Optimization Framework Research Articles

A mechanics-based data-free Problem Independent Machine Learning (PIML) model for large-scale structural analysis and design optimization

Machine learning (ML) enhanced fast structural analysis and design recently attracted considerable attention. In most related works, however, the generalization ability of the ML model and the massive cost of dataset generation are the two most criticized aspects. This work combines the advantages of the universality of the substructure method and the superior predictive ability of the operator learning architecture. Specifically, using a novel mechanics-based loss function, lightweight neural network mapping from the material distribution inside a substructure and the corresponding continuous multiscale shape function is well-trained without preparing a dataset. In this manner, a problem machine learning model (PIML) that is generally applicable for efficient linear elastic analysis and design optimization of large-scale structures with arbitrary size and various boundary conditions is proposed. Several examples validate the effectiveness of the present work on efficiency improvement and different kinds of optimization problems. This PIML model-based design and optimization framework can be extended to large-scale multiphysics problems.

Reliability-constrained configuration optimization for integrated power and natural gas energy systems: A stochastic approach

With the escalating dependence on electricity and natural gas infrastructure, ensuring both reliability and economic efficiency becomes paramount. It necessitates reliability centric measures to mitigate disruptions that could cascade between these interconnected systems. To address this challenges, this paper introduces a reliability-constrained two-stage stochastic model to optimize power-to-gas (P2 G) and gas-to-power (G2P) unit placement and sizing, aiming to enhance the reliability of both systems under stochastic scenarios. The proposed model, employing Sequential Monte Carlo (SMC) within its optimization framework, seeks to minimize investment, operation, and reliability costs. By addressing temporal uncertainties in component outages for both systems and considering uncertainties in power and gas system loads with a high temporal resolution and annual load growth, the model provides a comprehensive reliability perspective. Furthermore, sensitivity analysis is conducted to explore the impact of varying Values of Lost Load (VOLL) on the planning results. Numerical evaluation, using two integrated energy systems including IEEE 14-bus-10-gas node, and large-scale energy systems including IEEE 118-bus-85-gas node integrated power-gas system (IPGS), demonstrates a significant 12.53 % improvement in overall system reliability. Furthermore, a 2.81 % reduction in operation costs and a substantial 26.3 % reduction in reliability costs, validating the effectiveness of the proposed model.

Motion priority optimization framework towards automated and teleoperated robot cooperation in industrial recovery scenarios

In this study, we introduce an optimization framework to enhance the efficiency of motion priority design in scenarios involving automated and teleoperated robots within an industrial recovery context. The increasing utilization of industrial robots at manufacturing sites has been instrumental in reducing human workload. Nevertheless, achieving effective human–robot collaboration/cooperation (HRC) remains a challenge, especially when human workers and robots share a workspace for collaborative tasks. For instance, when an industrial robot encounters a failure, such as dropping an assembling part, it triggers the suspension of the corresponding factory cell for safe recovery. Given the limited capacity of pre-programmed robots to rectify such failures, human intervention becomes imperative, requiring entry into the robot workspace to address the dropped object while the robot system is halted. This discontinuous manufacturing process results in productivity loss. Robotic teleoperation has emerged as a promising technology enabling human workers to undertake high-risk tasks remotely and safely. Our study advocates for the incorporation of robotic teleoperation in the recovery process during manufacturing failure scenarios, which is referred to as “Cooperative Tele-Recovery”. Our proposed approach involves formulating priority rules designed to facilitate collision avoidance between manufacturing and recovery robots. This, in turn, ensures a continuous manufacturing process with minimal production loss within a configurable risk limitation. We present a comprehensive motion priority optimization framework composed of an HRC simulator and a cooperative multi-robot controller to identify optimal parameters for the priority function. The framework dynamically adjusts the allocation of motion priorities for manufacturing and recovery robots while adhering to predefined risk limitations. Through quantitative and qualitative assessments, we validate the novelty of our concept and demonstrate its feasibility.

