Advanced Materials Research Vols. 108-111

Abstract: T Washout algorithm parameters based on flight fidelity in pitch task are optimized. After a review of classical washout algorithm, optimization requirement and function are specified. The results of an example with pitch task show that the optimization process can reduce the difficulty of manipulating vehicle, and decrease the motion system requirement.

Abstract: A new active learning based on a new localized generalization error model is proposed in the paper for training RBFNN. The samples with largest local generalization error are selected and labelled. The experimental results show that the proposed algorithm is effective, which can select the most informative samples and fewer samples are necessary.

Abstract: This paper presents a Fuzzy Neural Network-based irrigation water control system for realizing accurate irrigation and water saving. The control system consists of a PC based server with a FNN-based control module, data sampling device, CAN bus communication network and remote irrigation units with sensors and controller. It can estimate irrigation water compensation according to the irrigation environments. By using Fuzzy Logic function and Fuzzy Neural Network, the controller module can fuse multi-source information that from Internet, sensors and other sources to make decision of irrigation water compensation on the control spot. Experimental result indicates that the compensation generated by FNN have higher precision than that by FL in this irrigation water control system.

Abstract: Centralized cluster system is widely used by many web sites to increase service availability and balancing workload among multiple servers. In order to reduce the possibility which balancer becomes performance bottleneck in clusters, a dynamic adaptive weighted load-balanced algorithm is presented in this paper. The objective of the algorithm is that every working server sends a corresponding ask signals at different rates to balancer based on the discrete uniform distribution principle and the balancer assigning user requests according to an ask queue which consists of these signals and achieves the dynamic negative feedback on workload. Experiment results show that the algorithm is better than the static algorithm, as compared with the dynamic algorithm, it can also reduce the burden on load balancer.

Abstract: A hybrid intelligent technique is proposed to identify Bouc-Wen hysteresis model parameters. This intelligent technique is based on a hybrid of genetic algorithm (GA) and Levenberg–Marquardt algorithm (LMA). In the hybrid intelligent technique, the GA, a popular evolutionary optimization method, firstly searches the entire problem space to get a set of roughly estimated solutions. The LMA, a well-known numerical method, then performs a local optima search in order to carry out further optimizations. The performance of the hybrid intelligent technique is compared with GA method in terms of parameter accuracy. The simulation experiments of Bouc-Wen hysteresis model with known parameters are illustrated to show that a high quality solution can be achieved by means of the hybrid intelligent technique. The concept of hybrid intelligent technique may benefit the parameter identification in diverse hysteresis model problems.

Abstract: Logistics domain ontology is the foundation of logistics information sharing. According to the knowledge of logistics, an eABC ontology model was be extended from ABC ontology, and a logistics domain ontology model (LDOM) based eABC was constructed. Then relations among logistics domain knowledge were represented in this model, and five key ontology elements of LDOM were described, such as concept, relation, function, axiom and instance. Finally, LDOM was validated by the analysis of a case.

Abstract: Accord to the vibration of the floor of one light bus ,the analysis based on experiment is done to confirm that the vibration source is resonance of the subframe. The modal analysis and optimal design is done with Hyper-OptiStruct .The experimental validation of real car is completed and the results show that the optimization is effective and reasonable

Abstract: The flood of spam promotes the development of anti-spam technology. In this paper, we bring forward the Bayesian filter technology based on the distributed architecture, which can realize the sharing of the Bayesian learning outcomes among servers within the system, so as to increase the accuracy of spam recognition. We, in the paper, discuss the sharing model of information with spam features under the distributed architecture and the spam identification process; analyze the Bayes algorithm and carry out the relevant improvements; design the Bayes Filter based on distributed architecture on the above basis and verify the effect of the filter by experiments.

Abstract: The flood of spam has caused more serious harm in society. In this paper, we bring forward an anti-spam system based on the service grid, which can dynamically organize the mail servers on the Internet together to against the spam. We discuss the working model and system architecture of the anti-spam grid and design the core algorithm for the status monitoring and load distribution of a dispatch server. And In this paper, we implement the system performance evaluation, and the experiment results show that the system is stable, efficient and scalable.

Abstract: Traditional load balancing strategy is based on the assumption that each node is valid ,in practice, with the growth of usage time, the failure rate of nodes will gradually increase, so we introduce software aging rejuvenation strategy into load balancing. We propose an algorithm to solve this problem. It is different from other algorithm that the function of collecting load information from real servers is moved to back-end in this algorithm, load balancer assigns tasks according to the calculation results, it could reduce the load of load balancer own. At the same time, in order to prevent unexpected event due to computer itself (e.g. cpu utilization rate sudden change), we also take into account decay factor. From the experiment result we can see that compared with non software aging rejuvenation strategy in load balancing, this strategy could balance the load of all the nodes more effectively.