Hierarchical Multi-objective Optimization Research Articles

Hierarchical multi-objective optimization of proton exchange membrane fuel cell with parameter uncertainty

It is important to improve the performance indexes including the net power and system efficiency for polymer electrolyte membrane fuel cell (PEMFC), especially considering the parameter uncertainty. To this end, this paper proposes a hierarchical multi-objective optimization framework including the off-line and on-line optimization. Firstly, a steadystate non-linear PEMFC system model is established as the base model. Secondly, an adaptive multi-objective particle swarm optimization (AMPSO) algorithm scheme is proposed in the off-line optimization procedure for the exact optimization of operation parameters. Meanwhile, an adaptive flight parameter strategy based on the particle dispersity (PD) information is proposed to balance the convergence and diversity of Pareto solutions. Finally, to decrease the influence caused by parameter uncertainty, the interval optimization method is proposed in the on-line optimization layer based on the results of AMPSO. The convergence condition of the proposed optimization scheme is verified by theory analysis. The proposed AMPSO is compared with different algorithms in the numerical simulation and hardware-in-loop (HIL) experiments. Meanwhile, the performance of the proposed method is tested on four representative benchmark problems. These results demonstrate that the PEMFC system with the proposed optimization scheme performs better than the base model and classical optimization algorithms in terms of the net power and system efficiency indexes, revealing the success of this hierarchical optimization approach in solving the accurate optimization and parameter uncertainty problems. By using statistical test methods, the proposed algorithm performs better hypervolume (HV) metric from the Pareto solution distribution.

Hierarchical Multi-Objective Optimization of a Multi-Sided Permanent Magnet Linear Synchronous Motor With Ring Structure Winding for Conveyor Systems

Hierarchical Multi-Objective Optimization of a Multi-Sided Permanent Magnet Linear Synchronous Motor With Ring Structure Winding for Conveyor Systems

Hierarchical Multi-Objective Optimization for Dedicated Bus Punctuality and Supply-Demand Balance Control.

Public transportation is a crucial component of urban transportation systems, and improving passenger sharing rates can help alleviate traffic congestion. To enhance the punctuality and supply-demand balance of dedicated buses, we propose a hierarchical multi-objective optimization model to optimize bus guidance speeds and bus operation schedules. Firstly, we present an intelligent decision-making method for bus driving speed based on the mathematical description of bus operation states and the application of the Lagrange multiplier method, which improves the overall punctuality rate of the bus line. Secondly, we propose an optimization method for bus operation schedules that respond to passenger needs by optimizing departure time intervals and station schedules for supply-demand balance. The experiments were conducted in Future Science City, Beijing, China. The results show that the bus line's punctuality rate has increased to 90.53%, while the retention rate for platform passengers and the intersection stop rate have decreased by 36.22% and 60.93%, respectively. These findings verify the effectiveness and practicality of the proposed hierarchical multi-objective optimization model.

Hierarchical multi-objective design of regional integrated energy system under heterogeneous low-carbon policies

Fossil-to-renewable energy system transition is an inevitable future trend considering carbon dioxide emissions mitigation burdens. However, huge investment costs and insufficient policy support are still the two main barriers to further expand renewable energy penetration. To transcend these barriers, from the perspective of both policy makers and energy suppliers, this study simultaneously integrates low-carbon policy formulations and regional integrated energy system (IES) reshaping to achieve an affordable decarbonization pathways. Mixed Integer Quadratically-Constrained Programs (MIQCP) models are constructed, and hierarchical multi-objective optimization is applied to quantify local government's decision-making costs and energy suppliers' system transition costs. A new rural community in Dalian (China) is taken as a case study to capture the relationship among energy structure adjustment, energy system reshaping, and renewable subsidy and carbon tax adjustment under different CO2 emission reduction regulations. The result shows that under the same CO2 emission reduction rate, portfolio policies can benefit both the policy maker and the energy supplier by balancing the local government revenue and expenditure, and reducing the annual total costs of IES. In addition, via IESs optimization, a higher renewable energy penetration rate could be obtained and the energy supplier could achieve a competitive cost within 40% CO2 emission reduction targets.

