Design Problem Research Articles

LQR controller performance via particle swarm optimization and neural networks

AbstractThe inverted pendulum‐cart (IPC) control problem has long been a benchmark in the field of control systems due to its inherent instability and nonlinear dynamics. The linear quadratic regulator (LQR) control technique has been proven to be effective in stabilizing the inverted pendulum; however, the challenge lies in finding the optimal control gains that provide the best performance. This article presents a method to address the LQR control design problem for the IPC system which is compared against a particle swarm optimization based LQR and a neural network optimized LQR. The method provides a deterministic approach to finding the weighing matrices Q and R in accordance with the time domain characteristics chosen by the designer, such as settling time and maximum peak overshoot. Results from MATLAB simulations indicate that the suggested strategy has good performance.

Techno-economic microgrid design optimization considering fuel procurement cost and battery energy storage system lifetime analysis

The importance of microgrids (MGs) lies in their capacity to enhance energy reliability, integrate renewable resources, and bolster resilience, yet their optimal design and sizing pose intricate challenges in balancing investment and operational cost while dispatching the MG equipment efficiently. This paper presents the optimal MG design problem formulated as an integer linear program (ILP) aimed at minimizing total investment and operational cost of MG as well as optimal hourly dispatch of MG asset over the project lifetime. Furthermore, a comprehensive analysis of the fuel procurement cost and the impact of ESS technical parameters, such as depth of discharge (Dod), ESS lifetime, and battery C-rate, on the optimal MG design, ESS lifetime and relevant costs is presented in this work for a MG operating in islanded and grid-connected modes. The simulation results demonstrate that the optimal design of a MG with the maximum load demand of 300 kW operating in the grid-connected mode can save about 28 k$ equivalent to 14% in total annual cost compared with the islanded mode operation thanks to the flexibility provided by the utility purchases and efficient MG equipment sizing. This study’s unique contributions provide significant insights into the practical implications of MG design optimization, potentially influencing regulatory policies and business strategies.

Exact approaches to solve the Transmission Expansion Planning Problem with Re-design

Expanding electrical transmission networks is a critical challenge faced by many developing economies. This process requires extensive investments, which must be strategically and methodically planned on a large scale, often encompassing regional or national considerations. Differing from typical Network Design Problems, efficiency in a transmission network can sometimes be enhanced by strategically deactivating certain transmission lines. Thanks to that, this paper focuses on a version of the transmission expansion planning problem that allows redesign during the network expansion. We propose using two exact methodologies as alternatives to the direct use of mixed integer programming formulations with commercial solvers. These methods are an application of Benders’ decomposition and a newly developed approach named Ring Space. Through comprehensive computational experiments, we assess and compare the efficacy of these methods against the conventional application of the mathematical formulation. Additionally, this study successfully obtained new high-quality solutions and proved their optimality for a subset of benchmark instances.

Cold chain distribution network design with heterogeneous fleet under carbon reduction policy application effect

ABSTRACT To integrate cold chain distribution network design problem with heterogeneous fleet, a multi-objective optimisation model is proposed to simultaneously optimise distribution routes, the quantity of internal combustion and electric refrigerated trucks (ERTs), and the number and location of distribution centres and charging stations with time window constraints, where the cold chain distribution operating cost and the customer dissatisfaction cost are minimised. A hybrid solution methodology, combining a multi-objective artificial fish swarm algorithm with a label-based optimisation algorithm, is presented to address this model. Extensive computational experiments verify the effectiveness of the proposed method. Moreover, a sensitivity analysis, focusing on the range of ERTs, the purchase subsidy for ERTs, the carbon tax and the carbon emission quota, informs the development of diverse carbon reduction policies. A comparative analysis of the implementation effect of carbon reduction policies can provide optimisation decision support for promoting sustainability in the electrification of cold chain distribution.

An integrative approach to protein sequence design through multiobjective optimization.

