The proceedings contain 32 papers. The special focus in this conference is on optimization and Learning. The topics include: We Won’t Get Fooled Again: When Performance Metric Malfunction Affects the Landscape o...
ISBN:
(纸本)9783031340192
The proceedings contain 32 papers. The special focus in this conference is on optimization and Learning. The topics include: We Won’t Get Fooled Again: When Performance Metric Malfunction Affects the Landscape of Hyperparameter optimization Problems;condition-Based Maintenance optimization Under Large Action Space with Deep Reinforcement Learning Method;an Application of Machine Learning Tools to Predict the Number of Solutions for a Minimum Cardinality Set Covering Problem;adaptative Local Search for a Pickup and Delivery Problem applied to Large Parcel Distribution;GRAPH Reinforcement Learning for Operator Selection in the ALNS Metaheuristic;multi-objective optimization of Adhesive Bonding Process in Constrained and Noisy Settings;evaluating Surrogate Models for Robot Swarm simulations;interactive Job Scheduling with Partially Known Personnel Availabilities;multi-armed Bandit-Based Metaheuristic Operator Selection: The Pendulum Algorithm Binarization Case;diagonal Barzilai-Borwein Rules in Stochastic Gradient-Like Methods;binary Black Widow with Hill Climbing Algorithm for Feature Selection;optimization of Fuzzy C-Means with Alternating Direction Method of Multipliers;partial K-Means with M Outliers: Mathematical Programs and Complexity Results;An optimization Approach for Optimizing PRIM’s Randomly Generated Rules Using the Genetic Algorithm;A Fast Methodology to Find Decisively Strong Association Rules (DSR) by Mining Datasets of Security Records;characterization and Categorization of Software Programs on X86 Architectures;robot-Assisted Delivery Problems and Their Exact Solutions;modeling and Analysis of Organizational Network Analysis Graphs Based on Employee Data;time Series Forecasting for Parking Occupancy: Case Study of Malaga and Birmingham Cities;e-scooters Routes Potential: Open Data Analysis in Current Infrastructure. Malaga Case;algorithm Selection for Large-Scale Multi-objective optimization;automatic Generation of Subtitles for Videos of the Governm
Optimizing a modeling tool such that it becomes capable to perform optimization is not a trivial task. An example where this was achieved successfully is the DLR Rover simulation Toolkit RST, a set of libraries implem...
详细信息
ISBN:
(纸本)9798350367232
Optimizing a modeling tool such that it becomes capable to perform optimization is not a trivial task. An example where this was achieved successfully is the DLR Rover simulation Toolkit RST, a set of libraries implemented in Modelica. The toolkit aims at helping engineers with elements to assemble models of planetary exploration rovers for simulation throughout all project phases. It also serves as control software of hardware prototypes and has been in use at DLR for the last years. optimization is now possible in the Modelica modeling environment or through export as executable in other software such as MATLAB. The results that now can be obtained, how RST was updated such that optimization on parameters and models can be applied, are good examples for engineers in modeling and simulation. This claim is substantiated first with a simple, academic example to verify the parameter optimization. Then, the usefulness is shown on a real-world example where the optimization capability successfully improved the accuracy of the DLR Scout rover wheel simulation model.
Applying simulation-based optimization to city-scale traffic signal optimization can be challenging due to the large search space resulting in high computational complexity. A divide-and-conquer approach can be used t...
详细信息
Applying simulation-based optimization to city-scale traffic signal optimization can be challenging due to the large search space resulting in high computational complexity. A divide-and-conquer approach can be used to partition the problem and optimized separately, which leads to faster convergence. However, the lack of coordination among the partial solutions may yield a poor-quality global solution. In this paper, we propose a new method for simulation-based optimization of traffic signal control, called spatially iterative coordination for parallel optimization (SICPO), to improve coordination among the partial solutions and reduce synchronization between the partitioned regions. The traffic scenario is simulated to obtain the interactions, which is used to spatially decompose the scenario into regions and identify interdependencies between the regions. Based on the regions, the problem is divided into subproblems which are optimized separately. To coordinate between the subproblems, the interactions between partial solutions are synchronized in two ways. First, multiple iterations of the optimization process can be executed to coordinate the partial solutions at the end of each optimization process. Second, the partial solutions can also be coordinated among the regions by synchronizing the trips across the regions. To reduce computational complexity, parallelism can be applied on two levels: each region is optimized concurrently, and each solution for a region is evaluated in parallel. We demonstrate our method on a real-world road network of Singapore, where SICPO converges to an average travel time 21.6% faster than global optimization at 62.8x shorter wall-clock time.
