Satisfactory Optimization Research Articles

Dynamic optimization design of thin plate structure based on surrogate model

Abstract To enhance dynamic optimization efficiency in traditional thin plate structures, surrogate models are integrated for structural dynamic size optimization, reducing dependence on high-precision finite element simulations. This paper focuses on optimizing the aluminum alloy sheet as the subject of research. The parameters of the specimen are optimized through sensitivity analysis, based on the comparison between the free mode Finite Element Analysis (FEA) and Experimental Modal Analysis (EMA) results. The modified specimen undergoes validation under a four-edge clamped boundary condition. Subsequently, random vibration analysis is conducted to determine the root mean square (RMS) value of the acceleration response. Employing surrogate model technology, four models are constructed using parametric modeling and experimental design methods. These models are then compared for their fitting effectiveness. The results indicate that the RBF model exhibits the highest accuracy in fitting each response, effectively replacing dynamic simulation. Finally, the RBF combined with the NSGA-II method, optimizes the specimen’s structural size, resulting in a 21.6% reduction in mass and an 18.60% decrease in acceleration response. These outcomes demonstrate satisfactory optimization results, ensuring accuracy while enhancing the dynamic characteristics of the specimen structure. This research offers valuable insights for related engineering applications.

A self-organized optimal scheduling approach for integrated energy systems using bottom-up modelling

In recent years, decentralized methods have received growing attentions in the field of optimal scheduling of integrated energy systems (IESs) due to their advantages in independent operation and privacy protection. However, the development of decentralized scheduling strategies is usually challenging and time-consuming in practice. To address this problem, this paper proposes a generic self-organized multi-agent-based decentralized scheduling framework for IESs. Four types of base agent models were defined firstly to characterize the individual benefits of various energy entities in IESs. They could construct customized decentralized scheduling models easily. A topology-based fully-decentralized coordination approach was further presented to resolve the conflicts of individual benefits among agents. The coordination rules could be generated in a self-organized manner according to the topological structure of target IESs. Finally, an improved fully-decentralized alternating direction method of multipliers was developed to achieve the fully-decentralized optimization. Simulation results indicate that the generated strategy can reach satisfactory optimization accuracy and computational efficiency. The relative error of optimal results with the benchmark strategy is lower than 0.02 %, and the computational time is reduced by 46.7 % compared with conventional decentralized strategies. This study provides a generic and cost-effective solution to generate customized decentralized scheduling strategies for various types of IESs.

Topology optimization for minimizing the mean compliance under thermo-mechanical loads using element-free Galerkin method

This paper presents a numerical model for addressing thermo-mechanical coupling problems in topology optimization, utilizing the element-free Galerkin (EFG) method. A multi-parameter density-based topology optimization framework is introduced to interpolate the thermal stress coefficient, encompassing thermal conductivity, thermal expansion coefficient, and elastic modulus. Two numerical examples are provided to investigate the model's characteristics and associated considerations, including the impact of EFG node distribution, quantity, calculation points layout, design variables, and filtering techniques on optimized results. Numerical findings indicates that the present model allows for adaptable adjustments in nodes number and distribution, calculation points layout, choice of design variables, and application of various filtering techniques while maintaining consistent background grids. Although these adjustments may affect convergence rates and final objective values, they can promise satisfactory optimization structures. Additionally, the study highlights the critical influence of filtering radius on optimization outcomes and the objective value, recommending a value of approximately 16∼20 calculation points.

Modelling and optimization method for energy saving of computer numerical control machine tools under operating condition

The widespread application of machine tools in industry results in very high energy consumption. In order to save energy, it is an effective means to reduce the processing power while improving the processing efficiency of machine tools. Since milling parameters directly affect machine performance, the selection of suitable milling parameters is particularly important. In this regard, this study proposes an integrated modelling and optimization method for energy saving of computer numerical control (CNC) machine tools under operating condition. Firstly, the operating condition of the CNC machine are analyzed and the processing power and efficiency mathematical models are established based on the four milling parameters (n, fz, ae, ap). Subsequently, the Pareto optimal solution is introduced to establish an optimization model integrating the processing power and time. To solve the optimization problem, an adaptive chaotic multi-objective chimp algorithm with fast convergence and satisfactory optimization is proposed considering three aspects: search space, convergence factor, and chaotic mapping. Afterwards, further degrees of freedom and sensitivity analyses were carried out on the constructed model using the Sobol method. Finally, comparative analysis and validation are conducted through simulation cases and actual processing experiments, respectively. The results show that both validation scenarios have the same conclusion. Compared with three existing algorithms, the proposed method reduces the processing power by 5.63 %, 4.65 % and 4.51 % and improves the processing efficiency by 27.88 %, 15.11 % and 15.31 %, respectively. Based on the above enhancements, the energy consumption is reduced by 58.21 %, 45.54 % and 41.45 % respectively.

