On the relational representation of complex specialization structures

The mapping of entity-relationship schemas (ER schemas) that contain complex specialization structures into the relational model requires the use of specific strategies to avoid inconsistent states in the final relational database. In this paper, we show that for this mapping to be correct it is required to enforce a special kind of integrity constraint, the key pairing constraint (KPC). We present a mapping strategy that use simple inclusion dependencies to enforce KPC and show that this strategy can be used to correctly map specialization structures that are more general than the simple specialization structures considered by previous strategies.     

A prediction-based adaptive grouping differential evolution algorithm for constrained numerical optimization

In this paper, a new adaptive grouping differential evolution (AGDE) algorithm is proposed to improve the optimization performance by implementing a prediction strategy of the constraints for constrained optimization problems. It is unnecessary to calculate the constraint values when dealing with the constraints in this method. The constraints are handled after a simple prediction according to the Lipschitz condition. When the constraints are very complex, the load arisen from the calculation of the constraint values is reduced dramatically and the feasibility of the solutions remains with great probability. In AGDE algorithm, the population is dynamically grouped to three subpopulations with respective newly-designed mutation strategy. Meanwhile, the mutation factor and crossover probability are adopted associated with the evolutionary process according to the information of the entire population. Both of the above improvements not only increase the diversity of population and speed up the convergence, but also reduce the complexity of the parameter selection. Four sets of comparative experiments are carried out to evaluate the feasibility and effectiveness of the proposed method that deals with the constraints.     

A numerical method for annuity-purchasing decision making to minimize the probability of financial ruin for regime-switching wealth models

This work develops an approximation procedure to find optimal annuity-purchasing strategies for minimizing the probability of lifetime ruin. The wealth is modelled as a regime-switching diffusion modulated by a continuous-time Markov chain. Based on Markov chain approximation techniques, a discrete-time controlled Markov chain with two components is constructed. Under simple conditions, the convergence of the approximation sequence to the wealth process is obtained. The convergence of the approximation to the value function is also established. Several examples are provided to demonstrate the performance of the algorithms.     

多目标约束处理方法求解泊位岸桥联合分配问题

为了制定合理高效的泊位岸桥联合分配方案,加快船舶周转,本文针对船舶动态到港的连续泊位建立了以船舶总在港时间最短为目标的泊位岸桥联合分配混合整数非线性模型.通过多目标约束处理策略将复杂约束的违反程度转化为另一个目标,从而将原单目标优化模型转化为双目标优化模型,并用基于快速非支配排序的多目标遗传算法(NSGA-II)对其进行求解.同时,针对问题特点,分别设计了基于调整、惩罚函数、可行解优先和综合约束处理策略的单目标遗传算法对原模型进行求解.通过多组不同规模的标准算例对本文的方法进行测试,验证了基于多目标约束处理策略的方法求解效果相较于单目标约束处理策略的方法更加高效和稳定.     

A Constrained Line Search Optimization Method for Discriminative Training of HMMs

In this paper, we propose a novel optimization algorithm called constrained line search (CLS) for discriminative training (DT) of Gaussian mixture continuous density hidden Markov model (CDHMM) in speech recognition. The CLS method is formulated under a general framework for optimizing any discriminative objective functions including maximum mutual information (MMI), minimum classification error (MCE), minimum phone error (MPE)/minimum word error (MWE), etc. In this method, discriminative training of HMM is first cast as a constrained optimization problem, where Kullback-Leibler divergence (KLD) between models is explicitly imposed as a constraint during optimization. Based upon the idea of line search, we show that a simple formula of HMM parameters can be found by constraining the KLD between HMM of two successive iterations in an quadratic form. The proposed CLS method can be applied to optimize all model parameters in Gaussian mixture CDHMMs, including means, covariances, and mixture weights. We have investigated the proposed CLS approach on several benchmark speech recognition databases, including TIDIGITS, Resource Management (RM), and Switchboard. Experimental results show that the new CLS optimization method consistently outperforms the conventional EBW method in both recognition performance and convergence behavior.     

