Multi-level Optimization Framework Research Articles

Improving long COVID-related text classification: a novel end-to-end domain-adaptive paraphrasing framework.

The emergence of long COVID during the ongoing COVID-19 pandemic has presented considerable challenges for healthcare professionals and researchers. The task of identifying relevant literature is particularly daunting due to the rapidly evolving scientific landscape, inconsistent definitions, and a lack of standardized nomenclature. This paper proposes a novel solution to this challenge by employing machine learning techniques to classify long COVID literature. However, the scarcity of annotated data for machine learning poses a significant obstacle. To overcome this, we introduce a strategy called medical paraphrasing, which diversifies the training data while maintaining the original content. Additionally, we propose a Data-Reweighting-Based Multi-Level Optimization Framework for Domain Adaptive Paraphrasing, supported by a Meta-Weight-Network (MWN). This innovative approach incorporates feedback from the downstream text classification model to influence the training of the paraphrasing model. During the training process, the framework assigns higher weights to the training examples that contribute more effectively to the downstream task of long COVID text classification. Our findings demonstrate that this method substantially improves the accuracy and efficiency of long COVID literature classification, offering a valuable tool for physicians and researchers navigating this complex and ever-evolving field.

High-fidelity multi-level efficiency optimization of propeller for high altitude long endurance UAV

This paper describes the development of a high-fidelity multi-level optimization framework to maximize the efficiency of a propeller. Specifically, the discrete adjoint method optimizes the airfoil cluster in a three-dimensional state. Subsequently, parametric perturbation and flow pattern reconstruction methods are applied to the optimization of the twist angle and chord length distribution. Additionally, we use the concepts of perturbation loss and additional efficiency to quantify the influence of the viscosity loss on the efficiency and the potential chord length optimization at each blade section. The precision during the optimization is consistent with that of Computational Fluid Dynamics (CFD), which agrees well with the experimental data. The results indicate that the proposed optimization framework can efficiently combine high fidelity and low computational costs and accurately quantify the influence of complex three-dimensional flow characteristics on the optimal shape of the propeller. With the aid of the proposed optimization criterion, the three-dimensional flow can reach the state that is most conducive to improving the propeller efficiency by adjusting the shape parameters. Overall, this research's methodology and general rules provide a reference for the propeller design of a high altitude long endurance UAV.

Multilevel optimization framework to support self-sustainability of industrial processes for energy/material recovery using circular integration concept

Climate change, resource scarcity, waste reduction, and the lack of sustainability in process industries are vital concerns that civilization confronts. The process integration methodology may aid such sustainability. In this study, a novel chemical exergy resource recovery network is proposed for optimal energy and waste recovery in high-salinity-gradient chemical industries using a pressure-retarded osmosis membrane while indicating a self-sustainability and allocation in complex industrial networks. The mathematical programming paradigm is expressed as a multilevel optimization model to introduce novel ideas for explicitly modeling the trade-offs between waste and energy flows in circular integration while demonstrating industrial network’s environmental impact assessment. This problem is decomposed into two subproblems (SPs) that must be addressed sequentially. The first SP is designed to decrease the total cost of the network while reducing external resource use. The second SP formulates a mixed-integer nonlinear programming model with the objective of minimizing the environmental effects and exergy consumption rate of the network. A case study of a naphthalene-methaforming plant demonstrates the efficacy of the proposed methodology. The results showed that using a tri-level optimization technique, a considerable improvement in flowrate, total annualized cost, and energy recovery is obtained while limiting the network's environmental impact. The operating phase accounts for approximately 75% of the global warming potential output. The proposed tri-level approach based on Benders’ decomposition approach reduced the overall cost of the network by 19.29%, and 47.42 MW net power is recovered in the case study. In addition, the circular exergy use rate and environmental factor were reduced from 400 to 0.037 MW and from 2.569 to 1.481 kgCO2/year, respectively, using the tri-level decomposition approach.