A Stochastic Energy-Efficient Robust Simulation-Based Truck Dispatching Optimization for Simultaneous GHG Mitigation and Operational Excellence in Open-Pit Mines

Almost all currently active large open-pit mines across the globe heavily rely on fleet of shovels and haul trucks for material handling. This material handling system has a significant contribution not only to the production cost of the mine (+50%), but also to its energy consumption (+41%) and greenhouse gas (GHG) emissions (+37%). However, traditionally, operation management in general and truck dispatching, in particular, focused primarily on production targets and operational efficiency, neglecting environmental considerations in open-pit mines. Nowadays, growing awareness of climate change and responsible resource extraction has shifted, however, the focus towards sustainable practices throughout the mining industry, open-pit mining not excluded. The significant contribution of this study lies in the development of an integrated simulation and optimization framework that simultaneously accounts for anthropogenic GHG emissions resulting from energy (fuel) consumption and enhances the operational efficiency of the material handling system, thus yielding direct economic and environmental benefits. The paper fulfills its goals with the aim of simultaneous achievement of four interdependent objectives: minimizing the deviations from target production rates set by strategic plans, minimizing the shovel idle time, minimizing the truck wait time, and minimizing the truck fuel consumption. The performance and the robustness of the proposed framework has been evaluated in a case study at Gol-E-Gohar mine in Iran, demonstrating a successful achievement of up to 6% reduction in fuel consumption per tonne of production leading to a significant overall decrease of up to 20,000 liters in fuel consumption, equivalent to 61,000 carbon dioxide (CO2) emissions, in a 10-day operation with 12 hours of operation per day.

Denoising diffusion probabilistic models for generative alloy design

Inverse material design is an extremely challenging optimization task made difficult by, in part, the highly nonlinear relationship linking performance with composition. Quantitative approaches have improved significantly owing to advances in high throughput experimentation and computational thermodynamics. However, existing physics-based tools are mostly forward models; input a chemistry and obtain a prediction. More recently the materials community has leveraged advances in the machine learning community to establish novel inverse design frameworks. Very recently denoising diffusion probabilistic models have been shown to be extremely powerful generators producing synthetic data of various modalities e.g. images, text, audio, tables, etc.. In this work a novel framework for alloy design and optimization is proposed leveraging these class of models. Five key generative tasks are demonstrated (1) unconditional generation (2) composition conditioned generation (3) property conditioned generation (4) multi-feedstock conditioned generation and (5) generative optimization. These methods were tested on three case studies: high entropy alloy design, superalloy binder jet additive manufacturing, and in-situ dual-feedstock wire-arc additive manufacturing. Results indicate that the established models are extremely flexible, expressive, and robust. The architecture’s flexibility and training procedure empower the model to learn complex intra-compositional and composition-property relationships. Furthermore, the probabilistic nature of these models makes them well suited for addressing solution non-uniqueness and tackling uncertainty quantification tasks. While the fidelity and quantity of the underlying training data is paramount, we envision that future alloy design frameworks will make extensive use of these kinds of machine learning models as “search” tools bolstering the utility of experimental and computational approaches.

A multi-objective optimization framework for regional land-use allocation: Fully utilizing terrestrial vegetation to mitigate carbon emissions

Human-induced changes in land use and land cover (LULC) considerably impact the global carbon cycle. Rational LULC optimization with full utilization of terrestrial vegetation facilitates reconciling carbon reduction and socio-economic development. To this end, this study develops a multi-objective framework for low-carbon LULC spatial optimization that incorporates carbon suitability, historical suitability, and socio-economic needs. The framework is applied to Hubei Province, China. The key findings include: (1) Considering spatial variations in vegetation carbon sinks could lead to remarkable carbon reductions of 2.75 × 106 t C by 2030 and 7.68 × 106 t C by 2060. It would also lower carbon emissions per unit of gross domestic product (GDP) and enhance carbon sequestration per unit of vegetation by 2030, signifying enhanced carbon reduction efficiency. (2) Integrating carbon sequestration potential and historical suitability can considerably minimize emissions while sustaining socio-economic growth. Our method, compared with strategies based solely on historical development laws, can reduce carbon emissions by approximately 1.3–1.6 × 106 t C·a−1. (3) We also propose targeted low-carbon development suggestions for different LULCs. This study provides optimal spatial LULC allocations and strategic advice for regional LULC planning aimed at low-carbon development, demonstrating the feasibility of reconciling the tension between carbon reduction and socio-economic growth at the regional scale.

