Optimization in dynamic environments is a challenging but important task since many real-world optimization problems are changing over time. Evolutionary computation and swarm intelligence are good ...tools to address optimization problems in dynamic environments due to their inspiration from natural self-organized systems and biological evolution, which have always been subject to changing environments. Evolutionary optimization in dynamic environments, or evolutionary dynamic optimization (EDO), has attracted a lot of research effort during the last 20 years, and has become one of the most active research areas in the field of evolutionary computation. In this paper we carry out an in-depth survey of the state-of-the-art of academic research in the field of EDO and other meta-heuristics in four areas: benchmark problems/generators, performance measures, algorithmic approaches, and theoretical studies. The purpose is to for the first time (i) provide detailed explanations of how current approaches work; (ii) review the strengths and weaknesses of each approach; (iii) discuss the current assumptions and coverage of existing EDO research; and (iv) identify current gaps, challenges and opportunities in EDO.
•This work presents potential neural network architectures which have been adapted to be compatible with control and optimization of large-scale industrial systems.•This work considers the modeling ...of an actual coal-fired boiler using neural networks and optimizes their hyperparameters for modeling the boiler.•This work introduces training techniques from the field of computer vision and adversarial machine learning to improve flexibility and reliability when operating on industrial systems.•This work describes the balance between model fidelity and model controllability when using input gradient regularization.•This work shows that input gradient regularization improves control variable optimization via particle swarm optimization.
This study proposes using neural networks, specifically gated recurrent unit (GRU), long-short-term memory (LSTM), and transformer networks, to improve control strategies in a 450 MW coal-fired power plant. However, neural networks face issues of becoming overly dependent on just a few variables to make predictions, which negatively impacts control decisions that rely on the model to determine the value of all manipulated variables. The paper introduces regularization techniques, including noise injection and input gradient regularization, during the training phase. The work presents novel contributions in adapting neural networks to control industrial systems and applying regularization techniques from computer vision to industrial process control. Results demonstrate the effectiveness of input gradient regularization in reducing model dependence on subsets of variables, emphasizing the balance between fidelity and controllability. Further exploration is recommended, including the development of recurrent transformers, closed-loop control testing, and a sensitivity analysis on computer models to provide further insight.
Display omitted
•Co-pyrolysis of coffee grounds (CG) and polyethylene (PE) was characterized.•Its blend ratio, heating rate and temperature dependencies were quantified.•Co-pyrolysis of 70% CG and ...30% PE had the lowest activation energy and emissions.•The co-pyrolysis yielded hydrocarbons and alcohols up to 93.61%.•The best co-pyrolysis conditions were 70% CG and 30% PE and 517–1000 ℃.
Spent coffee grounds (CG) and polyethylene (PE) are the two typical types of major solid wastes. Their co-pyrolysis may be leveraged to reduce their waste streams and pollution and valorize energy and by-products. In this study, their co-pyrolysis performances, interaction effects, kinetics, and products were characterized in response to the varying temperature and blend ratio. The co-pyrolysis exhibited the two main stages of (1) the degradation of CG (180–380 °C) and (2) the depolymerization of PE and the decomposition of lignin (380–550 °C). The pyrolysis performance rose from 1.34×10−4%3·min−2·°C−3 with the mono-pyrolysis of CG to 26.32×10−4%3·min−2·°C−3 with the co-pyrolysis of 10 % CG and 90 % PE. The co-pyrolysis of 70 % CG and 30 % PE (CP73) achieved a lower activation energy than did the mono-pyrolysis of the two fuels. The products of the CG pyrolysis included a large number of alcohols, ethers, ketones, esters, and other oxygen-containing compounds, with a proportion as high as 65.01 %. The products of CP73 at 550 °C resulted in the yields of hydrocarbons and alcohols up to 93.61 %, beneficial to the further utilization of the co-pyrolytic products. Not only did the co-pyrolysis valorize its products, but also it enhanced their co-circularity. Artificial neural network-based joint optimization showed CP73 in the range of 517–1000 °C as the best combination of the conditions. The study provides new insights into the co-pyrolytic disposal of spent coffee grounds and polyethylene so as to improve the waste stream reduction and the valorization of energy and products.