With recent methodological advances in the field of computational protein design, in particular those based on deep learning, there is an increasing need for frameworks that allow for coherent, direct integration of different models and objective functions into the generative design process. Here we demonstrate how evolutionary multiobjective optimization techniques can be adapted to provide such an approach. With the established Non-dominated Sorting Genetic Algorithm II (NSGA-II) as the optimization framework, we use AlphaFold2 and ProteinMPNN confidence metrics to define the objective space, and a mutation operator composed of ESM-1v and ProteinMPNN to rank and then redesign the least favorable positions. Using the two-state design problem of the foldswitching protein RfaH as an in-depth case study, and PapD and calmodulin as examples of higher-dimensional design problems, we show that the evolutionary multiobjective optimization approach leads to significant reduction in the bias and variance in RfaH native sequence recovery, compared to a direct application of ProteinMPNN. We suggest that this improvement is due to three factors: (i) the use of an informative mutation operator that accelerates the sequence space exploration, (ii) the parallel, iterative design process inherent to the genetic algorithm that improves upon the ProteinMPNN autoregressive sequence decoding scheme, and (iii) the explicit approximation of the Pareto front that leads to optimal design candidates representing diverse tradeoff conditions. We anticipate this approach to be readily adaptable to different models and broadly relevant for protein design tasks with complex specifications.

A Multi-strategy Improved Grasshopper Optimization Algorithm for Solving Global Optimization and Engineering Problems

This paper presents a multi-strategy improved grasshopper optimization algorithm (MSIGOA), which aims to address the shortcomings of the grasshopper optimization algorithm (GOA), including its slow convergence, vulnerability to trapping into local optima, and low accuracy. Firstly, to improve the uniformity of the population distribution in the search space, the MSIGOA uses circle mapping for the population initialization. A nonlinear decreasing coefficient is utilized instead of an original linear decreasing coefficient to improve the local exploitation and global exploration capabilities. Then, the modified golden sine mechanism is added during the position update stage to change the single position update mode of GOA and enhance the local exploitation capability. The greedy strategy is added to greedily select the new and old positions of the individual to retain a better position and increase the speed of convergence. Finally, the quasi-reflection-based learning mechanism is utilized to construct new populations to improve population multiplicity and the capability to escape from the local optima. This paper verifies the efficacy of MSIGOA by comparing it with other advanced algorithms on six engineering design problems, CEC2017 test functions, and 12 classical benchmark functions. The experimental results show that MSIGOA performs better than the original GOA and other compared algorithms and has stronger comprehensive optimization capabilities.

A Cost-Aware Multi-Agent System for Black-Box Design Space Exploration

Abstract Effective coordination of design teams must account for the influence of costs incurred while searching for the best design solutions. This paper introduces a cost-aware multi-agent system (MAS), a theoretical model to 1) explain how individuals in a team should search, assuming that they are all rational utility-maximizing decision-makers, and 2) study the impact of cost on the search performance of both individual agents and the system. First, we develop a new multi-agent Bayesian Optimization framework accounting for information exchange among agents to support their decisions on where to sample in search. Second, we employ a reinforcement learning approach based on the multi-agent deep deterministic policy gradient for training MAS to identify where agents cannot sample due to design constraints. Third, we propose a new cost-aware stopping criterion for each agent to determine when costs outweigh potential gains in search as a criterion to stop. Our results indicate that cost has a more significant impact on MAS communication in complex design problems than in simple ones. For example, when searching in complex design spaces, some agents could initially have low-performance gains, thus stopping prematurely due to negative payoffs, even if those agents could perform better in the later stage of the search. Therefore, global-local communication becomes more critical in such situations for the entire system to converge. The proposed model can serve as a benchmark for empirical studies to quantitatively gauge how humans would rationally make design decisions in a team.

Markov fracture models of inhomogeneous media and design of layered structures synthesized by additive technology

We consider the problem of studying the fracture of a composite macrolayer synthesized by additive technology on a solid substrate, to which the boundary of the lower microlayer is rigidly bonded. The materials of the microlayers are homogeneous and isotropic. The thicknesses of microlayers and the total thickness of the macrolayer are the control parameters of synthesis. The physical and mechanical characteristics of the microlayers and their combination in the macrolayer are also controlling parameters. The layer is fractured by a crack, which may appear in the top layer (or any other layer) and move toward the substrate or free surface, perpendicular to the boundaries of the microlayers. The crack can stop at the boundaries between the layers (then it is treated as a separate state) or pass through the boundaries without stopping. Based on the theory of Markov chains, the average lifetime of a stratified layer synthesized on a substrate using additive technology and the variance of the number of cycles (lifetime) characterizing the quality of the synthesized structure are calculated. On the basis of the materials available for the synthesis of the layered coating, the problems of optimal design of the composite body with maximum average resource, minimum resource dispersion, maximum specific strength for given geometric and physical-mathematical constraints on the composite parameters are solved. To find optimal solutions, dynamic programming algorithms are used, implemented on circuits (graphs) of sequential selection of a combination of layer materials according to their synergistic properties.