工业大系统双层结构预测控制的集中优化与分散控制策略

为降低工业大系统模型预测控制(Model predictive control,MPC)在线计算复杂度,同时保证系统的全局优化性能,提出一种集中优化、分散控制的双层结构预测控制策略.在稳态目标计算层(Steady-state target calculation, SSTC),基于全局过程模型对系统进行集中优化,将优化结果作为设定值传递给动态控制层;在动态控制层,将大系统划分为若干个子系统,每个子系统分别由基于各自子过程模型的模型预测控制进行控制,为减少各子系统之间的相互干扰,在各个子系统之间添加前馈控制器对扰动进行补偿,提高系统的总体动态控制性能.该策略的优点在于能确保系统全局最优性的同时降低了在线计算量,提高了工业大系统双层结构预测控制方法的实时性.仿真实例验证该方法的有效性.     

On multi-objective economic predictive control for cyclic process operation

In recent literature, a utopia-tracking strategy has been proposed for multi-objective model predictive control. This strategy tracks a vector of independently minimized objectives, evaluated at steady-state operation (the utopia point). The approach in the present work is based on the premise that cyclic process operation may in some cases outperform, on average, steady-state operation. We therefore concern ourselves with optimizing average performance for one cyclic period of operation. A dynamic utopia-tracking strategy is proposed, which generalizes steady-state utopia-tracking to systems which are optimally operated cyclically. The proposed control law minimizes the distance of its cost vector to a vector of independently minimized objectives, evaluating average cyclic performance (the dynamic utopia point). Recursive feasibility is established for a cyclic terminal state constraint formulation, however, conditions for stability are not given. The theoretical concepts are illustrated for a simple acetylene hydrogenation case, and a more complex oil production optimization case. The simulation study indicates novel operational insight for the oil production optimization case, by illustrating how simultaneous optimization of two objectives results in cyclic operation with improved performance.     

Window annealing for pixel-labeling problems

The pixel labeling problems in computer vision are often formulated as energy minimization tasks. Algorithms such as graph cuts and belief propagation are prominent; however, they are only applicable for specific energy forms. For general optimization, Markov Chain Monte Carlo (MCMC) based simulated annealing can estimate the minima states very slowly.This paper presents a sampling paradigm for faster optimization. First, in contrast to previous MCMCs, the role of detailed balance constraint is eliminated. The reversible Markov chain jumps are essential for sampling an arbitrary posterior distribution, but they are not essential for optimization tasks. This allows a computationally simple window cluster sample. Second, the proposal states are generated from combined sets of local minima which achieve a substantial increase in speed compared to uniformly labeled cluster proposals. Third, under the coarse-to-fine strategy, the maximum window size variable is incorporated along with the temperature variable during simulated annealing. The proposed window annealing is experimentally shown to be many times faster and capable of finding lower energy compared to the previous Gibbs and Swendsen-Wang cut (SW-cut) sampler. In addition, the proposed method is compared with other deterministic algorithms like graph cut, belief propagation, and spectral method in their own specific energy forms. Window annealing displays competitive performance in all domains.     

Regularization of shape optimization problems using FE-based parametrization

This paper introduces a general fully stabilized mesh based shape optimization strategy, which allows for shape optimization of mechanical problems on FE-based parametrization. The well-known mesh dependent results are avoided by application of filter methods and mesh regularization strategies. Filter methods are successfully applied to SIMP (Solid Isotropic Material with Penalization) based topology optimization for many years. The filter method presented here uses a specific formulation that is based on convolution integrals. It is shown that the filter methods ensure mesh independency of the optimal designs. Furthermore they provide an easy and robust tool to explore the whole design space with respect to optimal designs with similar mechanical properties. A successful application of optimization strategies with FE-based parametrization requires the combination of filter methods with mesh regularization strategies. The latter ones ensure reliable results of the finite element solutions that are crucial for the sensitivity analysis. This presentation introduces a new mesh regularization strategy that is based on the Updated Reference Strategy (URS). It is shown that the methods formulated on this mechanical basis result in fast and robust mesh regularization methods. The resulting grids show a minimum mesh distortion even for large movements of the mesh boundary. The performance of the proposed regularization methods is demonstrated by several illustrative examples.     

