This paper proposes an efficient training approach for support vector regression,called weighted constraints SVR(WCSVR),which will improve the ability to learn from the noisy samples and overcome the overfitting *** t...
详细信息
This paper proposes an efficient training approach for support vector regression,called weighted constraints SVR(WCSVR),which will improve the ability to learn from the noisy samples and overcome the overfitting *** the WCSVR learning process,the object function is modified by introducing a weight for each new training sample and the corresponding regularization *** weighted constraints value to each input sample is confirmed according to its kernel Mahalanobis distance to the center of the *** parameters help us to apply the importance of samples for eliminating the effect of the outliers in the *** overall experiments show prominence of out proposed method in comparison with the standard SVR.
In China, there are many actual problems in some enterprises' performance management, such as the target dislocation, the unscientific assessment indicator, the assessment results' not linking to rewards and p...
详细信息
Traditional Modbus/TCP server-side protocol includes in many inappropriate blocking I/O operations. These operations frequently lead to receiving thread blockage in the communication. To solve this problem, some impro...
详细信息
Traditional Modbus/TCP server-side protocol includes in many inappropriate blocking I/O operations. These operations frequently lead to receiving thread blockage in the communication. To solve this problem, some improvements such as asynchronous non-blocking and multi-threading technology are done on the TCP network layer of Modbus/TCP, which are completed in S3C2440 server-side by Visual Studio 2005. Furthermore, TCP / IP communications between server terminal and multiple clients are tested. The results show that the ping response time and the data sending or receiving time is less than 1ms and 1s respectively. Finally, a new TCP packet format in the process of TCP network communications is customized to carry out the Modbus/TCP functions on the Windows *** platform, so that a new method is established to provide an efficient and real-time master-slave dialogue.
As an important unit composing Wireless Sensor Networks, sensor node normally integrates sensor, data processing unit and communicating module. Power consuming and data transmitting rate are the key factors affecting ...
详细信息
As an important unit composing Wireless Sensor Networks, sensor node normally integrates sensor, data processing unit and communicating module. Power consuming and data transmitting rate are the key factors affecting the quality of it. Regarding the characteristics mentioned, we designed a new wireless sensor network node based on nRF2401. Since the node using low power consumption processor ATmega88 and wireless communication module nRF2401, it guarantees the system operating life and data transmission speed efficiently. Furthermore, the digital thermometer DS18B20 based on 1-wire technology can give the system more concision and more reliability. This node easily applies to Wireless Sensor Networks widely, with its characteristics of low-power, miniaturization and high integration.
It is shown in this paper that the quality of control in iron and steel process, hinges upon how well various disturbances are rejected. Characteristics of the disturbances and the current design solutions are present...
详细信息
作者:
Gong KunDeng FangMa TaoGong Kun is with School of Automation
Beijing Institute of Technology and Key Laboratory of Advanced Control of Iron and Steel Process (Ministry of Education) Beijing China Deng Fang is with School of Automation
Beijing Institute of Technology and Key Laboratory of Advanced Control of Iron and Steel Process (Ministry of Education) Beijing China Ma Tao is with School of Automation
Beijing Institute of Technology and Key Laboratory of Complex System Intelligent Control and Decision Ministry of Education Beijing China
In order to improve the precision of the azimuth measured by mobile robot's electronic compass, this paper proposes a new calibration method based on Fourier Neural Network trained by Modified Particle Swarm Optim...
详细信息
In order to improve the precision of the azimuth measured by mobile robot's electronic compass, this paper proposes a new calibration method based on Fourier Neural Network trained by Modified Particle Swarm optimization (MPSO-FNN). This method makes use of Fourier Neural Network (FNN) to establish the error compensation model of electronic compass's azimuth, and introduces Modified Particle Swarm optimization (MPSO) algorithm to optimize the weights of neural network. Thus the comparatively accurate error model of azimuth is obtained to compensate the output of electronic compass. This method not only has strong nonlinear approximation capability, but also overcomes the neural networks' shortcomings which are too slow convergence speed, oscillation, and easy to fall into local optimum and sensitive to the initial values. Experimental results demonstrate that after calibrated by this method, the range of azimuth error reduces to -0.35°~0.70° from -3.4°~25.2°, and the average value of absolute error is only 0.30°.
Service supply chain features human players as service vendor, service integrator, customer and service resource. It tends to be digitally connected, such as consulting, e-business and integrated enterprises. Our stud...
详细信息
Service supply chain features human players as service vendor, service integrator, customer and service resource. It tends to be digitally connected, such as consulting, e-business and integrated enterprises. Our study uses a formal model and simulations to develop the effect of a service supply chain on equilibrium computation. Two insights arise on how a network can obtain equilibrium computation: forming the network structure of service supply chain; exploring entities behavior and equilibrium conditions. These results highlight the importance for service supply chain of adapting its network structure to equilibrium and application.
To solve the strong nonlinearity and data deterioration due to missing, outliers contained in the training data, this paper combines robust EMPCA (Expectation Maximization Principle Component Analysis) and the error-b...
详细信息
An evolutionary algorithm based on the parallel evolution of multiple single objective populations and Pareto archive population is proposed, which is not only suitable for solving multi-objective optimization, but al...
详细信息
An evolutionary algorithm based on the parallel evolution of multiple single objective populations and Pareto archive population is proposed, which is not only suitable for solving multi-objective optimization, but also effective for multimodal function. For each single objective population, single objective evolutionary algorithm is applied to optimize separately each of multi-objective functions, where individuals generated by tournament selection from the union of single objective and Pareto Archive population form the single objective population of next generation. Especially, individuals in Pareto archive population also join evolutionary operations. Simulations manifest that the proposed method can realize the search from multiple directions to obtain the non-dominated solutions scattered more uniformly over the Pareto frontier with better convergence metric compared to well-known NSGA-II algorithm. Individual migration from Pareto archive population by tournament selection is also proved to have the advantage in improving the converging speed and converging precision. Moreover, for multimodal single objective function, simulations also show that ideal optimizing solution can be obtained by properly separating single objective function into multi-objective function and applying the above method.
Radial basis function (RBF) network is one of the significant neural networks. It has been used successfully in various fields. But in RBF network approximation algorithm, the initial value of the network weights, Gau...
详细信息
暂无评论