We consider the global energy efficiency (GEE) maximization problem for the general non-regenerative MIMO relay network, under the maximum power constraints for each user and each relay. The problem is reformulated th...
详细信息
ISBN:
(纸本)9781509041183
We consider the global energy efficiency (GEE) maximization problem for the general non-regenerative MIMO relay network, under the maximum power constraints for each user and each relay. The problem is reformulated through the fractional optimization technique, and the mean square error receiver filter is considered. By applying the alternating minimization method, we simplify the problem into several convex quadratic constrained quadraticprogramming subproblems, and solve the subproblems by the feasible shrinkage method combined with the sequential quadratic programming method. Because the result highly depends on the initialization, we design a deterministic initialization by introducing an auxiliary power minimization problem. Simulation results show that our proposed algorithm can achieve more than 10 times higher GEE than the previous works which are not tailored for GEE maximization.
Discrete mechanics and optimal control (DMOC) is a recent development in optimal control of mechanical systems that takes advantage of the variational structure of mechanics when discretizing the optimal control probl...
详细信息
ISBN:
(纸本)9781424477456
Discrete mechanics and optimal control (DMOC) is a recent development in optimal control of mechanical systems that takes advantage of the variational structure of mechanics when discretizing the optimal control problem. Typically, the discrete Euler-Lagrange equations are used as constraints on the feasible set of solutions, and then the objective function is minimized using a constrained optimization algorithm, such as sequential quadratic programming (SQP). In contrast, this paper illustrates that by reducing dimensionality by projecting onto the feasible subspace and then performing optimization, one can obtain significant improvements in convergence, going from superlinear to quadratic convergence. Moreover, whereas numerical SQP can run into machine precision problems before terminating, the projection-based technique converges easily. Double and single pendulum examples are used to illustrate the technique.
In this paper,a filter-trust-region method is used in nonlinear model predictive control(NMPC) *** means of simultaneous approach based on nonlinear programming,an SQP sub-problem,which treats the iterate step Au as a...
详细信息
In this paper,a filter-trust-region method is used in nonlinear model predictive control(NMPC) *** means of simultaneous approach based on nonlinear programming,an SQP sub-problem,which treats the iterate step Au as an optimal variable,is *** that,a trust region quadraticprogramming approach is used to solve the sub-problem,and the filter method is used to decide whether the trial point is better or not as an approximate solution to the optimization *** the Hessian matrix update method can also keep the sparse structure which is used to reduce the computational *** last, the simulation result proves that the nonlinear predictive control algorithm based on filter-trust-region SQP method can get feasible solution within limited iterations at each time instant.
We consider feature selection and weighting for nearest neighbor classifiers. A technical challenge in this scenario is how to cope with discrete update of nearest neighbors when the feature space metric is changed du...
详细信息
ISBN:
(纸本)9781618395993
We consider feature selection and weighting for nearest neighbor classifiers. A technical challenge in this scenario is how to cope with discrete update of nearest neighbors when the feature space metric is changed during the learning process. This issue, called the target neighbor change, was not properly addressed in the existing feature weighting and metric learning literature. In this paper, we propose a novel feature weighting algorithm that can exactly and efficiently keep track of the correct target neighbors via sequential quadratic programming. To the best of our knowledge, this is the first algorithm that guarantees the consistency between target neighbors and the feature space metric. We further show that the proposed algorithm can be naturally combined with regularization path tracking, allowing computationally efficient selection of the regularization parameter. We demonstrate the effectiveness of the proposed algorithm through experiments.
Low-dose computed tomography (CT) image sequences, obtained to reduce the risk of radiation exposure, can be seriously degraded by quantum noise and other kinds of mechanical and electrical effects. In order to overco...
详细信息
The large volume principle proposed by Vladimir Vapnik, which advocates that hypotheses lying in an equivalence class with a larger volume are more preferable, is a useful alternative to the large margin principle. In...