In this research, a one-dimensional heterogeneous model is developed to simulate the partial oxidation of methanol to formaldehyde over a molybdenum-iron catalyst in an industrial isothermal reactor at dynamic conditi...
详细信息
In this research, a one-dimensional heterogeneous model is developed to simulate the partial oxidation of methanol to formaldehyde over a molybdenum-iron catalyst in an industrial isothermal reactor at dynamic condition. The considered isothermal reactor is modelled based on the mass and energy balance equations considering catalyst deactivation. Based on the simulation results, decline in the catalyst activity from 1.0 to 0.73 decreases the rate of formaldehyde production rate from 94.9 kmol h-1 to 89.63 kmol h-1 during process run time. Subsequently, a multi-objective optimization problem is programmed to enhance formaldehyde productivity and minimize the production decline during process run time. To select the effective decision variables, a sensitivity analysis is performed based on the developed dynamic model. Then, the programmed multi-objective optimization problem is replaced with a single objective by the weighted sum method, and the problem is handled by the genetic algorithm method to determine the optimal trajectory of coolant temperature. The simulation results showed that the average formaldehyde production rate increases from 92.11 kmol h-1 to 95.22 kmol h-1 when the optimal conditions are applied to the reactor.
The proceedings contain 261 papers. The topics discussed include: MBSE-based modeling technology for aircraft assembly tooling design demand;modeling and implementation of distributed rain water storage and utilizatio...
The proceedings contain 261 papers. The topics discussed include: MBSE-based modeling technology for aircraft assembly tooling design demand;modeling and implementation of distributed rain water storage and utilization;modeling free surface elevation around tandem piers of the longitudinal bridge by computational fluid dynamics;L1-finite difference method for inverse source problem of fractional diffusion equation;EMD and singular value difference spectrum based bearing fault characteristics extraction;research on a distribution-outlier detection algorithm based on logistics distribution data;activation detection algorithms for pattern division multiple access uplink grant-free transmission;application of enhanced whale adaptive threshold noise reduction method in transformer ultrasonic 3D imaging detection;human body shape reconstruction from binary image using convolutional neural network;a pricing mechanism which implements allocation in Shannon formula of home cellular wireless communication;and research on power consumption information acquisition system based on broadband power line carrier technology.
The proceedings contain 13 papers. The special focus in this conference is on Dynamic Monitoring and optimization. The topics include: Constraint modeling for Forest Management;a Note On a Prey-Predator...
ISBN:
(纸本)9783031175572
The proceedings contain 13 papers. The special focus in this conference is on Dynamic Monitoring and optimization. The topics include: Constraint modeling for Forest Management;a Note On a Prey-Predator Model with Constant-Effort Harvesting;Discrete-Time System of an Intracellular Delayed HIV Model with CTL Immune Response;a Stochastic Capital-Labour Model with Logistic Growth Function;note On Optimal Control Problem applied to Irrigation with Sectioned Soil;on Non-linear optimization with a Perturbed Objective Function;approximation of the Solution Based on the Decoupling Transformation of Linear Time-Varying Singularly Perturbed System with Delay;optimal Cyclic Dynamic of Distributed Population Under Permanent and Impulse Harvesting;on the Growth and Oscillation of Fixed Points of Solutions of Linear Differential Equations with Meromorphic Coefficients;hesitant Fuzzy Sets Are Observers;on the Stability of Three-Time-Scale Linear Time-Invariant Singularly Perturbed Systems with State Delay.
This paper presents an innovative approach to solving complex multi-objective optimization problems through an asynchronous and distributed evolutionary game theory method. The proposed algorithm, an extension of the ...
详细信息
ISBN:
(纸本)9798331531317;9798331531300
This paper presents an innovative approach to solving complex multi-objective optimization problems through an asynchronous and distributed evolutionary game theory method. The proposed algorithm, an extension of the IMGAMO algorithm, optimizes individual criteria separately at varying computational speeds, thus significantly enhancing computational efficiency and adaptability. This unique structure enables independent criterion optimization, catering to real-world applications where different objectives demand varying computational resources. The algorithm's effectiveness is validated against traditional synchronous evolutionary multi-objective optimization algorithms, showing superior performance in handling diverse, real-world problems efficiently. The results underline the potential of the asynchronous approach in providing high-quality Pareto fronts, thus offering robust solutions for complex optimization challenges.
Urban modeling and simulation are critical for addressing urbanization challenges. Optimizing and scaling these simulations can significantly reduce simulation time and resource usage, enabling faster decision making ...