High-Throughput Screening of Optimal Process Parameters for PVD TiN Coatings With Best Properties Through a Combination of 3-D Quantitative Phase-Field Simulation and Hierarchical Multi-Objective Optimization Strategy

Physical vapor deposition (PVD) is one of the most important techniques for coating fabrication. With the traditional trial-and-error approach, it is labor-intensive and challenging to determine the optimal process parameters for PVD coatings with best properties. A combination of three-dimensional (3-D) quantitative phase–field simulation and a hierarchical multi-objective optimization strategy was, therefore, developed to perform high-throughput screening of the optimal process parameters for PVD coatings and successfully applied to technically important TiN coatings. Large amounts of 3-D phase-field simulations of TiN coating growth during the PVD process were first carried out to acquire the parametric relation among the model parameters, microstructures, and various coating properties. Experimental data were then used to validate the numerical simulation results and reveal the correlation between model parameters and process parameters. After that, a hierarchical multi-objective method was proposed for the design of multiple coating properties based on the quantitative phase–field simulations and key experimental data. Marginal utility was subsequently examined based on the identification of the Pareto fronts in terms of various combinations of objectives. The windows for the best TiN coating properties were, therefore, filtered with respect to the model/process parameters in a hierarchical manner. Finally, the consistent optimal design result was found against the experimental results.

Hierarchical Multi-Objective Optimization of Automobile Seat Frame Based on Grey Fuzzy Logic System

In development of the automobile industry, lightweight and safety performance are contradictory, yet they are of great significance.The optimization process of auto parts is a high-dimensional optimization problem which has a variety of regulations limits and safety tests at the same time. In order to address this issue, hierarchical multi-objective optimization of a passenger car seat frame is carried out in this research. Different from previous researches, in this paper, all car seat frame parts are listed as the optimization objects and are given different optimized attributes. Meanwhile, in order to achieve the goal of effectively reducing the sample sizes of the design of experiments, hierarchical optimization is proposed, and the optimization process is divided into three stages. In each stage, components with different optimized attributes are introduced into various safety tests and then conducting design of experiments. On the other hand, through adopting the grey fuzzy logic system to assign the appropriate optimized grade, the optimization process is simplified and the errors caused by the manual selection or unified optimization levels are avoided. The proposed method is considered to be an universal approach to solve the lightweight optimization of the auto parts. Design parameters of the car seat frame before and after the hierarchical multi-objective optimization are compared, it illustrated that total mass and material cost of the seat frame are reduced by 2.3kg (28.5%) and ¥ 13.8 (32.4%) respectively. Moreover, various comparisons are carried out to verify the validity of the optimization method proposed in this paper. In conclusion, the proposed method is quite promising yet with less sample points associated, is an effective mean to solve the multi-objective optimization problems of automobile component.

A Novel Fault Identification Method Driven by Knowledge and Data

In the field of intelligent manufacturing, fault identification is an effective way to improve product service by identifying the cause of failures. For addressing it, the Generalized Bayesian Network (GBN) model is extended based on the traditional Bayesian Network in this paper, which redefines the directed edges and probability parameters among nodes. Compared with Bayesian Network, the GBN model has the ability to simultaneously define causality and correlation of variables. In addition, the structure of network is not only based on statistical data but also driven by expert knowledge. In order to achieve the collaboration of data and knowledge while maintaining the consistency, a hierarchical collaborative framework is designed including the data layer and knowledge layer. Furthermore, a hierarchical multi-objective optimization algorithm, namely Hierarchical Non-dominated Sorting Genetic Algorithm II (HNSGA-II), is advanced to solve the proposed model. Finally, an industrial case study for fault cause identification targeting the product service helps illustrate all details.

Two-Stage Synthetic Optimization of Supercapacitor-Based Energy Storage Systems, Traction Power Parameters and Train Operation in Urban Rail Transit

The stationary supercapacitor energy storage system (SCESS) is one of effective approaches for the utilization of train's regenerative braking energy in urban rail systems. In this paper, the capacity configuration of SCESSs, the no-load voltage of substation, the control of onboard braking resistors and train operation diagrams are considered comprehensively. Based on the equivalent circuit model, the effects of traction power system parameters on the energy transmission between powering trains, braking trains and SCESSs are analyzed, and the relationships between the train operating parameters and system energy consumptions are revealed. A multi-variable synthetic optimization method is proposed to optimize the SCESS capacity, train operation diagrams and traction power system parameters collaboratively. In order to reduce the search space and improve the efficiency of the optimization algorithm, a hierarchical multi-objective optimization model is established, based on which the design variables and the control variables are optimized iteratively. And as the traffic density has a great impact on the regenerative braking energy distribution of the system, the optimization objective function fully considers the frequency distribution of headways throughout the day. Combined the Elitist Non-dominated Sorting Genetic Algorithm (NSGA-II) with the traction power flow simulator, the flowchart of the two-stage optimization algorithm is designed, and the pareto set of the multi-objective problem is obtained. The advantages of reducing system energy consumptions and configuration cost under the proposed algorithm are verified based on case studies of Beijing Batong Line.

