Sequential Optimization Research Articles

Stable Linear System Identification With Prior Knowledge by Riemannian Sequential Quadratic Optimization

Stable Linear System Identification With Prior Knowledge by Riemannian Sequential Quadratic Optimization

Treatment Algorithm for the Resorption of Calcific Tendinitis Using Extracorporeal Shockwave Therapy: A Data Mining Study.

Although evidence indicates that extracorporeal shockwave therapy (ESWT) is effective in treating calcifying shoulder tendinitis, incomplete resorption and dissatisfactory results are still reported in many cases. Data mining techniques have been applied in health care in the past decade to predict outcomes of disease and treatment. To identify the ideal data mining technique for the prediction of ESWT-induced shoulder calcification resorption and the most accurate algorithm for use in the clinical setting. Case-control study. Patients with painful calcified shoulder tendinitis treated by ESWT were enrolled. Seven clinical factors related to shoulder calcification were adopted as the input attributes: sex, age, side affected, symptom duration, pretreatment Constant-Murley score, and calcification size and type. The 5 data mining techniques assessed were multilayer perceptron (neural network), naïve Bayes, sequential minimal optimization, logistic regression, and the J48 decision tree classifier. A total of 248 patients with calcified shoulder tendinitis were enrolled in this study. Shorter symptom duration yielded the highest gain ratio (0.374), followed by smaller calcification size (0.336) and calcification type (0.253). With the J48 decision tree method, the accuracy of 3 input attributes was 89.5% by 10-fold cross-validation, indicating satisfactory accuracy. A treatment algorithm using the J48 decision tree indicated that a symptom duration of ≤10 months was the most positive indicator of calcification resorption, followed by a calcification size of ≤10.82 mm. The J48 decision tree method demonstrated the highest precision and accuracy in the prediction of shoulder calcification resorption by ESWT. A symptom duration of ≤10 months or calcification size of ≤10.82 mm represented the clinical scenarios most likely to show resorption after ESWT.

Enhancing clay content estimation through hybrid CatBoost-GP with model class selection

Excessive soil swelling, triggered by variations in moisture content, is recognized as one of the most severe geo-hazards. Swelling Potential (SP) is closely correlated with Clay Content (CC), Plasticity Index (PI), and Activity (A = PI/CC), which are crucial input features for SP prediction models. Traditional methods for determining CC, such as the hydrometer and pipette methods, are labor-intensive and time-consuming. Employing easily accessible soil properties to predict CC directly not only enhances the modeling efficiency of SP prediction models but also mitigates the impact of laboratory test errors on model performance. This study leveraged two databases encompassing a broad spectrum of soil classifications and developed three ensemble learning models (RF, XGBoost, and CatBoost) to predict CC efficiently. Sequential model-based optimization algorithms (SMBO) were employed to effectively identify the optimal hyperparameters for these models. To further refine the model, the Shapley additive explanation (SHAP) method was utilized to ascertain the most influential soil properties. The findings revealed that the CatBoost model, optimized via GP, achieved the highest level of predictive accuracy. Three easily measurable soil properties — the percentage of soil particles passing through 0.075 mm and 0.425 mm sieves, along with the plasticity index — were identified as key predictors for CC. The estimated CC values were instrumental in constructing various empirical and machine learning models to predict SP, all of which exhibited exceptional performance. This research advances the modeling efficiency of SP prediction models, making it particularly relevant for rapid-response scenarios in transportation infrastructure projects.

A Collaboratively Iterative Sequential Approximate Optimization Method for Aerodynamic Optimization

A Collaboratively Iterative Sequential Approximate Optimization Method for Aerodynamic Optimization

Performance analysis of data resampling on class imbalance and classification techniques on multi-omics data for cancer classification.