详细信息
ISBN:
(纸本)9798350385939;9798350385922
Urban modeling and simulation are critical for addressing urbanization challenges. Optimizing and scaling these simulations can significantly reduce simulation time and resource usage, enabling faster decision making and policy formulation. This paper investigates the performance optimization and scalability of the twin-building case. The study covers various optimization techniques, including compiler-level switches, MPI-specific binding flags, optimal decomposition strategies, and the utilization of vendor-specific compilers. Through extensive experimentation, a significant performance boost was achieved, increasing the performance by approximately 47 times and reducing the simulation time to 46 seconds. This was accomplished by scaling the prototype Twin Building 26.17M case across more than 1,000 CPU-only cores. These findings underscore the importance of performance optimization strategies in effectively harnessing computational resources for large-scale urban modelingsimulations. The study improves performance and serves as a basis for future research on heterogeneous computing platforms to enhance simulation time.
The proceedings contain 40 papers. The special focus in this conference is on Learning and Intelligent optimization. The topics include: Real-World Streaming Process Discovery from Low-Level Event Data;robust Neu...
ISBN:
(纸本)9783031445040
The proceedings contain 40 papers. The special focus in this conference is on Learning and Intelligent optimization. The topics include: Real-World Streaming Process Discovery from Low-Level Event Data;robust Neural Network Approach to System Identification in the High-Noise Regime;GPU for Monte Carlo Search;learning the Bias Weights for Generalized Nested Rollout Policy Adaptation;Heuristics Selection with ML in CP Optimizer;model-Based Feature Selection for Neural Networks: A Mixed-Integer Programming Approach;an Error-Based Measure for Concept Drift Detection and Characterization;predict, Tune and Optimize for Data-Driven Shift Scheduling with Uncertain Demands;on Learning When to Decompose Graphical Models;hyper-box Classification Model Using Mathematical Programming;inverse Lighting with Differentiable Physically-Based Model;repositioning Fleet Vehicles: A Learning Pipeline;bayesian Decision Trees Inspired from Evolutionary Algorithms;Towards Tackling MaxSAT by Combining Nested Monte Carlo with Local Search;relational Graph Attention-Based Deep Reinforcement Learning: An Application to Flexible Job Shop Scheduling with Sequence-Dependent Setup Times;experimental Digital Twin for Job Shops with Transportation Agents;learning to Prune Electric Vehicle Routing Problems;matheuristic Fixed Set Search applied to Electric Bus Fleet Scheduling;Class GP: Gaussian Process modeling for Heterogeneous Functions;surrogate Membership for Inferred Metrics in Fairness Evaluation;a Leak Localization Algorithm in Water Distribution Networks Using Probabilistic Leak Representation and Optimal Transport Distance;the BeMi Stardust: A Structured Ensemble of Binarized Neural Networks;discovering Explicit Scale-Up Criteria in Crisis Response with Decision Mining;job Shop Scheduling via Deep Reinforcement Learning: A Sequence to Sequence Approach;generating a Graph Colouring Heuristic with Deep Q-Learning and Graph Neural Networks;multi-task Predict-then-Optimize;Integrating Hyperparame
Multiobjective building design optimization is a challenging problem because it involves finding a set of solutions that simultaneously optimize multiple conflicting objectives. simulations-based optimization is widel...
详细信息
Multiobjective building design optimization is a challenging problem because it involves finding a set of solutions that simultaneously optimize multiple conflicting objectives. simulations-based optimization is widely used, but it is a computationally expensive process in terms of time, as it requires a large number of evaluations of the objective functions. Metamodel-based optimization is an alternative to reduce the time-consuming simulations during the optimization process. Metamodels can approximate the building simulation model with analytical expressions. However, the accuracy of metamodels depends on the number of simulations used to train the model and the sampling strategy used to select informative samples over the design space. This study proposes an efficient sequential sampling approach to fit the metamodels toward the regions of the design space where their accuracy is higher and can improve all objectives simultaneously. To demonstrate the effectiveness of this approach, it was applied to optimize the energy and investment costs of a multi-story residential building. The optimization results were compared with those obtained using a non-dominated sorted genetic algorithm II (NSGA-II). The results of this study show that the proposed method reduces the number of building energy simulations required by up to 50% while guaranteeing accurate optimization results. Fifteen energy-efficient buildings designs were proposed, with a wide range of trade-offs between energy and investment costs. This study highlights the potential of the proposed approach to achieve faster and accurate building design optimization and allowing for a larger design space, leading to more creative and innovative solutions.
暂无评论