UAV Task Allocation for Hierarchical Multiobjective Optimization in Complex Conditions Using Modified NSGA-III with Segmented Encoding

With the recent boom in unmanned aerial vehicle (UAV) technology, many UAV applications involving complex and risky tasks in military and civilian fields have emerged, such as military strikes and disaster monitoring. Task allocation for UAVs is the process of planning the division of work among UAVs, controlled from ground stations by human operators. This study formulates the UAV task-allocation problem as an extended traveling salesman problem and presents a novel UAV task-allocation model for complex air concentration monitoring tasks. Then, an optimized non-dominated sorting genetic algorithm III (NSGA-III) based on a twin-exclusion mechanism, hierarchical objective-domination operator, and segmented gene encoding (i.e., NSGA-III-TEHOD) is developed to solve complex task-allocation problems involving multiple UAVs, hierarchical objectives, obstacles, and ambient wind. The algorithm is tested in several simulations, and the results demonstrate that the new algorithm outperforms NSGA-III, non-dominated sorting genetic algorithm II (NSGA-II), and genetic algorithm (GA) in terms of efficiency of global convergence and early maturation prevention and is available for the hierarchical objective-optimization problems.

Behavioral Portfolio Management with Layered ESG Goals and Ai Estimation of Asset Returns

Portfolio managers and individual investors alike are in quest of efficient asset allocation models that simultaneously express environmental, social, and governance (ESG) considerations along with investor behavioral biases. The current study presents a novel approach to optimize the behavioral portfolio management model in the presence of investor biases for ESG sustainability, loss aversion, and cognitive dissonance. We extend the factor pricing literature by implementing a factor extraction protocol to identify three unique and pervasive ESG factors. Upon examining the interconnectedness of the factors, machine learning methods are applied to a production-theoretic six-factor Fama and French model to predict individual asset returns. Enumeration of efficient asset allocations is obtained by solving a hierarchical multiobjective portfolio optimization model. simulation results from solving alternate specifications of the layered goal ESG-driven model corroborate and extend emergent research on portfolio sustainability, network theory, and the interconnectedness of financial returns. Additionally, we provide results to amplify the existence of a hump-shaped ESG efficiency frontier. The results provide new information about the trade-offs available for resolving cognitive dissonance when investors are conflicted between holding ‘green’ versus ‘brown’ asset diversification plans.

Multi-objective optimization of traffic signals based on vehicle trajectory data at isolated intersections

Existing fixed-time traffic signal optimization methods mainly use traffic volumes collected by infrastructure-based detectors (e.g., loop detectors). These infrastructure-based detectors generally have high maintenance costs and low coverage. With the deployment of probe vehicles, vehicle trajectory data provide more information about traffic states and can be utilized for signal timing. However, most related studies assume high penetration rates of probe vehicles or short sampling intervals. This paper develops a hierarchical multi-objective optimization framework to optimize fixed-time traffic signals based on sampled vehicle trajectories at isolated signalized intersections, which is applicable to low-resolution trajectory data. Cycle length and green splits are optimized under both under- and slightly over-saturated traffic conditions. The number of over-saturated phases and average vehicle delays are adopted as the primary and the secondary objectives, respectively. Note that the queues of over-saturated phases cannot be discharged. The queue length and traffic delay of over-saturated phases will go infinite as time goes. The consideration of the over-saturated phase number helps increase vehicle throughput and reduce queue length and traffic delay. The aggregation of sampled trajectory data during the same period across multiple cycles and Same-ratio Principles (SRPs) are proposed to compensate for the limitations of low penetration rates of probe vehicles. The evolution of sampled trajectories with varying signal timings are formulated explicitly. A sampled-trajectory-density method is proposed to identify over-saturated phases. Then a mixed integer non-linear programming (MINLP) model is formulated and transformed to a series of mixed integer linear programming (MILP) models by linearization and enumerating cycle lengths. Simulation studies validate the advantages of the proposed model over the one in Synchro Studio. Sensitivity analysis shows that: (1) the proposed model is applicable to Poisson vehicle arrivals; (2) the proposed model can handle sampling intervals as long as 15 s when sufficient sampled vehicle trajectories are collected; (3) the number of sampled trajectories has impacts on the performance of the proposed model instead of probe vehicle penetration rates, especially with under-saturated traffic; (4) cycle lengths of initial signal timing plans have no noticeable impacts on the required number of sampled trajectories when a short sampling interval is applied; and (5) the proposed model is insensitive to the quality of initial signal timing plans with under- and slightly over-saturated traffic. The proposed model is also implemented with field data to demonstrate its applicability in the real world.

