This work aims to improve the reliability of dynamic systems by eliminating effect random control variables. At first, reliability-based optimization problem (RB-DOP) is introduced and defined account for with uncertainty associated Whereafter, in order solve RB-DOP efficiently, constraint function response shift scalar (CFRSS)-based method proposed, which nested decoupled into an equivalent deterministic DOP a CFRSS search problem, two problems are addressed iteratively until law converges....

In this article, an adaptive metamodel-based global optimization (AMGO) algorithm is presented to solve unconstrained black-box problems. the AMGO algorithm, a type of hybrid model composed kriging and augmented radial basis function (RBF) used as surrogate model. The weight factors are adaptively selected in process. To balance local search, sub-optimization problem constructed during each iteration determine new iterative points. As numerical experiments, six standard two-dimensional test...

Abstract High-dimensional model representation (HDMR), decomposing the high-dimensional problem into summands of different order component terms, has been widely researched to work out dilemma “curse-of-dimensionality” when using surrogate techniques approximate problems in engineering design. However, available one-metamodel-based HDMRs usually encounter predicament prediction uncertainty, while current multi-metamodels-based cannot provide simple explicit expressions for black-box...

Aiming to address the issue of 'curse dimensionality' encountered in approximating high-dimensional problems using surrogate-based methods, model representation (HDMR), decomposing problem into summands different-order component functions, has been widely studied. To reduce computational demands current HDMR metamodelling techniques, an improved framework (iHDMR) is presented, taking full advantage relationships between first-order and second-order functions Cut-HDMR theory. then, a novel...

Monitoring the state of battery, including charge (SOC) and health (SOH), is crucial for ensuring safety reliability electrical equipment. The paper presents a novel hybrid network that combines nonlinear autoregressive model with exogenous inputs (NARX) DS-attention. proposed DS-attention method establishes robust mapping relationship between outputs, it specialized recurrent neural enhances estimation performance by incorporating division function self-adaptive into attention mechanism....

Abstract When solving the black-box dynamic optimization problem (BDOP) in sophisticated system, finite difference technique requires significant computational efforts on numerous expensive system simulations to provide approximate numerical Jacobian information for gradient-based optimizer. To save budget, this work introduces a BDOP framework based right-hand side (RHS) function surrogate model (RHSFSM), which RHS derivative functions of state equation are approximated by models, and is...

The Kriging surrogate model in complex simulation problems uses as few expensive objectives possible to establish a global or local approximate interpolation. However, due the inversion of covariance correlation matrix and solving Kriging-related parameters, approximation process for high-dimensional is time consuming even impossible construct. For this reason, modeling method through principal component dimension reduction (HDKM-PCDR) proposed by considering parameters design variables...

In order to overcome the drawbacks of expensive function evaluation in practical reliability-based design optimization (RBDO) problem, researchers have proposed black box-based RBDO method. The algorithm flow commonly employed method for box problem consists outer construction loop surrogate model constraint and inner model-based solving loop. To improve ability this paper proposes a transformation-based improved kriging increase effectiveness two loops identified above. For loop, sample...

When solving the control co-design (CCD) problem using simultaneous strategy in a deterministic manner, uncertainty stemming from stochastic design variables is ignored, and might have negative influence on performance of dynamic system. In attempting to overcome undesirable effect uncertainty, this research investigates reliability-based (RB-CCD) presents single-loop framework for RB-CCD based modified model approach (SLA). Specifically, deduced by introducing additional equality...

The kriging optimization method that can only obtain one sampling point per cycle has encountered a bottleneck in practical engineering applications. How to find suitable generate multiple points at time while improving the accuracy of convergence and reducing number expensive evaluations been wide concern. For this reason, kriging-assisted multi-objective constrained global (KMCGO) proposed. sample data obtained from function evaluation is first used construct or update model each cycle....

When encountering the black-box dynamic co-design and optimization (BDCDO) problem in multidisciplinary system, finite difference technique is inefficient or even infeasible to provide approximate numerical gradient information for algorithm since it requires numerous original expensive evaluations. Therefore, a solving framework based on surrogate model of state equation introduced optimize BDCDO. To efficiently construct model, sequential sampling method presented basis successive relative...

The generalized efficient global optimization (GEGO) method is able to solve expensive black-box problems. However, selecting one sampling point per cycle may result in large time consumption and loss of convergence accuracy. To this end, a kriging-based multi-point unconstrained (KMUGO) proposed. It extends the GEGO with improved constant liar (CL) strategy. For each cycle, kriging model first constructed or updated by existing sampled data. Then, enhanced alternative CL strategy used find...