Cancer, in any of its forms, remains a significant public health concern worldwide. Advances in early detection and treatment could lead to a decline in the overall death rate from cancer in recent decades. Therefore, tumor prediction and classification play an important role in fighting cancer. This study built computational models for a joint analysis of RNA seq, copy number variation (CNV), and DNA methylation to classify normal and tumor samples across liver cancer, breast cancer, and colon adenocarcinoma from The Cancer Genome Atlas (TCGA) dataset. Total of 18 machine learning methods were evaluated based on the AUC, precision, recall, and F-measure. Besides, five techniques were compared to ameliorate problems of class imbalance in the cancer datasets. Synthetic Minority Oversampling Technique (SMOTE) demonstrated the best performance. The results indicate that the model applying Stochastic Gradient Descent (SGD) for learning binary class SVM with hinge loss has the highest classification results on liver cancer and breast cancer datasets, with accuracy over 99% and AUC greater than or equal to 0.999. For colon adenocarcinoma dataset, both SGD and Sequential Minimal Optimization (SMO) that implements John Platt's sequential minimal optimization algorithm for training a support vector machine shows an outstanding classification performance with accuracy of 100%, AUC, precision, recall, and F-measure all at 1.000.

Two-level optimization strategy for vehicle speed and battery thermal management in connected and automated EVs

The performance of the battery is affected by temperature, and the battery thermal management (BTM) system consumes considerable energy to maintain the temperature in the suitable range. The unnecessary acceleration and deceleration of electric vehicles (EVs) during driving causes higher energy consumption in the powertrain. The emergence of connected and automated vehicle (CAV) technology provides an opportunity for predictive control of thermal and energy management. To explore the coordination optimization between battery thermal and vehicle energy management, this article proposes a two-level optimization framework for the speed and BTM of EVs, which improves energy efficiency and battery safety. Each level consists of a sequential optimization of speed and battery thermal. In the upper layer, speed planning based on iterative dynamic programming (IDP) is first proposed to reduce powertrain energy consumption using intelligent traffic information. Then, based on the BTM system and the battery thermodynamics features, the long-term optimal trajectory of the battery temperature is derived according to optimized speed. In the lower layer, the model predictive controllers (MPC) are designed to track reference speed and temperature trajectories in real-time and enforce energy saving. Meanwhile, to improve the prediction accuracy of the system model, we integrate the Gaussian process (GP) model in the MPC and build the learning-based MPC strategy. Simulation results verify the performance of the proposed method which reduces the powertrain energy consumption by 20.95%. In the high and low temperature environment, compared with normal MPC, PID-based and Rule-based, it reduces BTM energy consumption by up to 15.69%, 29.68% and 38.73%.

RBF-Based Integrated Optimization Method of Structural and Turning Parameters for Low-Floor Axle Bridge

The axle bridge plays a crucial role in the bogie of low-floor light rail vehicles, impacting operational efficiency and fuel economy. To minimize the total cost of the structure and turning of axle bridges, an optimization model of structural and turning parameters was built, with the fatigue life, maximum stress, maximum deformation, and maximum main cutting force as constraints. Through orthogonal experiments and multivariate variance analysis, the key design variables which have a significant impact on optimization objectives and constraints (performance responses) were identified. Then the optimal Latin hypercube design and finite element simulation was used to build a Radial Basis Function (RBF) model to approximate the implicit relationship between design variables and performance responses. Finally, a multi-island genetic algorithm was applied to solve the integrated optimization model, resulting in an 8.457% and 1.1% reduction in total cost compared with the original parameters and parameters of sequential optimization, proving the effectiveness of the proposed method.

Two-stage recoverable robust optimization for an integrated location–allocation and evacuation planning problem

