Multi-Objective Surrogate-Assisted Stochastic Optimization for Engine Calibration

2021 ◽  
Author(s):  
Anuj Pal ◽  
Yan Wang ◽  
Ling Zhu ◽  
Guoming George Zhu
Keyword(s):  
Optimization Problems ◽  
Fuel Efficiency ◽  
Optimization Approach ◽  
Engine Fuel ◽  
Practical Applications ◽  
Engine Calibration ◽  
Constrained Problems ◽  
Calibration Problem ◽  
Uniform Noise ◽  
Measurement Noises

Abstract A surrogate assisted optimization approach is an attractive way to reduce the total computational budget for obtaining optimal solutions. This makes it special for its application to practical optimization problems requiring a large number of expensive evaluations. Unfortunately, all practical applications are affected by measurement noises, and not much work has been done to address the issue of handling stochastic problems with multiple objectives and constraints. This work tries to bridge the gap by demonstrating three different frameworks for performing surrogate assisted optimization on multiobjective constrained problems with stochastic measurements. To make the algorithms applicable to real-world problems, heteroscedastic (non-uniform) noise is considered for all frameworks. The proposed algorithms are first validated on several multiobjective numerical problems (unconstrained and constrained) to verify their effectiveness, and then applied to the diesel engine calibration problem, which is expensive to perform and has measurement noises. A GT-SUITE model is used to perform the engine calibration study. Three control parameters, namely variable geometry turbocharger vane position, exhaust-gas-recirculating valve position, and the start of injection, are calibrated to obtain the trade-off between engine fuel efficiency performance (brake specific fuel consumption) and NOx emissions within the constrained design space. The results show that all three proposed extensions can handle the problems well with different measurement noise levels at a reduced evaluation budget. For the engine calibration problem, a good approximation of the optimal region is observed with more than 80\% reduction in evaluation budget for all the proposed methodologies.

scholarly journals Non-smooth Optimization over Stiefel Manifolds with Applications to Dimensionality Reduction and Graph Clustering

2019 ◽  
Author(s):  
Fariba Zohrizadeh ◽  
Mohsen Kheirandishfard ◽  
Farhad Kamangar ◽  
Ramtin Madani
Keyword(s):  
Optimization Problems ◽  
Computationally Efficient ◽  
Practical Applications ◽  
Stiefel Manifolds ◽  
Convex Optimization Problems ◽  
Orthogonality Constraints ◽  
Constrained Problems ◽  
Convex Problem ◽  
Smooth Optimization ◽  
Penalization Technique

This paper is concerned with the class of non-convex optimization problems with orthogonality constraints. We develop computationally efficient relaxations that transform non-convex orthogonality constrained problems into polynomial-time solvable surrogates. A novel penalization technique is used to enforce feasibility and derive certain conditions under which the constraints of the original non-convex problem are guaranteed to be satisfied. Moreover, we extend our approach to a feasibility-preserving sequential scheme that solves penalized relaxation to obtain near-globally optimal points. Experimental results on synthetic and real datasets demonstrate the effectiveness of the proposed approach on two practical applications in machine learning.

scholarly journals JuPOETs: A Constrained Multiobjective Optimization Approach to Estimate Biochemical Model Ensembles in the Julia Programming Language

2016 ◽  
Author(s):  
David Bassen ◽  
Michael Vilkhovoy ◽  
Mason Minot ◽  
Jonathan T Butcher ◽  
Jeffrey D. Varner
Keyword(s):  
Programming Language ◽  
Optimization Problems ◽  
Training Data ◽  
Optimization Approach ◽  
Proof Of Concept ◽  
Test Functions ◽  
Biochemical Model ◽  
Constrained Problems ◽  
Optimal Ensemble ◽  
Model Ensembles

AbstractEnsemble modeling is a well established approach for obtaining robust predictions and for simulating course grained population behavior in deterministic mathematical models. In this study, we present a multiobjective based technique to estimate model ensembles, the Pareto Optimal Ensemble Technique in the Julia programming language (JuPOETs). JuPOETs integrates simulated annealing with Pareto optimality to estimate parameter ensembles on or near the optimal tradeoff surface between competing training objectives. We demonstrated JuPOETs on a suite of multiobjective problems, including test functions with parameter bounds and system constraints as well as for the identification of a proof-of-concept biochemical model with four conflicting training objectives. JuPOETs identified optimal or near optimal solutions approximately six-fold faster than a corresponding implementation in Octave for the suite of test functions. For the proof-of-concept biochemical model, JuPOETs produced an ensemble of parameters that gave both the mean of the training data for conflicting data sets, while simultaneously estimating parameter sets that performed well on each of the individual objective functions. JuPOETs can be adapted to solve many problem types, including mixed binary and continuous variable types, bilevel optimization problems and constrained problems without altering the base algorithm. JuPOETs can be installed using the Julia package manager from the JuPOETs GitHub repository at https://github.com/varnerlab/POETs.jl.