Optimal Control of Finite‐Valued Networks

Control of finite‐valued networks, including Boolean networks, is currently a hot topic. In this paper the optimization of the networks with time‐discounted performance criterion is discussed. The problem is formulated as a finite strategy game between human and machine. It is first proved that the optimal strategy can be found in the set of essentially periodic strategies, which makes the problem finitely computable, though the computational complexity of exhaustion might be a severe problem. Then an efficient numerical method is develossped to solve the problem. Some interesting examples are presented to demonstrate the efficiency of our results.     

Control variable parameterisation with penalty approach for hypersonic vehicle reentry optimisation

An efficient trajectory optimisation approach combining the classical control variable parameterisation (CVP) with a novel smooth technology and two penalty strategies is developed to solve the trajectory optimal control problems. Since it is difficult to deal with path constraints in CVP method, the novel smooth technology is firstly employed to transform the complex constraints into one smooth constraint. Then, two penalty strategies are proposed to tackle the converted path and terminal constraints to decrease the computational complexity and improve the constraints satisfaction. Finally, a nonlinear programming problem, which approximates the original trajectory optimisation problem, is obtained. Error analysis shows that the proposed method has good convergence property. A general hypersonic cruise vehicle trajectory optimisation example is employed to test the performance of the proposed method. Numerical results show that the path and terminal conditions are well satisfied and better trajectory profiles are obtained, showing the effectiveness of the proposed method.     

基于广义改进分解策略的多目标代理优化方法

为解决多目标代理优化方法中代理模型选择单一问题,提出基于广义改进函数分解策略的多目标代理优化方法.该方法充分利用模型预测信息构建广义改进多目标分解准则和广义改进R2指标准则,有效拓展多目标代理优化中代理模型的选择空间.所提两种准则通过随机均匀权重实现全局探索和局部搜索能力的自适应平衡.研究结果表明,所提方法在有限仿真条件下拥有良好的寻优性能,获得Pareto前沿在收敛性、多样性及空间分布性方面均具有一定优势.相比同类方法,该方法具有优势:1)不需要模型预测不确定性信息,适用于基于不同种类代理模型的代理优化方法; 2)实现简单且计算复杂度低,能够有效提升昂贵黑箱问题优化效率.     

A robust mean absolute deviation model for portfolio optimization

In this paper we develop a robust model for portfolio optimization. The purpose is to consider parameter uncertainty by controlling the impact of estimation errors on the portfolio strategy performance. We construct a simple robust mean absolute deviation (RMAD) model which leads to a linear program and reduces computational complexity of existing robust portfolio optimization methods. This paper tests the robust strategies on real market data and discusses performance of the robust optimization model empirically based on financial elasticity, standard deviation, and market condition such as growth, steady state, and decline in trend. Our study shows that the proposed robust optimization generally outperforms a nominal mean absolute deviation model. We also suggest precautions against use of robust optimization under certain circumstances.     

Finite optimal control for time-bounded reachability in CTMDPs and continuous-time Markov games

We establish the existence of optimal scheduling strategies for time-bounded reachability in continuous-time Markov decision processes, and of co-optimal strategies for continuous-time Markov games. Furthermore, we show that optimal control does not only exist, but has a surprisingly simple structure: the optimal schedulers from our proofs are deterministic and timed positional, and the bounded time can be divided into a finite number of intervals, in which the optimal strategies are positional. That is, we demonstrate the existence of finite optimal control. Finally, we show that these pleasant properties of Markov decision processes extend to the more general class of continuous-time Markov games, and that both early and late schedulers show this behaviour.     

面向复杂约束优化问题的进化算法综述

约束优化是多数实际工程应用优化问题的呈现方式.进化算法由于其高效的表现,近年来被广泛应用于约束优化问题求解.但约束条件使得问题解空间离散、缩小、改变,给进化算法求解约束优化问题带来极大挑战.在此背景下,融合约束处理技术的进化算法成为研究热点.此外,随着研究的深入,近年来约束处理技术在复杂工程应用问题优化中得到了广泛发展,例如多目标、高维、等式优化等.根据复杂性的缘由,将面向复杂约束优化问题的进化优化分为面向复杂目标的进化约束优化算法和面向复杂约束场景的进化算法两种类别进行综述,其中,重点探讨了实际工程应用的复杂性对约束处理技术的挑战和目前研究的最新进展,并最后总结了未来的研究趋势与挑战.     

