基于动态投入产出控制系统的最优跟踪

研究动态投入产出系统的最优产出跟踪控制问题。根据实际产出需求,采用鲁棒消费策略,修正产出同时使消费过程中的目标函数达到最优。利用离散广义系统的控制的有关理论实现了产出的最优跟踪控制,并给出了具体的优化消费调整策略。算例表明:这种策略实现了对实际产出的渐近跟踪。     

Trajectory optimization for maneuvering satellites

This paper proposes an algorithm to compute optimal trajectories for a maneuvering satellite by using a nonlinear programming representation of an optimal control problem. In this problem, a satellite must be located at a given final position with given velocity from initial position and velocity without passing a prohibited region such as the atmosphere while achieving minimum fuel consumption. Optimal control theory is applied to obtain a set of ordinary differential equations subject to two-point boundary conditions (TPB) on the adjoint system. Then an exact penalty function method is employed to obtain the optimal trajectories by solving the TPB problem as initial conditions for the adjoint system and an unknown final time are regarded as decision variables. This formulation, where the optimal control technique and the nonlinear programming method are incorporated, permits more systematic and flexible algorithm implementation.     

基于网格运动统计算法和最佳缝合线的密集重复结构图像快速拼接方法

针对常用的图像拼接算法对具有密集重复结构的图像会产生大量误匹配点从而出现明显鬼影且耗时较长的问题,将网格运动统计（GMS）算法与最佳缝合线算法相结合,提出了一种密集重复结构的图像快速拼接方法。首先,在图像的重叠区域提取大量粗匹配点;接着,采用GMS算法进行精匹配,然后在此基础上估计变换模型;最后,采用基于动态规划思想的最佳缝合线算法完成图像拼接。实验结果表明,将所提算法应用于两组具有密集重复结构的图像上,不仅可以有效消除鬼影,得到理想的拼接效果,而且显著减少了拼接时间;平均拼接速度分别是传统尺度不变特征变换（SIFT）和加速稳健特征（SURF）算法的7.4倍和3.2倍,分别是结合区域分块的SIFT算法和SURF算法的4.1倍和1.4倍。所提算法能够有效地消除密集重复结构拼接时的鬼影,同时缩短了拼接时间。     

Stabilization Problems of Development of Multisectoral Macroeconomic Systems and the Division of Control between Levels of the Hierarchy

Problems of the synthesis of optimal control are considered in multisectoral macroeconomic systems according to the criterion of the proportional and stable growth of the consumption of all categories of consumers in the production and the nonproduction sphere in the case of the centralized structure of control and a problem is treated of the optimal division of control between the central department and autonomous subsystems in the case of the decentralized structure. An example is given.     

Optimal injection of a satellite into geostationary orbit using a limited number of thruster firings

The problem of optimal injection of a satellite into a geostationary orbit using a limited number of thruster firings is considered. It is formulated as an optimal control problem for a continuous-discrete system with automaton-type discrete part. The application of necessary optimality conditions reduces this problem to a finite-dimensional minimization problem. Taking into account the practical accuracy of executing the thruster firing and shutdown commands, the constrained minimization problem to be solved becomes discrete. Various flight sequences with the number of main engine firings not exceeding nine are investigated. Each sequence is optimized so as to minimize fuel consumption and find the optimal times of the engine firings and shutdowns. Parameters of the transfer orbits and deviations of the final orbit from the geostationary one are calculated. The results are compared with the scheme that is normally used in practice.     

六足步行机节能最佳腿行程的研究

本文研究了步行机腿行程对其能耗的影响,分析了步行机低速和高速行走的最佳腿行程,就具体结构形式的腿得到了其耗能和耗能率计算公式.具体计算了缩放式腿的最佳腿行程,为步行机总体结构设计和行走时的控制提供了理论依据.     

Synthesis of elastic structures with controlled forces

An analytical model is presented for synthesis of elastic structural systems under external loads and desired “controlled forces”. In the first stage, the optimal force distribution and the final cross sectional dimensions are chosen, solving an optimization problem with elastic compatibility temporarily excluded. In the second stage, the force due to the external loads is determined for the optimal structure by elastic analysis, and the desired controlled force needed for the optimal cross sections and force distribution is obtained by subtracting the above result from the optimal force of the first stage. The final design which has been determined in the first stage, satisfies both conditions of equilibrium and compatibility. It is a lower-bound solution for a similar incontrolled elastic system. In some cases the first-stage optimization problem may be cast in linear programming form. Numerical examples of continuous beam and indeterminate truss demonstrate application of the proposed method.     