Co-Design of USV Control System Based on Fuzzy Satisfactory Optimization for Automatic Target Arriving and Berthing

Co-Design of USV Control System Based on Fuzzy Satisfactory Optimization for Automatic Target Arriving and Berthing

Many-objective optimization of operation strategy for district cooling system based on the per-unit value form

The operational strategy of the district cooling system (DCS) with an ice storage system must be carefully optimized for many objectives to achieve optimal operation. However, owing to the fact that the variables of DCS are in different orders of magnitude and there is a vast gap between the variables, it is not easy to perform arithmetic. In this paper, the DCS model based on the per-unit value form is formulated to develop a many-objectives operational strategy that is solved by the linear weighting method. And a multiple-attribute decision-making method is taken to get the optimal operating strategy. The simulation results are discussed for an existing cooling system with 619-node and 614-branch to illustrate that the model based on the per-unit value form can obtain satisfactory optimization results and a significant reduction in time consumption.

Optimal Operation of Domestic and Industrial Sewage Treatment Plants Using Machine Learning Methods

Purpose: This study aims to determine the economic and technical feasibility of operating and leasing sewage treatment plants through an application that uses mathematical modeling and Machine Learning techniques for process optimization. Theoretical Framework: Efficient operation of sewage treatment plants (STPs) is crucial to ensure water quality, minimize environmental impacts, and optimize costs. This study explores how Machine Learning (ML) can model and optimize sewage treatment processes, adapting to real-time conditions. Method/Design/Approach: The BSM1 model is combined with Machine Learning techniques to create simplified metamodels, enabling optimized results and the development of an application for evaluating economic and technical outcomes. Results and Conclusion: The reduced metamodel successfully reproduced the Simulink model, achieving satisfactory optimization. Research Implications: This research benefits water quality improvement, cost reduction, sustainability, innovation, water resource management, awareness, and resilience to extreme weather events, as well as influencing informed policies. Originality/Value: Efficiency, sustainability, economy, and quality of life are core values in this research, benefiting society, the environment, and the economy.

Multi-Objective Multi-Satellite Imaging Mission Planning Algorithm for Regional Mapping Based on Deep Reinforcement Learning

Satellite imaging mission planning is used to optimize satellites to obtain target images efficiently. Many evolutionary algorithms (EAs) have been proposed for satellite mission planning. EAs typically require evolutionary parameters, such as the crossover and mutation rates. The performance of EAs is considerably affected by parameter setting. However, most parameter configuration methods of the current EAs are artificially set and lack the overall consideration of multiple parameters. Thus, parameter configuration becomes suboptimal and EAs cannot be effectively utilized. To obtain satisfactory optimization results, the EA comp ensates by extending the evolutionary generation or improving the evolutionary strategy, but it significantly increases the computational consumption. In this study, a multi-objective learning evolutionary algorithm (MOLEA) was proposed to solve the optimal configuration problem of multiple evolutionary parameters and used to solve effective imaging satellite task planning for region mapping. In the MOLEA, population state encoding provided comprehensive population information on the configuration of evolutionary parameters. The evolutionary parameters of each generation were configured autonomously through deep reinforcement learning (DRL), enabling each generation of parameters to gain the best evolutionary benefits for future evolution. Furthermore, the HV of the multi-objective evolutionary algorithm (MOEA) was used to guide reinforcement learning. The superiority of the proposed MOLEA was verified by comparing the optimization performance, stability, and running time of the MOLEA with existing multi-objective optimization algorithms by using four satellites to image two regions of Hubei and Congo (K). The experimental results showed that the optimization performance of the MOLEA was significantly improved, and better imaging satellite task planning solutions were obtained.

Taguchi optimization and analysis of variance for thermoelectric generators with forced convection air cooling