A cascadic multilevel optimization framework for the concurrent design of the fiber-reinforced composite structure through the NURBS surface

A cascadic multilevel optimization framework for the concurrent design of the fiber-reinforced composite structure through the NURBS surface

Capacity withholding assessment of power systems considering coordinated strategies of virtual power plants and generation companies

This paper presents a multi-level optimization framework for power system operators' joint electricity markets capacity-withholding assessment. The main contribution of this research is that three capacity-withholding indices are introduced for day-ahead, intra-day, and real-time scheduling of the system that detect the capacity withholding and arbitrage opportunities of Virtual Power Plants (VPPs) and non-utility fossil-fueled GENeration COmpanies (GENCOs) in an ex-ante procedure. A three-level optimization process is used so that the system operator can estimate the coordinated bidding of VPPs/GENCOs in different energy and ancillary services markets to prevent the formation of withholding groups. The first level problem consists of two stages. The first stage estimates the optimal capacity withholding and arbitrage bidding strategy of VPPs/GENCOs, and the second stage determines the optimal system scheduling for the day-ahead horizon. A full competition algorithm is also carried out to evaluate the competition states of VPPs/GENCOs. The second and third level problems consist of two optimization stages for the intra-day and real-time optimization horizons. At the first stage of each level, the process estimates the coordinated bidding of VPPs/GENCOs, and at the second stage of each level, the system resources are optimally dispatched. The proposed method is applied to 30-bus and 118-bus IEEE test systems. The proposed algorithm reduced the maximum locational marginal prices of 30-bus and 118-bus test systems by about 57.04% and 44.73% compared to the normal and the worst-case contingency operating conditions, respectively. Furthermore, the proposed method reduced the average values of day-ahead, intra-day and real-time dynamic capacity withholding indices of the 118-bus test system by about 32.92%, 40.1%, and 46.85%, respectively.

Interactive Multi-level planning for energy management in clustered microgrids considering flexible demands

This paper presents a novel interactive multi-level planning strategy for the energy management of distribution networks with clustered microgrids (CMGs). CMGs are a group of microgrids with multiple renewable energy resources that comprise various technologies, such as photovoltaic systems, wind turbines, micro turbines and electric vehicles. This study develops an innovative multi-level optimization framework for the energy management coordination between microgrids and CMGs in the lower level, between clusters and distribution systems, and finally between distribution systems and upstream networks in the upper level. Accordingly, an hourly optimal energy management (HOEM) system is applied to minimize the multi-objective objective function for each level. The lower level may be operated in islanded or grid-connected mode in some hours. This is decided by changing switches between MGs, clusters, and grids, while the upper level is only operated in the grid-connected mode. Moreover, a demand response program that has a great effect on the hourly planning of switches is modeled in the upper level. The proposed model is tested on CMGs and actual distribution systems. The results show the significance of this planning strategy in the techno-economic aspects and optimal power transaction in the distribution system operation.

A simulation-based optimization model for watershed multi-scale irrigation water use with considering impacts of climate changes

It is attractive and advantageous to realize process-based optimization of irrigation water use on large scales. In this study, a simulation-based regional irrigation water optimization model (SRIOP) was developed for watershed-scale optimal spatial allocation based on dynamic responses of agro-hydrological process and its spatial differences. The SRIOP model tightly linked a field-scale agro-hydrological model (SWAP-EPIC) with an optimization algorithm (NSGA-II) in a regional multi-level optimization framework. It performed at three levels that sequentially dealt with optimal irrigation water allocation on field, irrigation district and irrigated watershed scales, with two objectives of maximizing yield/benefit and minimizing irrigation water. The model was tested with a case study in the major irrigation system (MIS) of middle reaches of the Heihe River basin, northwest China. The optimal spatial distribution of irrigation water showed a decreased variation coefficient (0.17) and increased irrigation benefit (9.8% higher) as compared with that of the present year, and better reflected the effects of multi-level collaboration and crop-soil field distribution. The impacts of climate condition changes on optimized results and resulting uncertainty were also quantified and evaluated through optimizations in different climate years. The case study indicated that the model is a computationally efficient and feasible approach for process-based optimization on large scales with considering impacts of climate condition changes.