Sources and uses of knowledge in a dynamic network technology

We estimate a dynamic network technology where new knowledge in the form of publications in STEM (science, technology, engineering, and mathematics) is an intermediate product. Knowledge is produced in the first stage of production and is used in the second stage of production to produce a final output of real consumption, which equals gross domestic product minus investment spending on physical capital minus research and development expenditures. Knowledge also spills over between producers as it becomes disseminated. The two stages of production are linked between periods as investments in research capital and physical capital enhance future production possibilities. Our model combines several theories of production: dynamic data envelopment analysis (DEA) and two‐stage network DEA. Using pooled data on 53 countries during 1999–2012, the model estimates indicate that dynamic efficiency averages about 70%. Countries could increase final consumption by about 25% via greater technical efficiency in production and by another 5% via an optimal intertemporal reallocation of investment spending.     

Optimal dictionaries of the final information on the basis of the solvability criterion and their applications in bioinformatics

Within the algebraic approach to the pattern recognition, the sets of the input (initial) and the output (final) information are defined on the basis of some additional expert information from the relevant subject area. Finding, in a sense, “optimal” definitions for the sets of the output information is crucial, for instance, for solving problems in bioinformatics. In this paper, we propose a formalism to establish optimal dictionaries for output information on the basis of the solvability criterion for the relevant problem. The formalism developed is applied to the problem of recognition of the secondary structure of proteins. The experiments have shown that the optimal dictionary to describe the secondary structure of proteins consists of four five-letter elements. The dictionary allows making a significant increase in the cross-validated recognition accuracy of the secondary structure.     

优化前馈神经网络模型结构的统计分析方法

利用统计分析学的方法对神经网络结构的优化问题进行研究,在神经网络模型输出节点的输出误差具有正态分析的假设下,推导出网络结构信息准则ＮＳＩＣ,并利用主成分分析方法确定了网络的初始结构。仿真结果表明了这一算法的有效性。     

Control of spatial reorientation of a space craft using the free trajectory method

The optimal control problem for spatial reorientation of a space craft for a given time is considered. The functional with the meaning of consumption of the working fluid is minimized. Variants of numerical solution of the formulated problem are presented. The optimal control of a turn of the space craft is determined in the class of controls with the structure “acceleration-free motion-deceleration”. The original optimal control problem is reduced to the solution of three problems, acceleration of the space craft to the required angular momentum, uncontrolled rotation of the space craft, and damping of angular velocity. For increasing accuracy of bringing the space craft to a required angular position, control methods that implement the free trajectory method are proposed. The synthesized controls are invariant with respect to external perturbations and parametric errors. The results of mathematical simulation demonstrating high efficiency of the developed control algorithms are presented. Numerical estimates of consumption of the working fluid for the programmed turn are given.     

Quadratic Optimal Control in Reorienting a Spacecraft in a Fixed Time Period in a Dynamic Problem Statement

A dynamic problem of a spacecraft (SC) turning from an arbitrary initial angular position to the required final angular position is considered and solved. The termination time of the maneuver is known. To optimize the turn control program, the quadratic criterion of quality is used; the minimized functional characterizes the energy consumption. The construction of the optimal control of the turn is based on quaternion variables and Pontryagin’s maximum principle. It is shown that during the optimal turn, the moment of forces is parallel to a straight line immobile in space and the angular momentum direction in the rotation process of the SC is constant relative to the inertial coordinate system. The formalized equations and calculation expressions to determine the optimal turn program are obtained. For a dynamically symmetric SC, a complete solution to the reorientation problem in the closed form is given. An example and results of the mathematical simulation of the dynamics of the SC’s motion under the optimal control are presented, which demonstrate the practical feasibility of the developed method to control the SC’s attitude.     

基于演化博弈的无线传感器网络分簇算法

针对无线传感器网络中节点负载过重与能耗不均衡而出现网络能量空洞的问题，基于演化博弈理论建立一种簇头竞选的博弈模型，同时提出一种基于演化博弈的无线传感器网络最优成簇算法。运用节点的剩余能量、数据接收能耗和数据转发能耗设计簇头演化博弈的收益函数，并将最优发射功率控制机制应用于簇成员的选择，从而形成稳定连通的网络分簇结构。仿真实验表明该算法平衡了节点负载，从而均衡网络能量，有效改善网络中过早出现能量空洞的问题，进而延长了网络生存时间。     

Efficient multiple objective optimal control of dynamic systems with integer controls