Optimal Operation Framework of the Hydrogen-electric Integrated Energy System by Considering the Power to Hydrogen Efficiency Factor

Optimal Operation Framework of the Hydrogen-electric Integrated Energy System by Considering the Power to Hydrogen Efficiency Factor

Techno-economic analysis and optimization of a hybrid solar-wind-biomass-battery framework for the electrification of a remote area: A case study

An off-grid hybrid energy framework on the basis of wind turbines and photovoltaic panels as the primary source of energy and a biogas generator and energy storage unit as a back-up system, was considered for clean energy generation in a rural area in Semnan, Iran. This study was aimed at designing an optimized hybrid renewable framework with minimal costs and CO2 emissions and highest reliability, utilizing comprehensive modeling based on two energy management strategies. The decision-making variables in this work were the type and number of wind turbines, number of storage units, capacity of solar panels, and the capacity of biogas generator. Then variations in turbine model, inflation rate and biofuel prices and their effect on the optimal design of the hybrid framework was investigated in a range of uncertainties (0, 1, and 2.5 %). Additionally, the powerplant emissions penalty cost rate was also considered. Eventually, the optimal hybrid frameworks were economically and environmentally compared with a stand-alone diesel generator. Optimization results confirmed the superiority of the proposed hybrid wind-solar-biomass-battery system in comparison with other investigated systems. The cost of energy for the optimal design including biogas generator (22 kW), photovoltaic panels (30.7 kW), a 10 kW wind turbine, 11 batteries and an inverter (15.1 kW), while considering the inflation rate of 10 % and the uncertainty rate of 0 %, is equal to 0.201 $/kWh. Additionally, the selected optimal design emitted 97 % less CO2 annually, in comparison with the similarly-sized diesel-based energy system.

Two-stage dynamic real-time optimization framework using parameter-dependent differential dynamic programming

The purpose of chemical process control includes proactive adjustment of the operation to make the most profit out of it. Within this context, real-time optimization (RTO) is proposed and extended to dynamic RTO (DRTO) in the hierarchical control structure, usually having model predictive control (MPC) below. However, online tractability confined the model complexity of RTO and MPC, which results in model inconsistency and, even, incompatible solutions. Here we use parameter-dependent differential dynamic programming (PDDP) to incorporate the closed-loop behavior of the controller in an RTO layer to reduce problem complexity and online computation time. The adaptive control performance of PDDP and the efficacy of closed-loop DRTO formulation with PDDP is demonstrated with the reaction-storage-separation network system control. Consequently, PDDP provides a useful parameterization method to express closed-loop system dynamics, which enables fast feedback control and integrated plant optimization.

PERF: Panoramic Neural Radiance Field From a Single Panorama.

Neural Radiance Field (NeRF) has achieved substantial progress in novel view synthesis given multi-view images. Recently, some works have attempted to train a NeRF from a single image with 3D priors. They mainly focus on a limited field of view with a few occlusions, which greatly limits their scalability to real-world 360-degree panoramic scenarios with large-size occlusions. In this paper, we present PERF, a 360-degree novel view synthesis framework that trains a panoramic neural radiance field from a single panorama. Notably, PERF allows 3D roaming in a complex scene without expensive and tedious image collection. To achieve this goal, we propose a novel collaborative RGBD inpainting method and a progressive inpainting-and-erasing method to lift up a 360-degree 2D scene to a 3D scene. Specifically, we first predict a panoramic depth map as initialization given a single panorama and reconstruct visible 3D regions with volume rendering. Then we introduce a collaborative RGBD inpainting approach into a NeRF for completing RGB images and depth maps from random views, which is derived from an RGB Stable Diffusion model and a monocular depth estimator. Finally, we introduce an inpainting-and-erasing strategy to avoid inconsistent geometry between a newly-sampled view and reference views. The two components are integrated into the learning of NeRFs in a unified optimization framework and achieve promising results. Extensive experiments on Replica and a new dataset PERF-in-the-wild demonstrate the superiority of our PERF over state-of-the-art methods. Our PERF can be widely used for real-world applications, such as panorama-to-3D, text-to-3D, and 3D scene stylization applications.