DEVELOPING SPECIALIZED SOFTWARE FOR PROGRESSIVE COLLAPSE ANALYSIS OF PLANE FRAME STRUCTURES USING THE PLASTIC DAMAGE INDEX

In this study, the authors developed software named Sonar to analyze the progressive collapse of structural elements. This software is capable of assessing the level of damage to each structural component and automatically updating the structural model for subsequent analyses. The software features a user-friendly interface, intuitive graphics, easy input of parameters, and presents computational results in 2D graphics. Through the use of the developed software, damaged structures were analyzed and studied to conclude that considering the progressive collapse of elements based on the plastic damage index serves as a basis for investigating specific design problems to propose more optimal design solutions.

A rhinopithecus swarm optimization algorithm for complex optimization problem

This paper introduces a novel meta-heuristic algorithm named Rhinopithecus Swarm Optimization (RSO) to address optimization problems, particularly those involving high dimensions. The proposed algorithm is inspired by the social behaviors of different groups within the rhinopithecus swarm. RSO categorizes the swarm into mature, adolescent, and infancy individuals. Due to this division of labor, each category of individuals employs unique search methods, including vertical migration, concerted search, and mimicry. To evaluate the effectiveness of RSO, we conducted experiments using the CEC2017 test set and three constrained engineering problems. Each function in the test set was independently executed 36 times. Additionally, we used the Wilcoxon signed-rank test and the Friedman test to analyze the performance of RSO compared to eight well-known optimization algorithms: Dung Beetle Optimizer (DBO), Beluga Whale Optimization (BWO), Salp Swarm Algorithm (SSA), African Vultures Optimization Algorithm (AVOA), Whale Optimization Algorithm (WOA), Atomic Retrospective Learning Bare Bone Particle Swarm Optimization (ARBBPSO), Artificial Gorilla Troops Optimizer (GTO), and Harris Hawks Optimization (HHO). The results indicate that RSO exhibited outstanding performance on the CEC2017 test set for both 30 and 100 dimension. Moreover, RSO ranked first in both dimensions, surpassing the mean rank of the second-ranked algorithms by 7.69% and 42.85%, respectively. Across the three classical engineering design problems, RSO consistently achieves the best results. Overall, it can be concluded that RSO is particularly effective for solving high-dimensional optimization problems.

Cause marketing decisions in supply chains: cause effect and marketing effect

This paper studies a cause marketing (CM) design problem incorporating donation amount, marketing expenditure, and product pricing as decision variables in a manufacturer-retailer supply chain context. We investigate five CM decision models: centralized supply chain conducting CM (Model C), manufacturer conducting CM (Model M), retailer conducting CM (Model R), manufacturer leading CM collaboration (Model MC), and retailer leading CM collaboration (Model RC). The models incorporate not only the ‘cause effect’ of consumers' increased willingness to purchase but also the ‘marketing effect’, meaning the market-base expansion associated with marketing efforts in a CM campaign. A key insight is that CM should never be adopted in the absence of marketing effect but can be adopted in the absence of cause effect. Comparing different models reveals that there can be consistency or conflict in CM mode choice among supply chain members. It is also found that an influencing parameter in CM collaboration is the allocation proportion of marketing expenditure rather than that of donation amount. These results indicate that the marketing effect, which has long been neglected in the literature, is a driving factor for firms to conduct CM, and the impact of corresponding marketing efforts should be emphasised.

Cause marketing decisions in supply chains: cause effect and marketing effect

This paper studies a cause marketing (CM) design problem incorporating donation amount, marketing expenditure, and product pricing as decision variables in a manufacturer-retailer supply chain context. We investigate five CM decision models: centralized supply chain conducting CM (Model C), manufacturer conducting CM (Model M), retailer conducting CM (Model R), manufacturer leading CM collaboration (Model MC), and retailer leading CM collaboration (Model RC). The models incorporate not only the ‘cause effect’ of consumers' increased willingness to purchase but also the ‘marketing effect’, meaning the market-base expansion associated with marketing efforts in a CM campaign. A key insight is that CM should never be adopted in the absence of marketing effect but can be adopted in the absence of cause effect. Comparing different models reveals that there can be consistency or conflict in CM mode choice among supply chain members. It is also found that an influencing parameter in CM collaboration is the allocation proportion of marketing expenditure rather than that of donation amount. These results indicate that the marketing effect, which has long been neglected in the literature, is a driving factor for firms to conduct CM, and the impact of corresponding marketing efforts should be emphasised.