In practical optimal control problems both integer control variables and multiple objectives can be present. The current paper proposes a generic and efficient solution strategy for these multiple objective mixed-integer optimal control problems (MO-MIOCPs) based on deterministic approaches. Hereto, alternative scalar multiple objective optimisation techniques as normal boundary intersection and normalised normal constraint are used to convert the original problem into a series of parametric single objective optimisation problems. These single objective mixed-integer optimal control problems are then efficiently solved through direct multiple shooting techniques which exploit convex relaxations of the original problem. Moreover, these relaxations enable to quickly approximate the final solution to any desired accuracy (without the need of solving integer problems). Consequently, the set of Pareto optimal solutions of the MO-MIOCP can be accurately obtained in highly competitive computation times. The proposed method is illustrated on (i) a testdrive case study with a complex car model which includes different gears and conflicting minimum time–minimum fuel consumption objectives, and (ii) a jacketed tubular reactor case study with conflicting conversion, heat recovery and installation costs.     

转型时期中国产业发展动力的SDA分析

自改革开放以来中国经济快速增长,高增长的同时产业结构发生了重大的变动,本文采用结构分解(SDA)方法利用投入产出表对转型期中国产业发展的动力进行研究,发现最终需求的增长是中国产业发展的最重要动力,同时,不同的最终需求明细项目对各产业有不同的拉动作用,正是这种拉动作用的不同使各产业获得了不同的发展速度并最终导致了产业结构的改变。     

基于蜂窝结构的改进混合无线传感器网络覆盖优化算法

基于蜂窝结构的混合无线传感器网络（HWSN）覆盖优化算法HWSNBCS存在移动节点平均移动距离较大的问题,为此,提出一种改进的HWSN覆盖优化算法IHWSNBCS。寻找移动传感器节点初始位置与通过HWSNBCS算法得出的候选目标位置之间的最优匹配,将移动节点移动距离之和最小化问题转化为二分图最优匹配问题,利用带权二分图匹配算法KM寻找该匹配问题的最优解,从而得到移动节点最终的目标位置,并实现对HWSNBCS算法移动节点平均移动距离的进一步优化。实验结果表明,IHWSNBCS算法在取得与HWSNBCS算法相同网络覆盖率的前提下,移动节点的平均移动距离减少幅度达到38.87%～43.28%,单个移动节点的最大移动距离减少幅度达到22.65%～66.58%,降低了系统因重新部署移动传感器节点所产生的能耗以及单个传感器节点因能量耗尽而失效的概率,从而延长了网络生命周期,同时,IHWSNBCS的ΔCov-Dist性能指标为HWSNBCS算法的1.64～1.76倍,表明移动节点移动相同距离时IHWSNBCS算法的网络覆盖率提升更大。     

基于能耗的分簇频谱感知方法

在协作频谱感知中,当信道经历衰落和阴影时,各用户的感知性能会有不同程度影响,某些受影响严重的次用户甚至可能会严重影响到整个感知网络的性能.同时参与协作的用户数越多,认知传感器网络面临的问题越大.为了解决这些问题,提出了一种基于能耗的分簇协作感知算法.该算法引入基于节点能量的LEACH分簇算法,将认知传感器网络中的各个节点按该分簇算法分入若干簇内;在每个簇内,构造了同时考虑检测性能和能量消耗的效用函数,基于该效用函数对最佳协作用户数和用户进行选择,使每个簇内的检测性能和能耗达到很好的折中;在汇聚节点运用OR规则将簇头发送过来的感知信息进行融合,得到最终判决结果.仿真结果表明,该基于能耗的分簇协作频谱感知算法可以有效地延长认知传感器网络的生存时间,节省网络的能耗,并且保证良好的检测性能.     

氧化铝配矿调度方法研究及软件实现

为了解决氧化铝配矿调度过程中不确定性因素多、计算工作量大、难以获取最优方案以及生产结构不合理造成能源浪费等问题,以能耗最小为目标函数,结合生产经验,建立了烧结法和拜耳法统一配矿的优化调度模型,并针对模型中存在的非线性、多约束和多目标特点,运用了一种引入变异算于的混合粒子群算法进行寻优.在充分考虑了软件工业实用性及界面友...     

基于改进蚁群算法的无线传感网络路由算法研究

针对无线传感网络能量消耗不均及节点过早死亡等问题,提出一种新的基于改进蚁群算法的路由算法。在网络结构方面,加入网络分隔带和搜索角,并结合节点剩余能量,共同限制下一跳节点的转移概率;同时改进启发函数,加入能量影响因子,增强算法寻优,避免陷入局部最优;在信息素更新方面,引入阈值机制并设立最优路径权重值来寻找最优路径。仿真结果表明,改进后的算法能够进一步降低网络能耗,延长网络生命周期。