Bayesian hierarchical multi-objective optimization for vehicle parking route discovery

Discovering an optimal route to the most feasible parking lot has been a matter of apprehension for any driver. Selecting the most optimal route to the most favorable lot aggravates further during peak hours of the day and at congested places. This leads to a considerable wastage of resources specifically time and fuel. This work proposes a Bayesian hierarchical technique for obtaining this optimal route. The route selection is based on conflicting objectives, and hence, the problem belongs to the domain of multi-objective optimization. A probabilistic data-driven method has been used to overcome the inherent problem of weight selection in the popular weighted sum technique. The weights of these conflicting objectives have been refined using a Bayesian hierarchical model based on multinomial and Dirichlet prior. Genetic algorithm has been used to obtain optimal solutions. Simulated data have been used to obtain routes which are in close agreement with real-life situations. Statistical analyses have shown the superiority of the weights obtained using the proposed algorithm based on Bayesian technique over the existing frequentist technique.

Post-earthquake medical evacuation system design based on hierarchical multi-objective optimization model: An earthquake case study

Post-earthquake medical evacuation (MEDEVAC) is a significant public health issue, and the transfer of injured people for medical care is time-sensitive. A well-designed MEDEVAC program, which deploys the emergency medical facilities and plans the evacuation routes, can reduce transport time and improve survival rates. In this paper, we develop models to guide the deployment of the MEDEVAC system with three echelons, i.e., the injured, first-line medical centers, and second-line hospitals. We generalize existing location models and formulate multi-objective coverage models by minimizing the relief budget and the total weighted distance. We use the multi-criteria decision approach to estimate the medical demand that is used for optimization and model evaluation. Multi-hazard risks and existing medical resources are incorporated to select a preferred solution. Using data from Wenchuan, China, results suggest that improvements in survival and cost-effectiveness are possible with optimization and sensitivity analysis highlights the robustness of our approach in seismic intensity and distribution of the injured.

Applied Research of Hierarchical Multi-objective Optimization Method in High Speed and High Precision Placement Machine

In practical engineering, placement optimization of multi-head in the high speed and high precision Surface Mount Technology (SMT) is a multi-objective optimization problem. It mainly includes the feeder distribution optimization, the component placement sequence optimization and the nozzle configuration optimization. Optimization problems are complicated and mutual coupling between them. In this paper, the hierarchical multi-objective optimization model is developed. In the hierarchical structure, the nozzle optimization is constructed as a 0-1 integer programming optimization model for the first master hierarchy. In the second hierarchical, component mounting sequence, feeder optimization and picking order are parallel optimized based on the first hierarchy optimization. The respective problems related algorithm strategies have been proposed. The engineering experiments show that this hierarchical multi-objective optimization method can significantly improve the overall performance of the multi-head placement machine with high speed and high precision.

Comprehensive Sensitivity and Cross-Factor Variance Analysis-Based Multi-Objective Design Optimization of a 3-DOF Hybrid Magnetic Bearing

Multi-degree-of-freedom (MDOF) magnetic bearings have been widely investigated and designed for various applications. However, a new design magnetic bearing cannot be directly used without optimization due to the not relatively excellent performance. Besides, there may be a lack of consideration of the interaction of parameters in the design process. Hence, in this article, a three-degree-of-freedom hybrid magnetic bearing (THMB) is optimized as an example. First, a comprehensive sensitivity analysis is carried out to show the relationship between the parameters and optimization objectives in detail. Second, a cross-factor variance analysis is considered due to the possibility of parameter interaction. And then, a hierarchical multi-objective optimization structure is used with the Kriging model and the non-dominated Sorting Genetic Algorithm II (NSGA II). The simulation results verify the validity of the proposed method, and the prototype is under manufacture for further evaluation.

Effective control allocation using hierarchical multi-objective optimization for multi-phase flight

For different flight phases in an overall flight mission, different control and allocation preferences should be pursued considering lift, drag or maneuverability characteristics. The multi-objective flight control allocation problem for a multi-phase flight mission is studied. For an overall flight mission, different flight phases namely climbing, cruise, maneuver and gliding phases are defined. Firstly, a multi-objective control allocation problem considering drag, lift or control energy preference is constructed. Secondly, considering different control preferences at different flight phases, the analytic hierarchical process method is used to construct a comprehensive performance index from different objectives such as lift or drag preferences. The active set based dynamic programming optimization method is used to solve the real-time optimization problem. For the validation, the Innovative Control Effector (ICE) tailless aircraft nonlinear model and the angular acceleration measurements based adaptive Incremental Backstepping (IBKS) are used to construct the validation platform. Finally, an overall flight mission is simulated to demonstrate the efficiency of the proposed multi-phase and multi-objective flight control allocation method. The results show that the comprehensive performance index for different phases, which are determined from the Analytic Hierarchy Process (AHP) method, can suitably satisfy the preference requirements for different flight phases.