Multi-level optimization framework for resilient distribution system expansion planning with distributed energy resources

A multi-stage optimization framework is proposed in this paper for the resilient electric distribution system expansion planning problem. The Non-utility Distributed Energy Resources (NDERs) can deliver their electricity to the distribution system in normal and external shock conditions. However, the NDERs bidding strategies in external shock conditions are an important issue and they can withhold their electricity generation in a contingent condition. The distribution system must tolerate the external shocks and determine the optimal contribution scenarios of NDERs in these conditions. The proposed algorithm determines the initial topology and system parameters of the planning horizon, at the first stage of optimization. Then, it explores the bidding strategies of NDERs in the second stage. At the third stage, the procedure calculates different market power indices to determine the optimal price of NDERs contributions in its different operational conditions and contracts with the selected NDERs. The problem has different sources of uncertainty that are modelled in the proposed algorithm. To assess the proposed method, 21-bus and 123-bus test systems are considered and the introduced procedure reduced the aggregated investment and operational costs of systems by about 11.82% and 23.74%, respectively, in comparison with the custom expansion planning exercise.

Investigation of microbial community interactions between Lake Washington methanotrophs using -------genome-scale metabolic modeling.

BackgroundThe role of methane in global warming has become paramount to the environment and the human society, especially in the past few decades. Methane cycling microbial communities play an important role in the global methane cycle, which is why the characterization of these communities is critical to understand and manipulate their behavior. Methanotrophs are a major player in these communities and are able to oxidize methane as their primary carbon source.ResultsLake Washington is a freshwater lake characterized by a methane-oxygen countergradient that contains a methane cycling microbial community. Methanotrophs are a major part of this community involved in assimilating methane from lake water. Two significant methanotrophic species in this community are Methylobacter and Methylomonas. In this work, these methanotrophs are computationally studied via developing highly curated genome-scale metabolic models. Each model was then integrated to form a community model with a multi-level optimization framework. The competitive and mutualistic metabolic interactions among Methylobacter and Methylomonas were also characterized. The community model was next tested under carbon, oxygen, and nitrogen limited conditions in addition to a nutrient-rich condition to observe the systematic shifts in the internal metabolic pathways and extracellular metabolite exchanges. Each condition showed variations in the methane oxidation pathway, pyruvate metabolism, and the TCA cycle as well as the excretion of formaldehyde and carbon di-oxide in the community. Finally, the community model was simulated under fixed ratios of these two members to reflect the opposing behavior in the two-member synthetic community and in sediment-incubated communities. The community simulations predicted a noticeable switch in intracellular carbon metabolism and formaldehyde transfer between community members in sediment-incubated vs. synthetic condition.ConclusionIn this work, we attempted to predict the response of a simplified methane cycling microbial community from Lake Washington to varying environments and also provide an insight into the difference of dynamics in sediment-incubated microcosm community and synthetic co-cultures. Overall, this study lays the ground for in silico systems-level studies of freshwater lake ecosystems, which can drive future efforts of understanding, engineering, and modifying these communities for dealing with global warming issues.

A multilevel optimization approach to route design and flight allocation taking aircraft sequence and separation constraints into account

This paper presents the development of a multilevel optimization framework for the design and selection of departure routes, and the distribution of aircraft movements among these routes, while taking the sequence and separation requirements for aircraft on runways and along selected routes into account. The main aim of the framework is to minimize aircraft noise impact on communities around an airport, and the associated fuel consumption. The proposed framework features two consecutive steps. In the first step, for each given Standard Instrument Departure (SID), multi-objective trajectory optimization is utilized to generate a comprehensive set of possible alternative routes. The obtained set is subsequently used as input for the optimization problem in the second step. In this step, the selection of routes for each SID and the distribution of aircraft movements among these routes are optimized simultaneously. To ensure the feasibility of optimized solutions for an entire operational day, the sequence and separation requirements for aircraft on runways and along selected routes are included in this second phase. In order to address these issues, three novel techniques are developed and added to a previously developed multilevel optimization framework, viz., a runway assignment model, a conflict detection algorithm, and a rerouting technique. The proposed framework is applied to a realistic case study at Amsterdam Airport Schiphol in the Netherlands, in which 599 departure flights and 13 different SIDs are considered. The optimization results show that the proposed model can offer conflict-free solutions, one of which can lead to a reduction in the number of people annoyed of up to 21%, and a reduction in fuel consumption of 8% relative to the reference case solution.