In recent years, humanity has experienced the effects of using unsustainable energy sources that pollute the environment while being inefficient. Most of the primary energy input used in energy conversion processes is lost in the form of heat. Thermoelectric generators (TEGs) have the potential to recover some of this wasted heat; however, their low energy conversion efficiency needs to be addressed. Previous literature has shown effective optimization of thermoelectric generation systems through statistical tools such as the Taguchi method that optimized all the parameters that affect the outcome of the system. To this date, no research has considered wind speed for convection heat transfer as one of the optimizable parameters. This study aims to optimize the operating conditions of commercially available TEGs that are cooled by forced convection simulating naturally occurring wind speeds and operate under low-quality waste heat temperatures to improve the TEG performance further. Two commercially available TEGs are optimized via the Taguchi method. The Taguchi method is implemented with three parameters and three levels, making an L9 orthogonal array. The objective function is the maximum output power. The optimized parameters are the hot side temperature, the heat sink size, and the wind airspeed. Analysis of variance (ANOVA) is used alongside the Taguchi orthogonal array for the statistical analysis of the results. The optimization results show that the hot side temperature is the most influential parameter to the output power. In addition, the change in the heat sink size also significantly influences the output power, whereas the impact of the wind speed is not as significant. In addition, interaction analysis between the parameters is performed. The Taguchi method yields satisfactory optimization results. The optimized system yields 0.65 W of power at 140 °C hot side temperature, Model K402 heat sink, and 3.7 m·s−1 wind speed. The influence of the air velocity of 1.73% is minimal on the output power compared to the hot side temperature, with a level of influence of 75.67% and a heat sink size of 22.43% for the shorter module. The results also show that the TEG with a smaller TE leg height yields the best output power at low load resistance ranging from 0.2 Ω to 2 Ω, as the TEG with a taller TE leg renders better performance at higher load resistance ranging from 3 Ω to 16 Ω at the optimal conditions. The results show that the hot side temperature remains the most important parameter in the optimization process. Furthermore, convection cooling from naturally occurring winds has a minimal but positive effect on the output of the presented TEG system.

Numerical Simulation and Optimization of Methane Steam Reforming to Maximize H2 Production: A Case Study

Research in renewable energy, the preservation of the environment, and the reduction of energy generation costs are themes that go hand in hand. In this work, a case study was carried out that aims to maximize the production of hydrogen through Methane Steam Reforming. For this, several numerical simulations, considering a laminar flow regime in a chemical reactor with a catalyst, were developed with COMSOL Multiphysics. After an exploratory study of the data, a systematic optimization was developed using multivariate regression models formed by combinations of input parameters in an idealized reactor. The results showed that the proposed approach is capable of satisfactory optimization.

Bi-Level Off-Policy Reinforcement Learning for Two-Timescale Volt/VAR Control in Active Distribution Networks

In Volt/Var control (VVC) of active distribution networks (ADNs), both slow timescale discrete devices (STDDs, e.g. on-load tap changers) and fast timescale continuous devices (FTCDs, e.g. distributed generators) are involved and should be coordinated in time sequence. Traditional two-timescale VVC optimizes STDDs and FTCDs based on accurate system models, but sometimes is impractical because of its unaffordable modeling effort. In this paper, a novel bi-level off-policy reinforcement learning (RL) method is proposed to solve this in a model-free manner. A Bi-level Markov decision process (BMDP) is defined and separate agents are set up for the slow and fast timescale sub-problems. For the fast timescale sub-problem, we adopt an off-policy RL method with high sample efficiency. For the slow one, we develop an off-policy multi-discrete soft actor-critic (MDSAC) algorithm to address the curse of dimensionality with various STDDs. To mitigate the non-stationary issue in the two agents’ training, we propose a multi-timescale off-policy correction (MTOPC) method by adopting the importance sampling technique. Comprehensive numerical studies not only demonstrate the proposed method can achieve stable and satisfactory optimization of both STDDs and FTCDs without any model information, but also support that the proposed method outperforms existing VVC methods involving both STDDs and FTCDs.

A hybrid XGBoost-SMOTE model for optimization of operational air quality numerical model forecasts

As a main technical tool, the air quality numerical model is widely used in the forecasts of atmospheric pollutants, and its development is of great significance to the atmospheric environment and human health. In this study, a hybrid XGBoost-SMOTE model has been developed and applied for the optimization of forecasted PM2.5 and O3 concentrations from the Chinese operational air quality forecasting model - CMA Unified Atmospheric Chemistry Environment model (CUACE), which automatically finds the optimal hyperparameters and features without human intervention. Supported by a knowledge base including the ground-observed, CUACE-forecasted pollutants and meteorological data as well as some auxiliary variables, and based on the evaluation analysis of 46 selected key national cities, it was found that the XGBoost-SMOTE model can achieve satisfactory optimization effects for the operational model, especially the significant improvement of the pollutant extreme values on high-pollution days. The results show that after optimization, the 5-day average correlation coefficient (R), mean error (ME) and root mean square error (RMSE) values can reach 0.87, 10.34 µg/m3 and 16.53 µg/m3 for PM25, and 0.89, 14.53 µg/m3 and 18.83 µg/m3 for O3, far better than those from original CUACE model and XGBoost model. Furthermore, the optimization of the spatial distribution of pollutants from the CUACE model and the impact analysis of the input features by the SHAP method were also explored. The developed hybrid model unveils a good application prospect in the field of environmental meteorology forecasts.

