Article
Mathematics, Applied
Anatoly M. Tsirlin, Sergey Karpesh, Alexander Balunov, Ilya O. Starodumov
Summary: This paper discusses the synthesis of heat exchange systems based on the criteria of minimum irreversibility and minimum contact surface. It considers various factors such as variable heat capacity of fluids, phase changes, and different hydrodynamics. The study also takes into account temperature limits at the inlet and outlet of the heat exchange system. To synthesize the system, it is necessary to choose the appropriate contact structure, free flow parameters, distribution of contact surfaces, and thermal loads between double-flow heat exchangers.
MATHEMATICAL METHODS IN THE APPLIED SCIENCES
(2022)
Article
Social Sciences, Interdisciplinary
Luis Felipe R. Murillo, Caitlin Wylie, Phil Bourne
Summary: In the context of ethical concerns about data-intensive analytics, data ethics has become a subject of debate over its definition. Recent scholarship in science and technology studies has provided opportunities to transform teaching methods. The emerging literature on critical data ethics has fostered interdisciplinary collaboration to examine issues in digital systems using a hands-on approach.
BIG DATA & SOCIETY
(2023)
Article
Mathematics, Applied
Rachid Malti, Abir Mayoufi, Stephane Victor
Summary: This paper discusses the experiment design problem in system identification, focusing on fractional models. The best input signal frequency is chosen by minimizing the D-optimality criterion of the covariance matrix. Different cases of unknown parameters are considered, and the difficulty of obtaining the optimum analytically is shown.
COMMUNICATIONS IN NONLINEAR SCIENCE AND NUMERICAL SIMULATION
(2022)
Article
Automation & Control Systems
Xavier Bombois, Federico Morelli, Hakan Hjalmarsson, Laurent Bako, Kevin Colin
Summary: In the least costly experiment design, the optimal spectrum of an identification experiment is determined to minimize the cost while maintaining accuracy on the identified parameter vector. By using a convex optimization approach based on a-priori uncertainty for the true system, issues like underestimating actual cost and lower accuracy of identified model can be prevented from happening. This method is particularly effective when identifying the spectrum of a multisine signal.
Article
Chemistry, Physical
Xiangrong Song, Dongyang Sun, Xuefeng Liang
Summary: This paper proposes a design method for a validation experiment to determine the ability of a deterioration model to simulate the actual deterioration process. A normalized area metric based on probability density functions is developed for validation results quantification. A design method for the validation experiment is also proposed using collaborative optimization and Latin hypercube sampling.
Article
Biochemical Research Methods
Gokce Uludogan, Elif Ozkirimli, Kutlu O. Ulgen, Nilgun Karal, Arzucan Ozgur
Summary: The article introduces a method that utilizes pretrained biochemical language models to initialize targeted molecule generation models and proposes two warm start strategies. The results show that utilizing pretrained models can improve model performance, and the one-stage strategy has better generalization ability in compound generation.
Article
Automation & Control Systems
Emanuele Fabbiani, Pulkit Nahata, Giuseppe De Nicolao, Giancarlo Ferrari-Trecate
Summary: The paper proposes an online learning procedure for estimating the network admittance matrix, improving on existing techniques and substantiating its effectiveness through numerical studies on an IEEE testbed.
IEEE TRANSACTIONS ON CONTROL SYSTEMS TECHNOLOGY
(2022)
Article
Energy & Fuels
Moritz Streb, Mathilda Ohrelius, Matilda Klett, Goran Lindbergh
Summary: This study presents an algorithmic method for designing current input signals to optimize parameter identifiability from voltage measurements in Li-ion batteries. The proposed approach significantly improves model performance and parameter accuracy, while lowering experimental burden.
JOURNAL OF ENERGY STORAGE
(2022)
Article
Computer Science, Artificial Intelligence
Tomas Kulvicius, Sebastian Herzog, Minija Tamosiunaite, Florentin Worgoetter
Summary: The article introduces a novel configuration of multilayer networks for solving path planning on a maze with multiple start and endpoint efficiently without the need for network training. This approach can quickly solve very large mazes with dense obstacle configurations and thousands of importance-weighted path endpoints in a single pass on parallel hardware, providing solutions identical to classical algorithms.
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
(2022)
Article
Physics, Multidisciplinary
Salah Haggag
Summary: A new approach to construct relativistic stellar models is presented, which formulates the problem as an optimal control problem with an objective function to be optimized. The results show that using this approach, the equation of state for a relativistic incompressible fluid can be determined. The article also explores extensions to other configurations.
Article
Engineering, Chemical
Dauda Ibrahim, Megan Jobson, Jie Li, Gonzalo Guillen-Gosalbez
Summary: The article introduces an optimization-based approach for the optimal design of distillation units considering a flexible operation under a range of potential feedstocks. By combining artificial neural networks and support vector machines, the method effectively models performance prediction and design filtering for crude oil distillation units. The proposed approach successfully identifies design alternatives capable of handling various feedstocks effectively.
CHEMICAL ENGINEERING RESEARCH & DESIGN
(2021)
Article
Automation & Control Systems
Hunmin Kim, Minghui Zhu
Summary: Entities in multiagent systems may compete due to conflicting subobjectives. To address performance degradation, a bilevel lottery model where a social planner selects a reward and players determine a Nash equilibrium is proposed. Efficiency losses may occur when the Nash equilibrium does not align with the social optimum. An optimal bilevel lottery design problem is formulated to find the least reward and perturbations that produce a socially optimal payoff. The price of anarchy and the behavior of public goods and Nash equilibrium are characterized with respect to the reward and perturbations. A convex approximation relaxes the problem, and conditions for exactness are identified.