RL-based adaptive control for a class of non-affine uncertain stochastic systems with mismatched disturbances

This paper investigates the reinforcement learning (RL) adaptive tracking control design problem for a class of mismatched stochastic nonlinear systems with non-affine structure. The stochastic system studied in this paper is more generally representative due to the presence of non-affine inputs, internal uncertainties, and mismatched external disturbances. Firstly, in order to solve the non-affine structure of stochastic systems, an extended stochastic differential equation is constructed. Based on the actor–critic framework, by generating reinforcement signals, the network is stimulated to evolve more quickly towards the desired direction, while compensating internal uncertainties and induced uncertainties in the stochastic system. Furthermore, aiming at the approximation errors and external disturbances, while satisfying stochastic stability, the adaptive laws for disturbances boundary estimator are established with the usage of higher powers of tracking errors. As a result, a novel non-affine adaptive tracking controller has been proposed by integrating RL, disturbances boundary estimation, and dynamic surface method. Through the stability analysis, it is proved that all closed-loop signals are bounded in probability, and the system output converges to a small neighborhood near the desired trajectory. Numerical simulations demonstrate the effectiveness and superiority of the proposed controller.

Optimization of electric vehicle design problems using improved electric eel foraging optimization algorithm

Optimization of electric vehicle design problems using improved electric eel foraging optimization algorithm

Flavoring search algorithm with applications to engineering optimization problems and robot path planning

In this paper, a human-based meta-heuristic algorithm, the Flavoring Search Algorithm, is proposed and mathematically modeled with the aim of providing an alternative optimization method for solving practical engineering problems. Flavoring Search Algorithm is inspired by the human behavior of flavoring in everyday life, including basic flavoring, formal flavoring, and auxiliary flavoring. By introducing a unique taste factor, it not only succeeded in making the Flavoring Search Algorithm corresponds to the real flavoring process but also balanced the exploration and exploitation of the algorithm. With the help of the taste factors, Flavoring Search Algorithm performs basic flavoring (initial flavoring and random flavoring) in the exploration phase and formal flavoring and auxiliary flavoring in the exploitation phase. In addition, theoretical analysis and experiments have led to the conclusion that the taste factor can be used as an effective and practical new threshold conversion mechanism for meta-heuristic algorithms. This study also establishes a Markov model to rigorously analyze the Flavoring Search Algorithm as a globally convergent algorithm from a mathematical point of view. Through experimental and analytical comparisons with other excellent optimizers on 30 test functions, as well as on 3 real-world engineering design problems and 1 path planning problem. The results show that the Flavoring Search Algorithm generally outperforms the tested competitors in solving benchmark functions and engineering problems, validating the utility of the proposed optimizer in solving challenging real-world problems.

Artificial intelligence in optical lens design

Traditional optical design entails arduous, iterative stages that significantly rely on the intuition and experience of lens designers. Starting-point design selection has always been the major hurdle for most optical design problem, and different designers might produce different final lens designs even if using the same initial specification. Lens designers typically choose designs from existing lens databases, analyse relevant lens structures, or explore patent literature and technical publications. With increased processing capability, producing automated lens designs using Artificial Intelligence (AI) approaches is becoming a viable alternative. Therefore, it is noteworthy that a comprehensive review addressing the latest advancements in using AI for starting-point design is still lacking. Herein, we highlight the gap at the confluence of applied AI and optical lens design, by presenting a comprehensive review of the current literature with an emphasis on using various AI approaches to generate starting-point designs for refractive optical systems, discuss the limitations, and suggest a potential alternate approach for further research.

Trajectory Design of Potentially Hazardous Asteroid Exploration with Reusable Probes from Cislunar Space

This article presents a trajectory design problem concerning the exploration of potentially hazardous near-Earth asteroids (PHAs) with reusable probes from cislunar space. A total of 20 probes, making round trips departing from and returning to a service space station in a lunar distant retrograde orbit, are expected to explore as many PHAs as possible by means of close flyby within a 10-year time window. The trajectory design problem was released in the 12th edition of China’s Trajectory Optimization Competition on 20 August 2022, and a total of 10 sets of trajectory solutions were submitted. As the authors who proposed the competition problem, we present in this article the problem descriptions, trajectory analysis, and design, as well as an impressive trajectory solution in which a total of 105 PHAs are explored. It is concluded that taking advantage of reusable probes from cislunar space is a promising option to efficiently explore large numbers of PHAs.