Cross regional online food delivery: Service quality optimization and real-time order assignment

Online food delivery (OFD) represents a rapidly evolving e-business application that leverages cloud computing data centers, playing a crucial role in meeting the demands of urban lifestyles. With diverse order fulfillment features and increasing expectations for service quality, the task of effectively assigning riders for timely long-distance, cross-regional deliveries presents a significant engineering challenge. Previous studies often relied on traditional rider allocation methods that fail to account for varying capacities, or they utilized non-intelligent systems that did not adequately address fluctuating order demands and service delays. In this study, we introduce a robust Mixed Integer Linear Programming (MILP) optimization framework designed to minimize the total service time and delivery cost for cross-regional orders. This framework divides a large OFD area into multiple regions and utilizes both transfer vehicles and riders to optimize deliveries. To enhance the predictive accuracy of our model, we incorporate advanced machine learning techniques. Specifically, we employ the Long Short-Term Memory (LSTM) model to forecast regional order demands accurately, reflecting the dynamic nature of the marketplace. Additionally, Extreme Gradient Boosting (XGBoost) is tailored to dynamically predict travel times from restaurants to customer locations, facilitating more precise scheduling and resource allocation within the MILP framework. These machine learning techniques significantly bolster the MILP framework by providing detailed, accurate predictions that improve decision-making processes and adaptability to real-time conditions. Acknowledging the complexity of this optimization problem, we further enhance our approach by integrating a meta-heuristic algorithm, Adaptive Large Neighbor Search (ALNS), which efficiently assigns orders to the appropriate transfer vehicles and riders within polynomial time. Our Cross Regional Online Food Delivery (XROFD) system is meticulously designed to optimize both customer satisfaction and rider incentives. Simulation experiments confirm that the XROFD system not only reduces service times and delivery costs but also markedly enhances customer satisfaction and provides superior incentives for riders, outperforming existing state-of-the-art methods.

Optimal design and dispatch of phosphoric acid fuel cell hybrid system with direct heat recovery through coupled calculation and artificial intelligence-based optimization

A phosphoric acid fuel cell (PAFC) is the first-generation of modern fuel cells that can use waste heat in various ways. This study presents a PAFC hybrid system with direct heat recovery. Direct heat recovery uses PAFC cooling water as the working fluid for the steam turbine, simplifying the system and improving performance. The PAFC in this hybrid system acts as a power generator and evaporator. The hybrid system operates in power generation mode (mode_pg) and combined heat and power mode (mode_chp). The overall behavior of the hybrid system was analyzed using a coupled calculation method. Under mode_pg, the electric power and efficiency of the hybrid system were 37.0% and 10.5% higher, respectively, than those of a conventional system. The thermal power and CHP efficiency in mode_chp were 35.7% and 8.8%p higher, respectively, than those of a conventional system. The optimal dispatch was obtained using an artificial intelligence-based optimization framework that considered two scenarios. As a result, the optimal demand was 25.8% higher when the hybrid system with the optimal dispatch was used. In addition, the economic and environmental indices were 7.3% and 6.8% lower, respectively, than the conventional system, highlighting the feasibility and eco-friendliness of the hybrid system.

A Novel Surrogate-Assisted Multi-Objective Well Control Parameter Optimization Method Based on Selective Ensembles

Multi-objective optimization algorithms are crucial for addressing real-world problems, particularly with regard to optimizing well control parameters, which are often computationally expensive due to their reliance on numerical simulations. Surrogate-assisted models help to reduce this computational burden, but their effectiveness depends on the quality of the surrogates, which can be affected by candidate dimension and noise. This study proposes a novel surrogate-assisted multi-objective optimization framework (MOO-SESA) that combines selective ensemble support-vector regression with NSGA-II. The framework’s uniqueness lies in its adaptive selection of a diverse subset of surrogates, established prior to iteration, to enhance accuracy, robustness, and computational efficiency. To our knowledge, this is the first instance in which selective ensemble techniques with multi-objective optimization have been applied to reservoir well control problems. Through employing an ensemble strategy for improving the quality of the surrogate model, MOO-SESA demonstrated superior well control scenarios and faster convergence compared to traditional surrogate-assisted models when applied to the SPE10 and Egg reservoir models.