A microbial treatment population dynamics optimization approach

2021 ◽  
pp. 1-15
Author(s):  
Jinding Gao
Keyword(s):  
Population Dynamics ◽  
Information Exchange ◽  
Information Diffusion ◽  
Optimization Problems ◽  
Microbial Control ◽  
Function Optimization ◽  
Optimization Approach ◽  
Dynamics Model ◽  
Population Dynamics Model ◽  
Number Of Individuals

In order to solve some function optimization problems, Population Dynamics Optimization Algorithm under Microbial Control in Contaminated Environment (PDO-MCCE) is proposed by adopting a population dynamics model with microbial treatment in a polluted environment. In this algorithm, individuals are automatically divided into normal populations and mutant populations. The number of individuals in each category is automatically calculated and adjusted according to the population dynamics model, it solves the problem of artificially determining the number of individuals. There are 7 operators in the algorithm, they realize the information exchange between individuals the information exchange within and between populations, the information diffusion of strong individuals and the transmission of environmental information are realized to individuals, the number of individuals are increased or decreased to ensure that the algorithm has global convergence. The periodic increase of the number of individuals in the mutant population can greatly increase the probability of the search jumping out of the local optimal solution trap. In the iterative calculation, the algorithm only deals with 3/500∼1/10 of the number of individual features at a time, the time complexity is reduced greatly. In order to assess the scalability, efficiency and robustness of the proposed algorithm, the experiments have been carried out on realistic, synthetic and random benchmarks with different dimensions. The test case shows that the PDO-MCCE algorithm has better performance and is suitable for solving some optimization problems with higher dimensions.

scholarly journals A LEXICOGRAPHIC OPTIMIZATION APPROACH FOR BERTH SCHEDULE RECOVERY PROBLEM IN CONTAINER TERMINALS

Transport ◽  
2016 ◽  
Vol 31 (1) ◽  
pp. 76-83 ◽  
Author(s):  
Qian Zhang ◽  
Qingcheng Zeng ◽  
Hualong Yang
Keyword(s):  
Container Terminals ◽  
Optimization Approach ◽  
Technical Problems ◽  
Trade Off ◽  
Practical Applications ◽  
Lexicographic Optimization ◽  
Multi Stage ◽  
Schedule Recovery ◽  
Equipment Failures ◽  
Recovery Problem

In container terminals, the planned berth schedules often have to be revised because of disruptions caused by severe weather, equipment failures, technical problems and other unforeseen events. In this paper, the problem of berth schedule recovery is addressed to reduce the influences caused by disruptions. A multi-objective, multi-stage model is developed considering the characteristics of different customers and the trade-off of all parties involved. An approach based on the lexicographic optimization is designed to solve the model. Numerical experiments are provided to illustrate the validity of the proposed Model A and algorithms. Results indicate that the designed Model A and algorithm can tackle the berth plan recovery problem efficiently because the beneficial trade-off among all parties involved are considered. In addition, it is more flexible and feasible with the aspect of practical applications considering that the objective order can be adjusted by decision makers.

An Efficient Chaotic Salp Swarm Optimization Approach Based on Ensemble Algorithm for Solving Class Imbalance Problem

2021 ◽  
Author(s):  
Rekha G ◽  
Krishna Reddy V ◽  
chandrashekar jatoth ◽  
Ugo Fiore
Keyword(s):  
Feature Selection ◽  
Optimization Algorithm ◽  
Optimization Problems ◽  
Fitness Function ◽  
Class Imbalance ◽  
Optimization Approach ◽  
Class Imbalance Problem ◽  
Swarm Optimization ◽  
Adaboost Algorithm ◽  
Wide Range