Many-Objective Whale Optimization Algorithm for Engineering Design and Large-Scale Many-Objective Optimization Problems

In this paper, a novel Many-Objective Whale Optimization Algorithm (MaOWOA) is proposed to overcome the challenges of large-scale many-objective optimization problems (LSMOPs) encountered in diverse fields such as engineering. Existing algorithms suffer from curse of dimensionality i.e., they are unable to balance convergence with diversity in extensive decision-making scenarios. MaOWOA introduces strategies to accelerate convergence, balance convergence and diversity in solutions and enhance diversity in high-dimensional spaces. The prime contributions of this paper are—development of MaOWOA, incorporation an Information Feedback Mechanism (IFM) for rapid convergence, a Reference Point-based Selection (RPS) to balance convergence and diversity and a Niche Preservation Strategy (NPS) to improve diversity and prevent overcrowding. A comprehensive evaluation demonstrates MaOWOA superior performance over existing algorithms (MaOPSO, MOEA/DD, MaOABC, NSGA-III) across LSMOP1-LSMOP9 benchmarks and RWMaOP1-RWMaOP5 problems. Results validated using Wilcoxon rank sum tests, highlight MaOWOA excellence in key metrics such as generational distance, spread, spacing, runtime, inverse generational distance and hypervolume, outperforming in 71.8% of tested scenarios. Thus, MaOWOA represents a significant advancement in many-objective optimization, offering new avenues for addressing LSMOPs and RWMaOPs’ inherent challenges. This paper details MaOWOA development, theoretical basis and effectiveness, marking a promising direction for future research in optimization strategies amidst growing problem complexity.

Feeling The Distance: Exploring Novice Student Designers' Perceptions of the Psychological Distance and Empathic Response Towards Variations of a Sustainable Design Problem

Abstract Research suggests novice designers find it challenging to integrate social context into engineering design. This challenge could be pronounced in sustainable design because the effects of climate change are perceived to be psychologically distant, which inhibits active engagement in sustainable behavior. However, prior work assesses psychological distance as an aggregate of its various subcomponents (e.g., temporal, social, spatial, and hypothetical). Limited research has explored how these components vary based on the socio-spatial context of the problem. Additionally, empathy development could be an effective mechanism to bridge psychological distance. However, little is known about the impact of empathy-focused problem formulation on student perceptions. We studied students' perceptions of problem formulations in a 2×2 factorial between-subjects experiment. The problems varied in (1) their socio-spatial context and (2) their empathy-focus. We measured the students' perceptions of the problem formulations using: (1) perceived psychological distance and (2) perceived empathic response. We find that participants reported the problem contextualized in India to be a more immediate threat compared to the one in the USA. Moreover, we find no significant differences in the other components of psychological distance or empathic response between the problem variants. This finding suggests that participants perceive environmental issues as currently affecting people in other parts of the world, and not themselves, but only from a temporal lens. These findings call for educators to carefully consider temporal framing when using socio-spatially far sustainable design problems as a lack of consideration could create faulty perceptions.

Ant Colony Optimization for solving Directed Chinese Postman Problem

AbstractThe Chinese Postman Problem (CPP) is a well-known optimization problem involving determining the shortest route, modeling the system as an undirected graph, for delivering mail, ensuring all roads are traversed while returning to the post office. The Directed Chinese Postman Problem (DCPP) extends the Chinese Postman Problem (CPP), where the underlying graph representing the system incorporates exclusively directed edges. Similarly to CPP, this problem has plenty of applications in route optimization, interactive system analysis, and circuit design problems. However, due to the added constraint (directionality of edges), DCPP results are more challenging to solve. Although methods to solve it in literature are proposed, typically by using minimum-cost-flow algorithms, the meta-heuristics approaches proposed to deal with it are very limited. In this paper, we propose an innovative meta-heuristic approach to solve DCPP by using an ant colony optimization (ACO) algorithm, i.e., an algorithm that simulates in a simplified way the behavior of some species of ants to solve optimization problems. The efficiency of our ant colony optimization for solving the Directed Chinese Postman Problem (ACO-DCPP) is measured by comparing the ACO outcomes with the results obtained by a recursive algorithm that explores all the possible solutions. Results show that ACO-DCPP is stable and gets the global optimum frequently by using an extremely limited number of solutions explored.