Energy management method for battery swapping station combining reinforcement learning and mathematical optimization

Abstract The rapid proliferation of electric vehicles has spurred the expansion of scalable Battery Swapping Station (BSS) to cater to the demand for swift charging. However, the current energy management faces challenges in coping with the fluctuation of charging batteries in scalable BSS, the unpredictability of electricity prices and battery demand, as well as the intricacy of demand response. So, this paper presents a two-layer optimization framework for energy management in scalable BSS. The framework decomposes the power scheduling problem in BSS into two subproblems and solves them with deep reinforcement learning and mathematical optimization. The upper layer uses deep reinforcement learning to schedule the BSS’s total power, while the lower layer uses mathematical optimization to allocate the power to each charging pile. The two layers cooperate to achieve an efficient solution. The experiments demonstrate that the proposed method can reduce the operating cost of the BSS, satisfy the safety and user’s demand, and facilitate grid demand response, in BSSs of different sizes. The method is an effective solution for power scheduling of scalable BSS.

Data-driven airfoil shape optimization framework for enhanced flutter performance

Data-driven airfoil shape optimization framework for enhanced flutter performance

Easy-to-use formulations based on the homogenization theory for vascular stent design and mechanical characterization

Background and objectivesVascular stents are scaffolding structures implanted in the vessels of patients with obstructive disease. Stents are typically designed as cylindrical lattice structures characterized by the periodic repetition of unit cells. Their design, including geometry and material characteristics, influences their mechanical performance and, consequently, the clinical outcomes. Computational optimization frameworks have proven to be effective in assisting the design phase of vascular stents, facilitating the achievement of enhanced mechanical performances. However, the reliance on time-consuming simulations and the challenge of automating the design process limit the number of design evaluations and reduce optimization efficiency. In this context, a rapid and automated method for the mechanical characterization of vascular stents is presented, taking the stent geometry, conceived as the periodic repetition of a unit cell, and material as input and providing the mechanical response of the stent as output. MethodsVascular stents were assumed to be thin-walled hollow cylinders sharing the same macroscopic geometrical characteristics as the cylindrical lattice structure but composed of an anisotropic homogenized material. Homogenization theory was applied to average the microscopic inhomogeneities at the stent unit cell level into a homogenized material at the macro-scale, enabling the calculation of the associated homogenized material tensor. Analytical formulations were derived to relate the stent mechanical behavior to the homogenized stiffness tensor, considering linear elastic theory for thin-walled hollow cylinders and three loading scenarios of relevance for vascular stents: radial crimping; axial traction; torsion. Validation was conducted by comparing the derived analytical formulations with results obtained from finite element analyses on typical stent designs. ResultsHomogenized stiffness tensors were computed for the unit cells of three stent designs, revealing insights into their mechanical performance, including whether they exhibit auxetic behavior. The derived analytical formulations were successfully validated with finite element analyses, yielding low relative differences in the computed values of foreshortening, radial, axial and torsional stiffnesses for all three stents. ConclusionsThe proposed method offers a rapid, fully automated procedure that facilitates the assessment of the mechanical behavior of vascular stents and is suitable for effective integration into computational optimization frameworks.

An efficient water quality index forecasting and categorization using optimized Deep Capsule Crystal Edge Graph neural network.

The world's freshwater supply, predominantly sourced from rivers, faces significant contamination from various economic activities, confirming that the quality of river water is critical for public health, environmental sustainability, and effective pollution control. This research addresses the urgent need for accurate and reliable water quality monitoring by introducing a novel method for estimating the water quality index (WQI). The proposed approach combines cutting-edge optimization techniques with Deep Capsule Crystal Edge Graph neural networks, marking a significant advancement in the field. The innovation lies in the integration of a Hybrid Crested Porcupine Genghis Khan Shark Optimization Algorithm for precise feature selection, ensuring that the most relevant indicators of water quality (WQ) are utilized. Furthermore, the use of the Greylag Goose Optimization Algorithm to fine-tune the neural network's weight parameters enhances the model's predictive accuracy. This dual optimization framework significantly improves WQI prediction, achieving a remarkable mean squared error (MSE) of 6.7 and an accuracy of 99%. By providing a robust and highly accurate method for WQ assessment, this research offers a powerful tool for environmental authorities to proactively manage river WQ, prevent pollution, and evaluate the success of restoration efforts. PRACTITIONER POINTS: Novel method combines optimization and Deep Capsule Crystal Edge Graph for WQI estimation. Preprocessing includes data cleanup and feature selection using advanced algorithms. Deep Capsule Crystal Edge Graph neural network predicts WQI with high accuracy. Greylag Goose Optimization fine-tunes network parameters for precise forecasts. Proposed method achieves low MSE of 6.7 and high accuracy of 99%.