► We develop optimization approaches for dynamic ride-sharing. ► We build a simulation environment to test different dynamic ride-sharing concepts. ► We show that dynamic ride-sharing may increase ...the efficiency of urban transportation. ► We demonstrate the value of sophisticated matching techniques.
Smartphone technology enables dynamic ride-sharing systems that bring together people with similar itineraries and time schedules to share rides on short-notice. This paper considers the problem of matching drivers and riders in this dynamic setting. We develop optimization-based approaches that aim at minimizing the total system-wide vehicle miles incurred by system users, and their individual travel costs. To assess the merits of our methods we present a simulation study based on 2008 travel demand data from metropolitan Atlanta. The simulation results indicate that the use of sophisticated optimization methods instead of simple greedy matching rules substantially improve the performance of ride-sharing systems. Furthermore, even with relatively low participation rates, it appears that sustainable populations of dynamic ride-sharing participants may be possible even in relatively sprawling urban areas with many employment centers.
Swarm intelligence (SI) algorithms, including ant colony optimization, particle swarm optimization, bee-inspired algorithms, bacterial foraging optimization, firefly algorithms, fish swarm ...optimization and many more, have been proven to be good methods to address difficult optimization problems under stationary environments. Most SI algorithms have been developed to address stationary optimization problems and hence, they can converge on the (near-) optimum solution efficiently. However, many real-world problems have a dynamic environment that changes over time. For such dynamic optimization problems (DOPs), it is difficult for a conventional SI algorithm to track the changing optimum once the algorithm has converged on a solution. In the last two decades, there has been a growing interest of addressing DOPs using SI algorithms due to their adaptation capabilities. This paper presents a broad review on SI dynamic optimization (SIDO) focused on several classes of problems, such as discrete, continuous, constrained, multi-objective and classification problems, and real-world applications. In addition, this paper focuses on the enhancement strategies integrated in SI algorithms to address dynamic changes, the performance measurements and benchmark generators used in SIDO. Finally, some considerations about future directions in the subject are given.
•Existing dynamic optimisation surveys focus entirely on evolutionary algorithms and little on swarm intelligence algorithms. This survey provides a comprehensive survey dedicated to swarm intelligence algorithms to fill in the gap in the dynamic optimisation domain.•In addition to the mainstream ant colony optimisation and particle swarm optimisation algorithms; recent swarm intelligence applications to dynamic optimisation problems (DOPs) are included.•Provides several classifications related to both the algorithmic components and the application problem.
Dynamic interval multiobjective optimization problems (DI-MOPs) are very common in real-world applications. However, there are few evolutionary algorithms (EAs) that are suitable for tackling DI-MOPs ...up to date. A framework of dynamic interval multiobjective cooperative co-evolutionary optimization based on the interval similarity is presented in this paper to handle DI-MOPs. In the framework, a strategy for decomposing decision variables is first proposed, through which all the decision variables are divided into two groups according to the interval similarity between each decision variable and interval parameters. Following that, two subpopulations are utilized to cooperatively optimize decision variables in the two groups. Furthermore, two response strategies, i.e., a strategy based on the change intensity and a random mutation strategy, are employed to rapidly track the changing Pareto front of the optimization problem. The proposed algorithm is applied to eight benchmark optimization instances as well as a multiperiod portfolio selection problem and compared with five state-of-the-art EAs. The experimental results reveal that the proposed algorithm is very competitive on most optimization instances.
Dynamic changes are an important and inescapable aspect of many real-world optimization problems. Designing algorithms to find and track desirable solutions while facing challenges of dynamic ...optimization problems is an active research topic in the field of swarm and evolutionary computation. To evaluate and compare the performance of algorithms, it is imperative to use a suitable benchmark that generates problem instances with different controllable characteristics. In this article, we give a comprehensive review of existing benchmarks and investigate their shortcomings in capturing different problem features. We then propose a highly configurable benchmark suite, the generalized moving peaks benchmark, capable of generating problem instances whose components have a variety of properties, such as different levels of ill-conditioning, variable interactions, shape, and complexity. Moreover, components generated by the proposed benchmark can be highly dynamic with respect to the gradients, heights, optimum locations, condition numbers, shapes, complexities, and variable interactions. Finally, several well-known optimizers and dynamic optimization algorithms are chosen to solve generated problems by the proposed benchmark. The experimental results show the poor performance of the existing methods in facing new challenges posed by the addition of new properties.