Optimization of nanosecond laser processing for microgroove on TC4 surface by combining response surface method and genetic algorithm

Drag reduction is one of the main problems in aircraft design and a meaningful way to save fuel and improve efficiency. Surface microgrooves can reduce the frictional resistance of aircraft walls and become one of the main ways to reduce aircraft drag. A nanosecond laser can prepare microgrooves. However, the recast layers under high laser energy during processing deteriorate the processing quality. How to improve processing quality while ensuring processing efficiency is a challenge. By replacing the target feature analysis model in the original genetic algorithm with an optimization target model based on the response surface method, this study proposes an optimization method for the nanosecond laser processing of the microgroove. This method uses the difficult-to-machine aerospace material TC4 titanium alloy as the target. The Box–Behnken design method is used to design the laser processing experiment of the microgroove. The validity and reliability of the proposed optimization method are verified. It is worth mentioning that the proposed method can obtain satisfactory optimization results with fewer parameter variables and fewer experiments, with low experimental cost and high optimization efficiency compared to genetic algorithms. Furthermore, the optimized process parameters will provide initial parameters for the high-quality preparation of large-scale microgrooves on the TC4 surface through the nanosecond pulsed laser layer-by-layer scanning processing way. And the optimization method is suitable for optimizing short-pulse and long-pulse laser processing parameters of microgrooves on metal surfaces and has practical engineering value.

Research on optimized layout of bridge sensors based on MAC stepwise addition and subtraction algorithm

The difference in the location and number of sensors induces the acquisition effect of structural state information in the bridge health monitoring system to a certain extent, which is also one of the most difficult problems restricting the development of bridge health monitoring technology. Based on the Modal Assurance Criterion, this paper took a cantilever simply supported beam as an example and adopted the Effective Independence method, Kinetic Energy method and QR decomposition method to optimize a sensor layout. Taking the number of sensors as 5, the layout and MAC matrix factors were analyzed by comparing the MAC matrix and its distribution diagrams obtained by different methods. The results showed that the stepwise subtraction method based on QR decomposition was the best. And by changing the number of sensor, it was found that the optimal number of sensor was 9. It was also found that the existing methods had the shortcomings of over-reliance on the selection of initial measuring points. In order to improve this defect, cyclic iteration of stepwise addition and subtraction was introduced, and an optimal layout method based on the stepwise addition and subtraction of the number of sensors was proposed. Through screening and optimization of this method, the average off-diagonal element in the calculation example was reduced from 0.118 to 0.006 to achieve the satisfactory optimization effect. At the same time, this method was employed to optimize the layout of a concrete continuous rigid-frame bridge with 9 and 19 sensors, and the results showed that the sensor layout was more uniform. Therefore, the proposed method provides the opportunity to obtain fair optimization results and to have a considerable engineering application value.

New approach for optimizing the consumed energy of the TDEEC protocol based on the firefly algorithm in HWSN

The energy consumption optimization in the routing protocols of wireless sensor networks has always been at the center of scientific research. In this context, an improved version of the TDEEC protocol entitled ITDDEC allowing maximization lifetime of the network by excluding the closest nodes to the base station from the election process of the cluster heads. In fact, this approach has led to a new energy balance that depends on two critical parameters: the number of excluded nodes and the number of clusters. However, the implementation of this proposition by traditional optimization methods, must take into consideration the implementing cost and its speed in each sensor node because it has low memory and modest processing capacities. In this paper, we present a comparative study between different classical optimization methods and one nature-inspired algorithm entitled Firefly. This in the goal to highlight the different indicators, such as the number of iterations, the execution time, the convergence, the precision and the memory capacity. Simulations under Matlab proves that this last optimization method offers an improvement of lifetime by 48% compared to the TDEEC protocol and 13.58% compared to the ITDEEC protocol and an increase in the number of packets received of 178% compared to the TDEEC protocol and 3% compared to the ITDEEC protocol. The proposed solution is suitable for application because it offers satisfactory optimization enhancement of the total network energy consumption by recovering the energy lost in the node's calculation.