Abstract Class imbalance problems have attracted the research community but a few works have focused on feature selection with imbalanced datasets. To handle class imbalance problems, we developed a novel fitness function for feature selection using the chaotic salp swarm optimization algorithm, an efficient meta-heuristic optimization algorithm that has been successfully used in a wide range of optimization problems. This paper proposes an Adaboost algorithm with chaotic salp swarm optimization. The most discriminating features are selected using salp swarm optimization and Adaboost classifiers are thereafter trained on the features selected. Experiments show the ability of the proposed technique to find the optimal features with performance maximization of Adaboost.

scholarly journals Improved Quantum Particle Swarm Optimization for Mangroves Classification

2016 ◽  
Vol 2016 ◽  
pp. 1-8
Author(s):  
Zhehuang Huang
Keyword(s):  
Particle Swarm Optimization ◽  
Optimization Algorithm ◽  
Optimization Problems ◽  
Quantum Particle ◽  
Particle Swarm ◽  
Stable Convergence ◽  
Network Parameter ◽  
Swarm Optimization ◽  
Practical Applications ◽  
Quantum Particle Swarm Optimization

Quantum particle swarm optimization (QPSO) is a population based optimization algorithm inspired by social behavior of bird flocking which combines the ideas of quantum computing. For many optimization problems, traditional QPSO algorithm can produce high-quality solution within a reasonable computation time and relatively stable convergence characteristics. But QPSO algorithm also showed some unsatisfactory issues in practical applications, such as premature convergence and poor ability in global optimization. To solve these problems, an improved quantum particle swarm optimization algorithm is proposed and implemented in this paper. There are three main works in this paper. Firstly, an improved QPSO algorithm is introduced which can enhance decision making ability of the model. Secondly, we introduce synergetic neural network model to mangroves classification for the first time which can better handle fuzzy matching of remote sensing image. Finally, the improved QPSO algorithm is used to realize the optimization of network parameter. The experiments on mangroves classification showed that the improved algorithm has more powerful global exploration ability and faster convergence speed.

A Comparative Study of Global-Best Harmony Search and Bat Algorithms on Optimization Problems

2013 ◽  
Vol 464 ◽  
pp. 352-357
Author(s):  
Pasura Aungkulanon
Keyword(s):  
Optimization Problems ◽  
Search Algorithm ◽  
Harmony Search ◽  
Computation Time ◽  
Finite Sequence ◽  
Bat Algorithm ◽  
Harmony Search Algorithm ◽  
Response Surfaces ◽  
Sample Mean ◽  
Constrained Problems

The engineering optimization problems are large and complex. Effective methods for solving these problems using a finite sequence of instructions can be categorized into optimization and meta-heuristics algorithms. Meta-heuristics techniques have been proved to solve various real world problems. In this study, a comparison of two meta-heuristic techniques, namely, Global-Best Harmony Search algorithm (GHSA) and Bat algorithm (BATA), for solving constrained optimization problems was carried out. GHSA and BATA are optimization algorithms inspired by the structure of harmony improvisation search process and social behavior of bat echolocation for decision direction. These algorithms were implemented under different natures of three optimization, which are single-peak, multi-peak and curved-ridge response surfaces. Moreover, both algorithms were also applied to constrained engineering problems. The results from non-linear continuous unconstrained functions in the context of response surface methodology and constrained problems can be shown that Bat algorithm seems to be better in terms of the sample mean and variance of design points yields and computation time.

scholarly journals A Distributed Multifactorial Particle Swarm Optimization Approach

2021 ◽  
Author(s):  
Ahlem Aboud ◽  
Nizar Rokbani ◽  
Seyedali Mirjalili ◽  
Abdulrahman M. Qahtani ◽  
Omar Almutiry ◽  
...  
Keyword(s):  
Particle Swarm Optimization ◽  
Optimization Problems ◽  
Beta Function ◽  
Particle Swarm ◽  
Search Space ◽  
Optimization Approach ◽  
Multi Objective Optimization ◽  
Swarm Optimization ◽  
Multi Objective ◽  
New System