详细信息
The large volume principle proposed by Vladimir Vapnik, which advocates that hypotheses lying in an equivalence class with a larger volume are more preferable, is a useful alternative to the large margin principle. In this paper, we introduce a new discriminative clustering model based on the large volume principle called maximum volume clustering (MVC), and then propose two approximation schemes to solve this MVC model: A soft-label MVC method using sequential quadratic programming and a hard-label MVC method using semi-definite programming, respectively. The proposed MVC is theoretically advantageous for three reasons. The optimization involved in hard-label MVC is convex, and under mild conditions, the optimization involved in soft-label MVC is akin to a convex one in terms of the resulting clusters. Secondly, the soft-label MVC method possesses a clustering error bound. Thirdly, MVC includes the optimization problems of a spectral clustering, two relaxed k-means clustering and an information-maximization clustering as special limit cases when its regularization parameter goes to infinity. Experiments on several artificial and benchmark data sets demonstrate that the proposed MVC compares favorably with state-of-the-art clustering methods.
sequential quadratic programming is used to solve both a minimum- time and a minimum-noise aircraft landing-approach problem for two dynamic models. The "exact" model features the usual point-mass equations ...
详细信息
sequential quadratic programming is used to solve both a minimum- time and a minimum-noise aircraft landing-approach problem for two dynamic models. The "exact" model features the usual point-mass equations of motion for flight in a vertical plane. The other model is based on two simplifying assumptions: (1) small angle of attack and flight path angle and (2) no flight path angle dynamics (lift equals weight). Range is used to replace time as the independent variable. The resulting models are of order three and two, respectively; each model involves two control functions. The primary objective is to compare the solutions for each model with regard to accuracy and computational effort. Numerical results are presented for a variety of boundary conditions and path constraints.
In this research, a new approach has been suggested for providing an optimization bidding strategy in the day-ahead market for case research in China. This research uses a newly developed version of Emperor Penguin Op...
详细信息
In this research, a new approach has been suggested for providing an optimization bidding strategy in the day-ahead market for case research in China. This research uses a newly developed version of Emperor Penguin Optimizer (CEPO) to govern the fitness function of all individuals based on the Market Clearing Price (MCP) probability function. The resemblance amounts between each day and the next day are used for clustering. The clustering is performed based on the well-known subtractive clustering methodology. A simulation model using the probability function in MCP estimates the fitness function of the generated strategies. The results indicate that this method proposed is a statistically effective bidding design in China’s day-ahead market related to two other plans from the literature.
Loading distribution for heavy plate mill is to find optimal control solutions under the granted performance indicators and constraints including mill capacity and hypothesis of rolling *** solutions are quite differe...
详细信息
Loading distribution for heavy plate mill is to find optimal control solutions under the granted performance indicators and constraints including mill capacity and hypothesis of rolling *** solutions are quite different for different performance *** the article,the performance indicators and sequential quadratic programming(SQP for short below) methods employed in 5 000 mm heavy plate mill of BaoSteel are penetratingly ***,the SQP method is an effective and fast way to solve the nonlinear programming problems with small or medium scale *** in 1976,Han put forward the SQP method for the first time and Powell made it perfect and accomplished the algorithm in *** fact, SQP method was to turn a nonlinear programming problem to a series of sub set of quadraticprogramming *** the algorithm,each iteration step is to solve one quadraticprogramming *** optimal solutions will be gradually approached after quadraticprogramming problems were totally *** solving the quadraticprogramming problem,the active set strategy were employed which turned the constrained quadraticprogramming problem to unconstrained quadraticprogramming *** active set strategy made the whole quadraticprogramming problem be solved by a least square *** finally, the matrix of the least square problem would be decomposed by Q matrix and R *** Q matrix and R matrix were obtained,the optimal solutions would be finally *** loading distribution,the performance indicators were composed by plate shape and draft of each *** shape is represented by rolling force gradually reduced pass by pass with a tunable *** mill capacity is another performance indicator represented by draft of each *** heavy plate mill,the mill capacity here is the motor moment. For heavy draft,the motor would be overloaded especially for the first several passes;for small draft,the motor would be loaded ***
暂无评论