We consider an integrated location–allocation and evacuation planning problem in a disaster context, where the effects of a disaster, including the uncertain capacities of relief facilities (rescue centers and distribution centers), uncertain demands for relief supplies and casualty treatment services, and uncertain availability of transportation links are characterized by a discrete scenario set. Instead of complete failures, we allow the disrupted relief facilities only lose part of their capacity. To deal with the uncertainties, we propose a two-stage recoverable robust optimization model, where the location decision of relief facilities, the allocation decision of delivering relief supplies from relief facilities to affected areas, the transfer decision of transporting casualties from affected areas to rescue centers etc are defined in two stages where the first-stage solution should be robust against the possible effects of a disaster that are revealed in the second stage, and the second-stage solution involves some recovery actions, which we term as multi-mitigation strategies: re-opening and re-operation, re-allocation, and relief supply sharing, to mitigate the effects. To solve the model to optimality, we develop a nested two-stage decomposition algorithm that iterates between a master problem considering only a subset of disaster scenarios solved by a Benders decomposition algorithm that incorporates some non-trivial acceleration strategies, and an adversarial separation problem that identifies disaster scenarios to enhance the worst-case recovery cost of the master problem. We introduce some warm-start techniques to accelerate the convergence of the solution algorithm. We conduct numerical studies on simulation instances to assess the performance of the solution algorithm, and analyze the robustness and recoverability of the model. We also conduct extensive numerical studies on realistic instances from Ya’an and Ganzi to demonstrate the benefits of accounting for recoverable robustness over a stochastic policy and a robust policy without recovery actions, the benefits of considering integrated optimization over sequential optimization, and the benefits of considering partial capacity loss and multi-mitigation strategies.

Sequential simulated annealing for life-cycle optimization of nonlinear stochastic systems via arbitrary polynomial chaos expansion

Quantifying the uncertainties of engineering systems modeled as nonlinear oscillators subject to random excitation is a theoretically complex and computationally demanding task. Consequently, finding an optimal design of such systems considering their life-cycle performance is prohibitive with Monte Carlo simulation methods. In this paper, an efficient performance-based design optimization approach is developed for finding the optimal parameters of engineering systems modeled as nonlinear/hysteretic oscillators subject to stationary and non-stationary excitation. This novel approach utilizes an arbitrary polynomial chaos expansion to estimate the expected cost of failure without performing a computationally expensive integration over the hazard levels. Moreover, we introduce a novel sequential heuristic optimization scheme based on simulated annealing to minimize the total expected cost over the structure life-cycle. Three examples are included in the paper to assess the developed optimization scheme. First, we use the developed framework to optimize a linear single-degree-of-freedom oscillator subject to broadband excitation. Second, a multi-degree-of-freedom oscillator with cubic nonlinearity in damping and stiffness, subject to stationary broadband excitation, is optimized to show the influence of the problem dimensionality in the optimization process. In the last example, a multi-story reinforced concrete shear building modeled as a multi-degree-of-freedom Bouc-Wen oscillator with stiffness and strength degradation and subject to multi-hazards modeled as stationary (wind excitation) and non-stationary (earthquake) stochastic processes, is optimized.

Evaluation and control of energy‐efficient processes for separating epichlorohydrin‐oriented quaternary system

AbstractBACKGROUNDThe direct epoxidation of allyl chloride is a clean and efficient synthesis method for producing epichlorohydrin (ECH), which is a crucial organic intermediate widely used in the fields of renewable energy and aerospace. However, the presence of multiple azeotropes in the synthesized product using this method complicates the separation process and results in high energy consumption. To efficiently separate high‐purity ECH, this paper investigates and analyzes three separation schemes based on phase‐equilibrium analysis: hybrid extractive distillation (HED), pressure swing distillation (PSD) and three‐column batch distillation (TCBD).RESULTSThe operating parameters of the three separation processes are optimized by the sequential iterative optimization method such that the minimum total annual cost can reach $493 491 yr−1. The thermal integration method is used for process energy‐saving optimization, and the total annual cost can be further reduced by 6.9%. In addition, a comprehensive evaluation based on economic, energy, environmental and exergy analysis is conducted, and reveals that the TCBD process with thermal integration is optimal. A control structure is designed for the TCBD process to enhance its robustness such that the purity of ECH remains above 99.9 mol% under ±10% disturbances in feed flowrate and composition.CONCLUSIONSCompared with the HED and PSD processes, the TCBD process has better economic and environmental benefits, and its control structure can effectively resist disturbances. It is reasonable to believe that the TCBD process can be an excellent solution for the industrial production of ECH. © 2024 Society of Chemical Industry (SCI).