<p>Multifactorial Optimization (MFO) and Evolutionary Transfer Optimization (ETO) are new optimization challenging paradigms for which the multi-Objective Particle Swarm Optimization system (MOPSO) may be interesting despite limitations. MOPSO has been widely used in static/dynamic multi-objective optimization problems, while its potentials for multi-task optimization are not completely unveiled. This paper proposes a new Distributed Multifactorial Particle Swarm Optimization algorithm (DMFPSO) for multi-task optimization. This new system has a distributed architecture on a set of sub-swarms that are dynamically constructed based on the number of optimization tasks affected by each particle skill factor. DMFPSO is designed to deal with the issues of handling convergence and diversity concepts separately. DMFPSO uses Beta function to provide two optimized profiles with a dynamic switching behaviour. The first profile, Beta-1, is used for the exploration which aims to explore the search space toward potential solutions, while the second Beta-2 function is used for convergence enhancement. This new system is tested on 36 benchmarks provided by the CEC’2021 Evolutionary Transfer Multi-Objective Optimization Competition. Comparatives with the state-of-the-art methods are done using the Inverted General Distance (IGD) and Mean Inverted General Distance (MIGD) metrics. Based on the MSS metric, this proposal has the best results on most tested problems.</p>

scholarly journals RANGE EXTENDER ICE MULTI-PARAMETRIC MULTI-OBJECTIVE OPTIMIZATION

2021 ◽  
Vol 18 (1) ◽  
pp. 10
Author(s):  
Mikuláš Adámek ◽  
Rastislav Toman
Keyword(s):  
Simulation Model ◽  
Combustion Engine ◽  
Raw Materials ◽  
Single Point ◽  
Fuel Efficiency ◽  
Optimization Approach ◽  
Multi Objective Optimization ◽  
Electric Generator ◽  
Design And Optimization ◽  
Multi Objective

Range Extended Electric Vehicles (REEV) are still one of the suitable concepts for modern sustainable low emission vehicles. REEV is equipped with a small and lightweight unit, comprised usually of an internal combustion engine with an electric generator, and has thus the technical potential to overcome the main limitations of a pure electric vehicle – range anxiety, overall driving range, heating, and air-conditioning demands – using smaller battery: saving money, and raw materials. Even though several REx ICE concepts were designed in past, most of the available studies lack more complex design and optimization approach, not exploiting the advantageous single point operation of these engines. Resulting engine designs are usually rather conservative, not optimized for the best efficiency. This paper presents a multi-parametric and multi-objective optimization approach, that is applied on a REx ICE. Our optimization toolchain combines a parametric GT-Suite ICE simulation model, modeFRONTIER optimization software with various optimization strategies, and a parametric CAD model, that first provides some simulation model inputs, and second also serves for the final designs’ feasibility check. The chosen ICE concept is a 90 degrees V-twin engine, four-stroke, spark-ignition, naturally aspirated, port injected, OHV engine. The optimization goal is to find the thermodynamic optima for three different design scenarios of our concept – three different engine displacements – addressing the compactness requirement of a REx ICE. The optimization results show great fuel efficiency potential by applying our optimization methodology, following the general trends in increasing ICE efficiency, and power for a naturally aspirated concept.

Numerical Optimization of the Thermoelectric Cooling Devices

2006 ◽  
Vol 129 (3) ◽  
pp. 339-347 ◽  
Author(s):  
Boris Abramzon
Keyword(s):  
Electric Current ◽  
Optimization Problems ◽  
Random Search ◽  
Thermal Conductance ◽  
Numerical Approach ◽  
Physical Parameters ◽  
Optimization Approach ◽  
Independent Variables ◽  
Adaptive Random Search ◽  
Optimization Search

The present study proposes a unified numerical approach to the problem of optimum design of the thermoelectric devices for cooling electronic components. The standard mathematical model of a single-stage thermoelectric cooler (TEC) with constant material properties is employed. The model takes into account the thermal resistances from the hot and cold sides of the TEC. Values of the main physical parameters governing the TEC performance (Seebeck coefficient, electrical resistance, and thermal conductance) are derived from the manufacturer catalog data on the maximum achievable temperature difference, and the corresponding electric current and voltage. The optimization approach is illustrated with several examples for different design objective functions, variables, and constraints. The objective for the optimization search is the maximization of the total cooling rate or the performance coefficient of the cooling device. The independent variables for the optimization search are as follows: The number of the thermoelectric modules, the electric current, and the cold side temperature of the TEC. Additional independent variables in other cases include the number of thermoelectric couples and the area-to-height ratio of the thermoelectric pellet. In the present study, the optimization problems are solved numerically using the so-called multistart adaptive random search method.

Sign in / Sign up

Export Citation Format

Share Document