stock market have created an urgent need for efficient methods to help stockholders take appropriate decisions about their stocks. Since large fluctuations occur in the stock market over time and there are many parameters which influence this, it seems difficult to make good decisions that are also well-timed. The purpose of this study is to apply artificial neural networks (ANNs), which can deal with time series data and nonlinear parameters, to predict the next day's stock price. This research has trained the proposed ANN with a meta-heuristic bat algorithm which has a fast and powerful convergence. The recommended method has been applied to stock price forecasting for the first time. This work has used a seven-year dataset of a private bank stocks in order to prove the performance of the suggested method. After data pre-processing, three types of ANNs (back propagation-ANN, particleswarmoptimization-ANN and bat-ANN) were employed to predict the stocks' closing price. Afterwards, MATLAB was used to evaluate the performance of these three methods by scoring the target of the mean absolute percentage error (MADE). This paper indicates that the bat algorithm adjusts the weight matrix of ANN more precisely than the two other algorithms. The results may be adapted to other companies' stocks.
In the late three decades, grid computing has emerged as a new field providing a high computing performance to solve larger scale computational demands. Because Directed Acyclic Graph (DAG) application scheduling in a...
详细信息
ISBN:
(数字)9783319257440
ISBN:
(纸本)9783319257440;9783319257433
In the late three decades, grid computing has emerged as a new field providing a high computing performance to solve larger scale computational demands. Because Directed Acyclic Graph (DAG) application scheduling in a distributed environment is a NP-Complete problem, meta-heuristics are introduced to solve this issue. In this paper, we propose to hybridize two well-known heuristics. The first one is the Heterogeneous Earliest Finish Time (HEFT) heuristic which determines a static scheduling for a DAG in a heterogeneous environment. The second one is particleswarmoptimization (PSO) which is a stochastic meta-heuristic used to solve optimization problems. This hybridization aims to minimize the makespan (i.e., overall completion time) of all the tasks within the DAG. The experimental results that have been conducted under hybridization show that this approach improves the scheduling in terms of completion time compared to existing algorithms such as HEFT.
This paper proposes a novel application of a chaos particleswarmoptimization (PSO) algorithm for economic dispatch (ED) of a hybrid power system including the solar and wind energy sources. The algorithm is seeking ...
详细信息
ISBN:
(纸本)9781509012374
This paper proposes a novel application of a chaos particleswarmoptimization (PSO) algorithm for economic dispatch (ED) of a hybrid power system including the solar and wind energy sources. The algorithm is seeking to minimize total operating costs of the hybrid power system. The proposed chaos PSO algorithm is one of the standard PSO algorithm variants which has been used a logistic map for initializing random values of generators, as well as the inertia weight in the velocity update equation of the standard PSO algorithm. This results in the best convergence capability and search performance during the evolution process of the algorithm. The chaos PSO algorithm based ED problem of the hybrid power system with and without solar and wind powers is considered on a standard IEEE 30-bus 6-generator 41-transmission line test power system. The simulation results demonstrate the capabilities of the proposed algorithm to generate optimal dispatch solutions of the ED problem considering the renewable energy resources. The comparison with the standard PSO algorithm demonstrates the superiority of the proposed algorithm and confirms its potential to solve the ED problem.
暂无评论