First-Order Algorithms Without Lipschitz Gradient: A Sequential Local Optimization Approach

Most first-order methods rely on the global Lipschitz continuity of the objective gradient, which fails to hold in many problems. This paper develops a sequential local optimization (SLO) framework for first-order algorithms to optimize problems without Lipschitz gradient. Operating on the assumption that the gradient is locally Lipschitz continuous over any compact set, SLO develops a careful scheme to control the distance between successive iterates. The proposed framework can easily adapt to the existing first-order methods, such as projected gradient descent (PGD), truncated gradient descent (TGD), and a parameter-free variant of Armijo linesearch. We show that SLO requires [Formula: see text] gradient evaluations to find an ϵ-stationary point, where Y is certain compact set with [Formula: see text] radius, and [Formula: see text] denotes the Lipschitz constant of the i-th order derivatives in Y. It is worth noting that our analysis provides the first nonasymptotic convergence rate for the (slight variant of) Armijo linesearch algorithm without globally Lipschitz continuous gradient or convexity. As a generic framework, we also show that SLO can incorporate more complicated subroutines, such as a variant of the accelerated gradient descent (AGD) method that can harness the problem’s second-order smoothness without Hessian computation, which achieves an improved [Formula: see text] complexity. Funding: J. Zhang is supported by the MOE AcRF [Grant A-0009530-04-00], from Singapore Ministry of Education. M. Hong is supported by NSF [Grants CIF-1910385 and EPCN-2311007]. Supplemental Material: The online appendix is available at https://doi.org/10.1287/ijoo.2021.0029 .

Uncertainty‐oriented thermoelastic topology optimization with stress constraint

AbstractThe material redistribution abilities of traditional deterministic topology optimization are effective in addressing stress‐related design issues in thermal elastic structures. However, uncertainties are inevitable in real‐world. The structural strength of a design, achieved through deterministic topology optimization, is highly susceptible to these uncertainties, which may result in failure. This article introduces a novel method for topology optimization in stress‐constrained thermoelastic structures, taking into account the uncertainties associated with heat sources and loads. We employ the Kieisselmeier–Steinhauser function to aggregate the stress constraint when constructing the performance function. In order to improve optimization efficiency, the sequential optimization and reliability assessment method is used to decouple the double‐layer loop reliability‐based topology optimization. Initially, we derive the derivative of stress‐based performance functions with respect to heat source and load uncertainty variables, thereby facilitating the use of modified chaos control for assessing structural reliability and imposing constraints. The adjoint method and chain rule are utilized to obtain the derivative information of the performance function with respect to density variables, guiding the topology updates. We present five design examples to demonstrate the effectiveness of the presented method. Monte Carlo simulations for the optimized results are performed to show that the presented method can obtain a structure that meets reliability requirements.

Lung cancer diagnosis based on weighted convolutional neural network using gene data expression

Lung cancer is thought to be a genetic disease with a variety of unknown origins. Globocan2020 report tells in 2020 new cancer cases identified was 19.3 million and nearly 10.0 million died owed to cancer. GLOBOCAN envisages that the cancer cases will raised to 28.4 million in 2040. This charge is superior to the combined rates of the former generally prevalent malignancies, like breast, colorectal, and prostate cancers. For attribute selection in previous work, the information gain model was applied. Then, for lung cancer prediction, multilayer perceptron, random subspace, and sequential minimal optimization (SMO) are used. However, the total number of parameters in a multilayer perceptron can become extremely large. This is inefficient because of the duplication in such high dimensions, and SMO can become ineffective due to its calculating method and maintaining a single threshold value for prediction. To avoid these difficulties, our research presented a novel technique including Z-score normalization, levy flight cuckoo search optimization, and a weighted convolutional neural network for predicting lung cancer. This result findings show that the proposed technique is effective in precision, recall, and accuracy for the Kent Ridge Bio-Medical Dataset Repository.

Enhancement of α-galactosidase production using novel Actinoplanes utahensis B1 strain: sequential optimization and purification of enzyme.