An optimization framework to provide volunteers with task selection autonomy and group opportunities

Nonprofit Organizations (NPOs) rely on volunteers to support community needs but struggle with making strategic volunteer-to-task assignments to enable volunteer satisfaction, and completion of complex tasks. Creation of volunteer groups and their assignment to NPO tasks can help achieve these goals by providing volunteers with opportunity for networking, collaboration, and peer learning. However, strategically creating ideal assignments is challenging because (i) there are exponentially many ways a set of volunteers can be assigned in groups; and (ii) NPOs tend to have limited and uncertain data concerning volunteers’ personal preferences, availabilities, and motivations to participate. To address these challenges, this research contributes by introducing an integer programming framework to offer volunteers a menu of tasks to choose from and then based on volunteers’ willingness information, creates ideal homogenous volunteer group assignments. These groups are created such that the group collectively meet a task’s skill requirements and groups of volunteers of similar skill and affinity levels are prioritized. We apply the developed methodology to a case study based on a partner NPO that works with remote volunteers from multiple countries to produce online educational content. The menu creation method can improve NPO and volunteer-based performance metrics, where the most improvement is observed when a NPO is faced with very picky volunteers. Presenting volunteers with larger menus of tasks also leads to an improvement in ideal group creations. Implementing the group creation methodology helps obtain a statistically significant increase in ideal group creations but results in a tradeoff of decreased benefits to volunteers and the NPO. Finally, implementing a minimum desired group size does not severely impact most KPIs and would be beneficial for an NPO to implement as it encourages the creation and assignment of volunteer groups to tasks.

Conductive paths generation using topology optimization for worst-case thermal design in space systems

Designing thermal systems for space platforms is a challenging task due to the wide variability in thermal conditions during the orbit and the strict constraints imposed by other subsystems. Several strategies have been developed to address these challenges while minimizing the thermal control subsystem’s signature, encompassing different algorithms optimize components positions in the platform. When relocation is not possible, thermal couplings between components can be modified to enhance heat transfer and achieve more uniform temperature distribution. To design optimal thermal paths in 2D space structures, in this paper a topology optimization framework based on the Heaviside Projection Method is introduced, using the Lumped Parameter Method (LPM) and taking the radiation heat transfer into account. The algorithm’s performance is tested on a Printed Circuit Board (PCB) by designing an optimal copper layer to meet specific thermal requirements under both extreme hot and cold conditions. Results show a significant improvement in thermal compliance for all components with the addition of this high-conductivity layer. Additionally, a reduction method is introduced to transfer the material distribution from the optimization to a coarser mesh of any size, which is useful to facilitate its implementation in existing software, where a large number of degrees of freedom can be a limitation.

Safe-Enhanced Autonomous Driving Technology Using Conformal Prediction Results

Abstract Autonomous driving technology significantly improves road safety, mitigates traffic congestion, and paves the way for a more efficient and connected transportation future. However, the uncertain driving scenarios poses a great challenge to the safe control of autonomous vehicles (AVs). Therefore, this paper proposes an enhanced safety control framework by integrating the quantified prediction results of human driving behaviors into the path-planning process. First, a Long Short-Term Memory (LSTM) network is employed for driver behavior prediction, based on which, the conformal prediction, a mathematical statistical method, is introduced to quantify the uncertainties of the prediction results with a probabilistic presentation. Then, a model predictive model controller (MPC) is designed to enable the path-planning ability. By using the Lipschitz constant, the inequality for obstacle avoidance is reformulated as a hard constraint in the MPC optimization framework while considering the conformal prediction results. Finally, some simulation test cases are conducted to validate the proposed method. The results demonstrate the effectiveness to guarantee the safety of AVs with uncertain prediction results.