Search

The majority of large-scale data intensive applications designed by MapReduce model are deployed and executed on a large-scale distributed Hadoop system. Running such application on a large cluster requires a large amount of energy. Therefore improving energy efficiency and minimizing energy consumption when executing each MapReduce job is a critical concern for data centers. We propose a control model based on model prediction control (MPC) for improving energy efficiency of Hadoop cluster while satisfying performance goal. The controller can dynamically scale CPU frequency in facing the workloads changes. The wavelet neural network is utilized to build the prediction model, which can accurately estimate energy efficiency ratio based on the features extracting from the variation of Hadoop workloads. With the help of a wavelet neural network, the controller can guarantee the energy efficiency ratio converges to an ideal set point within a finite settling time. We perform the extensive experiments on a Hadoop cluster to compare our controller with the previous MPC controller. The results demonstrate that our controller outperforms the baseline on energy efficiency control and adapts to workloads of Hadoop cluster.

The modular multilevel converter (MMC) has the advantages of high efficiency, low harmonic, modular design, and easy cascade, which has been widely used in the field of high voltage and large capacity energy conversion. In order to improve the dynamic response speed of the MMC-HVDC based on the modular multi-level converter, a novel method which combines a model predictive control (MPC) of MMC-HVDC system with improved sub-module voltage balanced control strategy is proposed. The method, which utilises the prediction model, feedback correction and rolling optimisation to obtain the optimal voltage control, overcomes the difficulties in the traditional way of setting PI parameters of the internal loop current controller and the outer loop controller and tackles the problem of low dynamic response. Finally, a 21-level MMC-HVDC system simulation model is built on PSCAD-EMTDC software platform. The simulation results indicate the effectiveness and feasibility of the control strategy.

Cooperative terminal guidance with impact angle constraint is a key technology to achieve a saturation attack and improve combat effectiveness. The present study envisaged cooperative terminal guidance with impact angle constraint for multiple missiles. In this pursuit, initially, the three-dimensional cooperative terminal guidance law with multiple constraints was studied. The impact time cooperative strategy of virtual leader missile and follower missiles was designed by introducing virtual leader missiles. Subsequently, based on the distributed model prediction control combined with the particle swarm optimization algorithm, a cooperative terminal guidance algorithm was designed for multiple missiles with impact angle constraint that met the guidance accuracy. Finally, the effectiveness of the algorithm was verified using simulation experiments.