α-Galactosidase is an important exoglycosidase belonging to the hydrolase class of enzymes, which has therapeutic and industrial potential. It plays a crucial role in hydrolyzing α-1,6 linked terminal galacto-oligosaccharide residues such as melibiose, raffinose, and branched polysaccharides such as galacto-glucomannans and galactomannans. In this study,Actinoplanes utahensis B1 was explored for α-galactosidase production, yield improvement, and activity enhancement by purification. Initially, nine media components were screened usingthe Plackett-Burman design (PBD). Among these components, sucrose, soya bean flour, and sodium glutamate were identified as the best-supporting nutrients for the highest enzyme secretion by A. Utahensis B1. Later, the Central Composite Design (CCD) was implemented to fine-tune the optimization of these components. Based on sequential statistical optimization methodologies, a significant, 3.64-fold increase in α-galactosidase production, from 16 to 58.37 U/mL was achieved. The enzyme was purified by ultrafiltration-I followed by multimode chromatography and ultrafiltration-II. The purity of the enzyme was confirmed by Sodium Dodecyl Sulphate-Polyacrylamide Agarose Gel Electrophoresis (SDS-PAGE) which revealed a single distinctive band with a molecular weight of approximately 72kDa. Additionally, it was determined that this process resulted in a 2.03-fold increase in purity. The purified α-galactosidase showed an activity of 2304 U/mL with a specific activity of 288 U/mg. This study demonstrates the isolation of Actinoplanes utahensis B1 and optimization of the process for the α-galactosidase production as well as single-step purification.

Sequential niching particle swarm optimization algorithm for localization of multiple damage locations using fiber bragg grating sensors

Structural Health Monitoring (SHM) systems have a potential to reduce lifecycle costs of structures through maintenance planning and lifetime extension. Hence, there is a lot of active research in the area for SHM of civil and mechanical structures. The SHM system should be low cost, suitable for continuous monitoring, able to detect small levels of damage. Guided waves (GW) based SHM techniques allow monitoring of large thin-walled structures with a few sensors and have been identified as the most promising of techniques for SHM. The GW based techniques allow mapping of damage very easily, and hence are a powerful tool for assessing the damage. The mapping of damage with high resolution is of course computationally expensive and hinders real-time decision making. Hence, methods are being developed to minimize the processing time.In this paper, the authors implement a sequential niching particle swarm optimization (PSO) for multiple damage detection. It is shown that the niching PSO is able to detect multiple damage scenarios effectively. The processing time is significantly better (order of magnitude) hence making it suitable for real-time assessment. The paper also presents some sensitivity studies, for determination of the thresholds. The validation has been conducted experimentally on an aluminum plate instrumented with piezo-actuators for actuation and FBG sensor for sensing. The results indicate that the niching PSO is suitable and indeed a better alternative than the brute-force techniques commonly used in literature.

Developing Cheap but Useful Machine Learning-Based Models for Investigating High-Entropy Alloy Catalysts.

This work aims to address the challenge of developing interpretable ML-based models when access to large-scale computational resources is limited. Using CoMoFeNiCu high-entropy alloy catalysts as an example, we present a cost-effective workflow that synergistically combines descriptor-based approaches, machine learning-based force fields, and low-cost density functional theory (DFT) calculations to predict high-quality adsorption energies for H, N, and NHx (x = 1, 2, and 3) adsorbates. This is achieved using three specific modifications to typical DFT workflows including: (1) using a sequential optimization protocol, (2) developing a new geometry-based descriptor, and (3) repurposing the already-available low-cost DFT optimization trajectories to develop a ML-FF. Taken together, this study illustrates how cost-effective DFT calculations and appropriately designed descriptors can be used to develop cheap but useful models for predicting high-quality adsorption energies at significantly lower computational costs. We anticipate that this resource-efficient philosophy may be broadly relevant to the larger surface catalysis community.