A Multi-Level Optimization Framework for Simultaneous Grasping and Motion Planning

We present an optimization framework for grasp and motion planning in the context of robotic assembly. Typically, grasping locations are provided by higher level planners or as input parameters. In contrast, our mathematical model simultaneously optimizes motion trajectories, grasping locations, and other parameters such as the pose of an object during handover operations. The input to our framework consists of a set of objects placed in a known configuration, their target locations, and relative timing information describing when objects need to be picked up, optionally handed over, and dropped off. To allow robots to reason about the way in which grasping locations govern optimal motions, we formulate the problem using a multi-level optimization scheme: the top level optimizes grasping locations; the mid-layer level computes the configurations of the robot for pick, drop and handover states; and the bottom level computes optimal, collision-free motions. We leverage sensitivity analysis to compute derivatives analytically (how do grasping parameters affect IK solutions, and how these, in turn, affect motion trajectories etc.), and devise an efficient numerical solver to generate solutions to the resulting optimization problem. We demonstrate the efficacy of our approach on a variety of assembly and handover tasks performed by a dual-armed robot with parallel grippers.

Efficient Design Optimization of Acoustic Liners for Engine Noise Reduction

Engine noise reduction using acoustic liner panels is an established technology in the design of bypass turbofan engines. Using acoustic impedance modeling techniques in aeroacoustic simulations, quick assessment of different liner configurations can be made. To minimize the engine noise in the best possible way, optimal liner configurations are ideally determined by employing numerical optimization methods. In general, to exhaust the full potential of the lining technology, it is desired to use a large number of design parameters in the optimization process. In the present work, a multilevel optimization framework is presented for the design of acoustic liner panels, which can efficiently handle large design vectors. The multilevel method combines a global search strategy with a local gradient-based optimization method. The discrete adjoint solver, which is used to evaluate the liner sensitivities required in the local search, has been developed using algorithmic differentiation techniques. This approach allows consistent treatment of boundary conditions in the adjoint part, thereby increasing the robustness and accuracy of the adjoint solver. In this way, the discrete adjoint solver fully incorporates the features of the underlying aeroacoustic solver, such as the impedance modeling and nonreflecting boundary conditions. The feasibility and the efficiency of the multilevel optimization strategy are demonstrated by finding the optimal liner parameters in a turbofan engine bypass duct configuration.

Personalized collaborative plans for robot-assisted dressing via optimization and simulation

Robots could be a valuable tool for helping with dressing but determining how a robot and a person with disabilities can collaborate to complete the task is challenging. We present task optimization of robot-assisted dressing (TOORAD), a method for generating a plan that consists of actions for both the robot and the person. TOORAD uses a multilevel optimization framework with heterogeneous simulations. The simulations model the physical interactions between the garment and the person being dressed, as well as the geometry and kinematics of the robot, human, and environment. Notably, the models for the human are personalized for an individual’s geometry and physical capabilities. TOORAD searches over a constrained action space that interleaves the motions of the person and the robot with the person remaining still when the robot moves and vice versa. In order to adapt to real-world variation, TOORAD incorporates a measure of robot dexterity in its optimization, and the robot senses the person’s body with a capacitive sensor to adapt its planned end effector trajectories. To evaluate TOORAD and gain insight into robot-assisted dressing, we conducted a study with six participants with physical disabilities who have difficulty dressing themselves. In the first session, we created models of the participants and surveyed their needs, capabilities, and views on robot-assisted dressing. TOORAD then found personalized plans and generated instructional visualizations for four of the participants, who returned for a second session during which they successfully put on both sleeves of a hospital gown with assistance from the robot. Overall, our work demonstrates the feasibility of generating personalized plans for robot-assisted dressing via optimization and physics-based simulation.