Optimizing the Reliability of Streaming Applications Under Throughput Constraints

Mapping a pipelined application onto a distributed and parallel platform is a challenging problem. The problem becomes even more difficult when multiple optimization criteria are involved, and when the target resources are heterogeneous (processors and communication links) and subject to failures. This paper investigates the problem of mapping pipelined applications, consisting of a linear chain of stages executed in a pipeline way, onto such platforms. The objective is to optimize the reliability under a performance constraint, i.e., while guaranteeing a threshold throughput. In order to increase reliability, we replicate the execution of stages on multiple processors. We compare interval mappings, where the application is partitioned into intervals of consecutive stages, with general mappings, where stages may be partitioned without any constraint, thereby allowing a better usage of processors and communication network capabilities. However, the price to pay for general mappings is a dramatic increase in the problem complexity. We show that computing the period of a given general mapping is an NP-complete problem, and we give polynomial bounds to determine a (conservative) approximated value. On the contrary, the period of an interval mapping obeys a simple formula, and we provide an optimal dynamic programming algorithm for the bi-criteria interval mapping problem on homogeneous platforms. On the more practical side, we design a set of efficient heuristics, and we compare the performance of interval and general mapping strategies through extensive simulations.     

Two-level multipoint constraint approximation concept for structural optimization

A two-level multipoint approximation concept is proposed. Based upon the values and the first-order derivatives of the critical constraint functions at the points obtained in the procedure of optimization, explicit functions approximating the primal constraint functions have been created. The nonlinearities of the approximate functions are controlled to be near those of the constraint functions in their expansion domains. Based on the principle above, the first-level sequence of explicitly approximate problems used to solve the primarily structural optimization problem are constructed. Each of them is approximated again by the second-level sequence of approximate problems, which are formed by using the linear Taylor series expansion and then solved efficiently with dual theory. Typical numerical examples including optimum design for trusses and frames are solved to illustrate the power of the present method. The computational results show that the method is very efficient and no intermediate/generalized design variable is required to be selected. It testifies to the adaptability and generality of the method for complex problems.     

Application of non-probabilistic sensitivity analysis in the optimization of aeronautical hydraulic pipelines

This paper investigates the reliability design optimization of an aeronautical hydraulic pipeline system, in which the constraint locations are treated as design parameters. To reduce the size of the optimization problem, two non-probabilistic global sensitivity indices are introduced and modified to screen out those constraint locations which have no or little effect on the optimization target. Considering the rest of constraint locations as design variables, the complexity of the optimization problem is dramatically reduced. The optimization of the pipeline systems demonstrates that the proposed method is superior to the traditional direct optimization method in both the optimization efficiency and results. This work indicates that introduction of sensitivity analysis can greatly improve the efficiency and performance of optimization, especially in those complex engineering problems involving a large number of design variables.     

An efficient RBDO process using adaptive initial point updating method based on sigmoid function

Using Kriging model in the reliability-based design optimization (RBDO) process can reduce the computational cost effectively. However, the constraints in practical problems are often highly nonlinear and black box functions, and the cost of evaluations at design points is very high, such as the finite element analysis (FEA). So building accurate Kriging models will consume a huge amount of computing resources. Moreover, complex constraint functions will lead to the local minimum in the design space, which makes it difficult to get the global optimum. To cope with this problem, an adaptive sampling method based RBDO process (AS-RBDO) is proposed by introducing two new sampling criterions. The first criterion is built based on the support vector machine (SVM) and the sigmoid function. And the second criterion is built based on the improvement of the constraint boundary sampling (CBS) method. With the use of new strategies, AS-RBDO can not only guide the optimization to the global optimal direction, but also update the Kriging model only in the local range that has the greatest impact on the results of RBDO. Thus the unnecessary sampling and evaluations can be avoided effectively. Several examples are selected to test the computation capability of the proposed method. The results show that AS-RBDO can effectively improve the efficiency of the RBDO process.