Multidisciplinary Design of Electric Vehicles Based on Hierarchical Multi-Objective Optimization

A method for the optimal design of complex systems is developed by effectively combining multi-objective optimization and analytical target cascading techniques. The complex systems with high dimensionality are partitioned into manageable subsystems that can be optimized using dedicated algorithms. The multiple objective functions in each subsystem are treated simultaneously, and the interactions between subsystems are managed using linking variables and shared variables. The analytical target cascading algorithm ensures the convergence of the optimal solution that meets the system level targets while complying with the subsystem level constraints. A design optimization of electric vehicles with in-wheel motors is formulated as a two-level hierarchical scheme where the top level has a model representing the electric vehicle and the bottom level contains models of battery and suspension. The vehicle model includes an electric motor model and a power electronics model. Pareto-optimal solutions are derived holistically. The effectiveness of the proposed method for optimizing the complex systems is compared against the conventional all-in-one optimization approach.

Hierarchical Multiobjective H-Infinity Robust Control Design for Wireless Power Transfer System Using Genetic Algorithm

In inductive wireless power transfer (IPT) system, parameter perturbation and disturbances not only cause power fluctuation but also worsen system performances. Thus, robust control methods are introduced for improving the performances of uncertain IPT system. In robust control design, robust performance is often considered first. However, the enhancement of robust performance may lead to inferior time-domain performance. To deal with this contradiction, the weighting functions in the robust control design structure shall be chosen wisely. Considering the complex mutual dependence of weighting parameters, the selection of weighting parameters will become a complicated and practically time-consuming problem. Therefore, this brief presents a hierarchical multiobjective optimization method to select weighting parameters of H-infinity controller automatically. The inner level designs the H-infinity controller and guarantees robust performance. The outer level improves the system’s time-domain performance and robust performance at the same time with a specially designed objective function using genetic algorithm. An experimental IPT prototype is established for the verification of the proposed optimization method. The optimization results show both good robust performance and fast time response about 5–8ms in different experiment conditions.

Current-flow and current-density-aware multi-objective optimization of analog IC placement

In this paper, the concept of hierarchical multi-objective optimization is applied to analog integrated circuit placement automation, where current-flow and current-density considerations are taken to improve the reliability and, reduce post-layout routing-induced parasitics of the circuit. The current-flow constraints are satisfied by forcing a monotonic routing directly in an absolute placement representation, while the impact of current-intensive interconnects is mitigated with the electromigration-aware optimization of the optimal wiring topology for all nets of the circuit. The problem׳s complexity is reduced using the hierarchy in the circuit׳s partitions, combining, bottom-up, Pareto fronts of placements that explore the tradeoffs between the design objectives. The approach is demonstrated in analog circuit structures for the United Microelectronics Corporation 130nm design process. Post-layout simulations show the importance of considering both current-flow and current-density considerations for an effective fully-automatic placement.

Ensemble-based hierarchical multi-objective production optimization of smart wells

In an earlier study, two hierarchical multi-objective methods were suggested to include short-term targets in life-cycle production optimization. However, this earlier study has two limitations: (1) the adjoint formulation is used to obtain gradient information, requiring simulator source code access and an extensive implementation effort, and (2) one of the two proposed methods relies on the Hessian matrix which is obtained by a computationally expensive method. In order to overcome the first of these limitations, we used ensemble-based optimization (EnOpt). EnOpt does not require source code access and is relatively easy to implement. To address the second limitation, we used the Broyden-Fletcher-Goldfarb-Shanno (BFGS) algorithm to obtain an approximation of the Hessian matrix. We performed experiments in which water flood was optimized in a geologically realistic multilayer sector model. The controls were inflow control valve settings at predefined time intervals. Undiscounted net present value (NPV) and highly discounted NPV were the long-term and short-term objective functions used. We obtained an increase of approximately 14 % in the secondary objective for a decrease of only 0.2–0.5 % in the primary objective. The study demonstrates that ensemble-based hierarchical multi-objective optimization can achieve results of practical value in a computationally efficient manner.