Multilevel Optimization Framework for Hierarchical Stiffened Shells Accelerated by Adaptive Equivalent Strategy

In order to improve the post-buckling optimization efficiency of hierarchical stiffened shells, a multilevel optimization framework accelerated by adaptive equivalent strategy is presented in this paper. Firstly, the Numerical-based Smeared Stiffener Method (NSSM) for hierarchical stiffened shells is derived by means of the numerical implementation of asymptotic homogenization (NIAH) method. Based on the NSSM, a reasonable adaptive equivalent strategy for hierarchical stiffened shells is developed from the concept of hierarchy reduction. Its core idea is to self-adaptively decide which hierarchy of the structure should be equivalent according to the critical buckling mode rapidly predicted by NSSM. Compared with the detailed model, the high prediction accuracy and efficiency of the proposed model is highlighted. On the basis of this adaptive equivalent model, a multilevel optimization framework is then established by decomposing the complex entire optimization process into major-stiffener-level and minor-stiffener-level sub-optimizations, during which Fixed Point Iteration (FPI) is employed to accelerate convergence. Finally, the illustrative examples of the multilevel framework is carried out to demonstrate its efficiency and effectiveness to search for the global optimum result by contrast with the single-level optimization method. Remarkably, the high efficiency and flexibility of the adaptive equivalent strategy is indicated by compared with the single equivalent strategy.

Multilevel framework for large-scale global optimization

Large-scale global optimization (LSGO) algorithms are crucially important to handle real-world problems. Recently, cooperative co-evolution (CC) algorithms have successfully been applied for solving many large-scale practical problems. Many applications have imbalanced subcomponents where the size of subcomponents and their contribution to the objective function value are different. CC algorithms often lose their efficiency on LSGO problems with the imbalanced subcomponents; since they do not consider the imbalance aspect of variables. In this paper, we propose a multilevel optimization framework based on variables effect (called MOFBVE) which optimizes several subcomponents of the most important variables at earlier stages of optimization procedure before optimizing the problem with the original search space at its last stage. Sensitivity analysis (SA) method determines how the variation in the outputs of the model can be influenced by the variation of its input parameters. MOFBVE computes the main effect of variables using an SA method, Morris screening, and then it employs the k-means clustering method to construct groups including variables with the similar effects on the fitness value. The constructed groups are sorted in the descending order based on their contribution on the fitness value and the top groups are selected as the levels of the important variables. MOFBVE can reduce the complexity of search space to work with a simplified model to achieve an efficient exploration. The performance of MOFBVE is benchmarked on the imbalanced LSGO problems, i.e., two individually modified CEC-2010 and the CEC-2013 LSGO benchmark functions. The simulated experiments confirmed that MOFBVE obtains a promising performance on the majority of the imbalanced LSGO test functions. Also, MOFBVE is compared with state-of-the-art CC algorithms; and the results show that it is better than or at least comparable to CC algorithms.

Techniques for Multilevel Design Optimization of Permanent Magnet Motors

The multilevel method has been presented for design optimization of electrical machines and drive systems for optimal system performances and efficiency in our previous work. For framework design of the multilevel optimization method, four techniques are presented in this paper, including the sizing equation, local sensitivity analysis, global sensitivity analysis, and design of experiments techniques. For each technique, a general and theoretical analysis procedure is presented before the application study. To demonstrate the effectiveness, a permanent magnet claw-pole motor with soft magnetic composite core and 3-D finite-element analysis model is investigated to minimize the material cost and maximize the output power while keeping the volume constant. The calculated motor performance based on this 3-D finite-element model has been verified by the experimental results. As shown, these techniques are simple to implement, and the resultant multilevel optimization framework can significantly improve the motor performance and reduce the required sample number of finite-element analysis.