Machine learning and sequential subdomain optimization for ultrafast inverse design of 4D-printed active composite structures

Shape transformations of active composites (ACs) depend on the spatial distribution and active response of constituent materials. Voxel-level complex material distributions offer a vast possibility for attainable shape changes of 4D-printed ACs, while also posing a significant challenge in efficiently designing material distributions to achieve target shape changes. Here, we present an integrated machine learning (ML) and sequential subdomain optimization (SSO) approach for ultrafast inverse designs of 4D-printed AC structures. By leveraging the inherent sequential dependency, a recurrent neural network ML model and SSO are seamlessly integrated. For multiple target shapes of various complexities, ML-SSO demonstrates superior performance in optimization accuracy and speed, delivering results within second(s). When integrated with computer vision, ML-SSO also enables an ultrafast, streamlined design-fabrication paradigm based on hand-drawn targets. Furthermore, ML-SSO empowered with a splicing strategy is capable of designing diverse lengthwise voxel configurations, thus showing exceptional adaptability to intricate target shapes with different lengths without compromising high speed and accuracy. As a comparison, for the benchmark three-period shape, the finite element and evolutionary algorithm (EA) method was estimated to need 219 days for the inverse design; the ML-EA achieved the design in 54 min; the new ML-SSO with splicing strategy requires only 1.97 s. By further leveraging appropriate symmetries, the highly efficient ML-SSO is employed to design active shape changes of 4D-printed lattice structures. The new ML-SSO approach thus provides a highly efficient tool for the design of various 4D-printed, shape-morphing AC structures.

Development and optimization of Clitoria teratea synthesized silver nanoparticles and its application to nanogel systems for wound healing

Objective The present research deals with sequential optimization strategy based on Central Composite Design to optimize the process variables for efficient production of Clitoria teratea (CLT) synthesized silver nanoparticles (AgNPs) using biological synthesis. Methods Two substantial factors influencing the dependent variables viz UV-visible absorbance, particle size, zeta potential and polydispersity index (PDI) were identified as NaOH concentration, RH concentration, temperature as independent variables. In-vitro and ex-vivo studies of prepared CLT-AgNPs gel and marketed gel were carried out using dialysis membrane and egg membrane, respectively. In addition, antimicrobial study was also performed on the bacterial strains. Results The particles size (114 nm), PDI (0.45), and zeta potential (−29.5 mV) of optimized formulation were found, respectively. In-vitro profile of AgNPs from prepared CLT-AgNPs gel was noted (95.6%) in 8 h. It was found that the prepared CLT-AgNPs gel stimulates fibroblast and agranulocytosis development resulting better and timely wound healing. Conclusions The prepared CLT-AgNPs gel can be as a potential substitute in the management and treatment of acute and chronic wounds.

Application of multiple pressure management strategies in urban water distribution networks using sequential optimization

Controlling pressure in water distribution networks is an important management issue. Excess pressure in a network can lead to leakage, causing economic losses to water utilities. To address this problem, pressure management methods are used to reduce and balance pressure in the network. Popular methods include the pressure-reducing valve, variable-speed pump, and district metered area. Any of the above methods can be applied to a single network. However, considering all methods in a single optimization can lead to a very complex problem because each method behaves differently. To mitigate this problem, sequential optimization can be performed to simplify the optimization process. The objective of this study is to present the methodology of applying these methods sequentially and analyze the best sequence. The optimization was performed considering the benefit-cost ratio to gage the performance gain versus the investment cost. Then, the behavior of each method in the sequence was analyzed and compared using multi-criteria decision analysis, considering the economic, hydraulic, and environmental factors. The results show that the sequential optimization succeeded in iteratively reducing the leakage of the network and that there was an optimal path to a better optimization result.

An appropriate approach of design and machining of mechanical part by means of probabilistic multi‐objective optimization

AbstractAppropriate design and machining of mechanical part are of great importance to guarantee the comprehensive quality of product. In this paper, a rational approach of design and machining of mechanical part with multiple objectives is formulated by combining the probability‐based method for multi‐objective optimization with sequential optimization. The probability‐based method for multi‐objective optimization is employed to transfer the problem of multiple objectives into a single objective one in the respect of probability theory, while the “sequential algorithm for optimization” is used to conduct successive deep optimization quantitatively. By performing the assessment of preferable probability of each scheme and the “sequential algorithm for optimization”, the optimal design of materials processing is thus completed rationally. Furthermore, the approach is used to deal with thermal design of spindle system for computerized numerical control grinder, optimal designs of large floor‐type milling and boring machine spindle box and bending performance of B‐pillar under condition of side impact as examples, respectively. The results indicate the rationality of the approach.