Elastic parameters characterization of multilayered structures by air-coupled ultrasonic transmission and genetic algorithm

This paper describes a non-contact method to characterize isotropic and anisotropic planar multilayer structures using a genetic algorithm. The method is based on the determination of critical angles, where the maxima of the modulus of transmission coefficient of the structure appear, and which correspond to the generation of guided waves. The optimization process minimizes the error between the reference critical angles and associated amplitudes of the transmission coefficient, with the corresponding estimated ones. The estimation of elastic parameters is demonstrated for acrylic and oak plates as well as for a bi-layered structure composed of oak and a thin layer of gesso. It is shown that to obtain satisfactory optimization results, it is necessary for guided modes of higher order than the zero ones to be taken into account. Results also show that some elastic constants such as C33 and C55 retrieved from the transmission coefficient are very sensitive to the optimization.

Finite Control Set Model Predictive Direct Power Control of Single-Phase Three-Level PWM Rectifier Based on Satisfactory Optimization

In this paper, a finite control set model predictive direct power control (FCS-MPDPC) method based on satisfactory optimization, for single-phase three-level PWM rectifier, is proposed to achieve global optimization that takes into account all its objectives and constraints. Satisfaction optimization aims to obtain the satisfactory solution after coordination of multiple goals, instead of the optimal solution of a single goal. By replacing “optimal” with “satisfaction”, more control degrees of freedom are acquired, so that low-priority auxiliary control objectives can participate in the optimization process. Meanwhile, in order to enhance the description accuracy of FCS-MPDPC method for the future trend of PWM rectifier, the prediction time domain is extended from the traditional single-step to multi-step, and the average switching frequency model is established to realize the low switching frequency control. The satisfactory optimized FCS-MPDPC method proposed in this paper is compared, through simulation and experimental results, with the standard FCS-MPDPC method, which verifies the effectiveness and superiority of the proposed algorithm.

NAAM‐MOEA/D‐Based Multitarget Firepower Resource Allocation Optimization in Edge Computing

In the edge environment, the multiobjective evolutionary algorithm based on decomposition (MOEA/D) has been widely used in the research of multitarget firepower resource allocation. However, as the MOEA/D algorithm uses a fixed neighborhood update mechanism, it is impossible to rationally allocate computing resources based on the difficulty of each subproblem optimization, which results in some problems such as reduced population evolution efficiency and poor evolution quality during the calculation process. In order to solve these problems, a decision mechanism for subproblems and population evolution stages is designed, and on this basis, a MOEA/D algorithm based on the neighborhood adaptive adjustment mechanism is proposed to adapt to the edge environment. The optimization model of multiobjective firepower resource allocation based on the maximization of damage effect and the minimization of strike cost is constructed and solved. Using the ZDT series of test functions for comparative experiments, the simulation results show that the proposed algorithm can balance the distribution and convergence of population evolution and obtain satisfactory optimization results.

Optimization of Transfer Layout and Organization on Urban Rail Transit —— A Case Study on Split Scheme of Beijing Rail Transit Line 13

Based on the practical case “the Consultation on Split Scheme of Beijing Rail Transit Line 13”, this paper collects and forecasts the passenger flow information, to combine the situation of the line with the surrounding land use, to probe into those possible problems of the existing split scheme deeply. By adopting digital construction technology, three sets of transfer layout optimization schemes are studied and designed, in order to realize the seamless connection between 13A Line and 13B Line after split. Then, this paper is supposed to establish the index evaluation system, and the satisfactory optimization scheme is obtained by the comprehensive comparison and selection from the aspects of convenience for passengers, feasibility of scheme implementation and so on. AnyLogic software is used to build a simulation model of human-vehicle interaction, which realizes the effect display and dynamic index acquisition. And the optimization effect is analyzed from the perspective of transfer platform security.

Fatigue Life Assessment of the Shell Structure of Purified Terephthalic Acid Filter Press.

The filter press is one of the most important devices in purified terephthalic acid (PTA) refinement, and it is of great significance to ensure the fatigue strength of the structure in operation. In this study, the fatigue life prediction of the shell structure of the PTA filter press was investigated through numerical and experimental methods. Firstly, the accurate stress at the critical area of the stiffener was obtained based on the thermomechanical model and submodel approach proposed. Subsequently, the fatigue life was evaluated by the fatigue strength reduction method and hot spot stress method. Finally, the shell structure is optimized by increasing the size of the axial stiffener and continuous hoop stiffener. The results unveil that both thermal load and outer structure constraints have little effect on the radial displacement and stress amplitude of the shell structure. Through modifying the fatigue design curve of the fatigue strength reduction method, the shell structure of the PTA filter press has 42.0% and 0.3% failure probabilities in the previous and present cyclic pressure conditions. Furthermore, the hoop stiffener plays an important role in reducing radial displacement and stress amplitude, among which three hoop stiffeners exhibit the most satisfactory optimization.