A multilevel layout algorithm for visualizing physical and genetic interaction networks, with emphasis on their modular organization

BackgroundGraph drawing is an integral part of many systems biology studies, enabling visual exploration and mining of large-scale biological networks. While a number of layout algorithms are available in popular network analysis platforms, such as Cytoscape, it remains poorly understood how well their solutions reflect the underlying biological processes that give rise to the network connectivity structure. Moreover, visualizations obtained using conventional layout algorithms, such as those based on the force-directed drawing approach, may become uninformative when applied to larger networks with dense or clustered connectivity structure.MethodsWe implemented a modified layout plug-in, named Multilevel Layout, which applies the conventional layout algorithms within a multilevel optimization framework to better capture the hierarchical modularity of many biological networks. Using a wide variety of real life biological networks, we carried out a systematic evaluation of the method in comparison with other layout algorithms in Cytoscape.ResultsThe multilevel approach provided both biologically relevant and visually pleasant layout solutions in most network types, hence complementing the layout options available in Cytoscape. In particular, it could improve drawing of large-scale networks of yeast genetic interactions and human physical interactions. In more general terms, the biological evaluation framework developed here enables one to assess the layout solutions from any existing or future graph drawing algorithm as well as to optimize their performance for a given network type or structure.ConclusionsBy making use of the multilevel modular organization when visualizing biological networks, together with the biological evaluation of the layout solutions, one can generate convenient visualizations for many network biology applications.

OptCom: A Multi-Level Optimization Framework for the Metabolic Modeling and Analysis of Microbial Communities

Microorganisms rarely live isolated in their natural environments but rather function in consolidated and socializing communities. Despite the growing availability of high-throughput sequencing and metagenomic data, we still know very little about the metabolic contributions of individual microbial players within an ecological niche and the extent and directionality of interactions among them. This calls for development of efficient modeling frameworks to shed light on less understood aspects of metabolism in microbial communities. Here, we introduce OptCom, a comprehensive flux balance analysis framework for microbial communities, which relies on a multi-level and multi-objective optimization formulation to properly describe trade-offs between individual vs. community level fitness criteria. In contrast to earlier approaches that rely on a single objective function, here, we consider species-level fitness criteria for the inner problems while relying on community-level objective maximization for the outer problem. OptCom is general enough to capture any type of interactions (positive, negative or combinations thereof) and is capable of accommodating any number of microbial species (or guilds) involved. We applied OptCom to quantify the syntrophic association in a well-characterized two-species microbial system, assess the level of sub-optimal growth in phototrophic microbial mats, and elucidate the extent and direction of inter-species metabolite and electron transfer in a model microbial community. We also used OptCom to examine addition of a new member to an existing community. Our study demonstrates the importance of trade-offs between species- and community-level fitness driving forces and lays the foundation for metabolic-driven analysis of various types of interactions in multi-species microbial systems using genome-scale metabolic models.

Economic-Oriented Stochastic Optimization in Advanced Process Control of Chemical Processes

Finding the optimal operating region of chemical processes is an inevitable step toward improving economic performance. Usually the optimal operating region is situated close to process constraints related to product quality or process safety requirements. Higher profit can be realized only by assuring a relatively low frequency of violation of these constraints. A multilevel stochastic optimization framework is proposed to determine the optimal setpoint values of control loops with respect to predetermined risk levels, uncertainties, and costs of violation of process constraints. The proposed framework is realized as direct search-type optimization of Monte-Carlo simulation of the controlled process. The concept is illustrated throughout by a well-known benchmark problem related to the control of a linear dynamical system and the model predictive control of a more complex nonlinear polymerization process.

Integrated structure/control law design by multilevel optimization

A new approach to integrated structure/control law design based on multilevel optimization is presented. This new approach is applicable to aircraft and spacecraft and allows for the independent design of the structure and control law. Integration of the designs is achieved through use of an upper level coordination problem formulation within the multilevel optimization framework. The method requires the use of structure and control law design sensitivity information. A general multilevel structure/control law design problem formulation is given, and the use of Linear Quadratic Gaussian (LQG) control law design and design sensitivity methods within the formulation is illustrated. Results of three simple integrated structure/control law design examples are presented. These results show the capability of structure and control law design tradeoffs to improve controlled system performance within the multilevel approach.