IEEE TRANSACTIONS ON SYSTEMS MAN CYBERNETICS-SYSTEMS
(2023)
Article
Chemistry, Multidisciplinary
Guoqiang Dun, Ning Mao, Xinxin Ji, Fuli Zhang, Wenyi Ji
Summary: In this study, a corn-overlapped strip fertilizer spreader was designed to achieve uniform spreading of fertilizer in hilly areas. A model was developed to determine the factors affecting fertilizer distribution pattern, and simulation tests were conducted to optimize these factors. The test results showed that the optimized combination of parameters resulted in high uniformity coefficients for both transversal and longitudinal fertilizer distribution. The validation test confirmed the accuracy of the simulation results. This study provides evidence for further improving the performance of corn mid-tillage fertilizer applicators.
APPLIED SCIENCES-BASEL
(2023)
Article
Chemistry, Multidisciplinary
Qing Zhao, Lijun Li, Zechao Wu, Xin Guo, Jun Li
Summary: In this paper, a four-degree-of-freedom camellia-pollen-picking manipulator is proposed and designed to solve the problem of lacking mechanized equipment for picking camellia pollen in agricultural machinery. A structural-parameter-optimization method is used to make the manipulator reach the target space quickly and efficiently. The stability and working space of the manipulator are verified through static analysis, simulation, and experiments.
APPLIED SCIENCES-BASEL
(2022)
Article
Statistics & Probability
Belmiro P. M. Duarte, Anthony C. Atkinson, Jose F. O. Granjo, Nuno M. C. Oliveira
Summary: The article discusses the theory of optimal design of experiments for implicit algebraic models, utilizing nonlinear programming to convert the problem and ensure the global optimality of continuous optimal designs.
JOURNAL OF THE AMERICAN STATISTICAL ASSOCIATION
(2022)
Article
Automation & Control Systems
Yang Zheng, Aivar Sootla, Antonis Papachristodoulou
Summary: This article introduces a new notion of block factor-width-two matrices and builds a new hierarchy of inner and outer approximations of the cone of positive semidefinite matrices. This improves the inner-approximation of the PSD cone. In the context of SOS optimization, this leads to a block extension of the scaled diagonally dominant sum-of-squares polynomials. Numerical experiments confirm the theoretical findings.
IEEE TRANSACTIONS ON AUTOMATIC CONTROL
(2023)
Article
Automation & Control Systems
Jared Miller, Yang Zheng, Mario Sznaier, Antonis Papachristodoulou
Summary: This paper presents a concept of decomposed structured subsets to approximate an SDP with structured subsets after an equivalent conversion. The lower/upper bounds found by approximation after conversion become tighter than the bounds obtained by approximating the original SDP directly.
Article
Automation & Control Systems
Giorgio Valmorbida, Antonis Papachristodoulou
Summary: This article presents strategies for the state-feedback control law design of nonlinear control laws with saturating inputs. The input constraints are handled by considering a generalized local sector inequality, allowing the study of nonsymmetric saturation bounds. A numerical formulation based on solving Lyapunov inequalities with sum-of-squares programming is presented for polynomial systems.
IEEE TRANSACTIONS ON AUTOMATIC CONTROL
(2022)
Article
Automation & Control Systems
Han Shu, Xuan Zhang, Na Li, Antonis Papachristodoulou
Summary: This article presents a control reconfiguration approach to improve the performance of two classes of dynamical systems. The approach involves a three-step procedure of reverse-engineering, forward-engineering, and comparing dynamics to redesign and improve the given system. Internet congestion control and distributed proportional-integral control are used as applications to showcase the effectiveness of the proposed approach.
IEEE TRANSACTIONS ON AUTOMATIC CONTROL
(2022)
Article
Biochemical Research Methods
Boon Lim, Yutong Yin, Hua Ye, Zhanfeng Cui, Antonis Papachristodoulou, Wei E. Huang
Summary: Advances in synthetic biology have enabled the reprogramming of bacteria to specifically target tumors and release anticancer drugs in a controlled manner. SimCells, chromosome-free bacteria controlled by gene circuits, have been engineered to display antibodies on their surface and effectively bind and induce death of cancer cells expressing specific biomarkers. These findings highlight the potential of SimCells for targeted cancer therapy and the application of synthetic biology in medicine.
ACS SYNTHETIC BIOLOGY
(2022)
Article
Automation & Control Systems
Han Wang, Kostas Margellos, Antonis Papachristodoulou
Summary: This paper introduces a novel Time-Triggered Dimension Reduction Algorithm (TTDRA) for solving the task assignment problem. The algorithm accelerates computational speed by convexifying the constraints and cost function. Numerical simulations verify the optimality and computational efficiency of the algorithm.
EUROPEAN JOURNAL OF CONTROL
(2022)
Article
Automation & Control Systems
Elias August, Antonis Papachristodoulou
Summary: This paper presents a controller design method for nonlinear systems, achieving global asymptotic stability and a certain cost using sum of squares programming. Compared to previous approaches, the new method is less restrictive and requires less memory.
EUROPEAN JOURNAL OF CONTROL
(2022)
Article
Automation & Control Systems
Licio Romao, Kostas Margellos, Antonis Papachristodoulou
Summary: In this study, we propose a removal scheme with a less conservative bound on constraint violation probability. By extending the number of discarded scenarios, we improve the theoretical properties of the sampling-and-discarding scheme. Our results show that our feasibility guarantees outperform the standard approach, but also highlight the inherent property of introducing additional conservatism when the number of discarded scenarios is not an integer multiple of the decision space dimension.
Review
Biotechnology & Applied Microbiology
Miroslav Gasparek, Harrison Steel, Antonis Papachristodoulou
Summary: Living organisms produce a variety of metabolites that are of interest to the pharmaceutical industry due to their potential medicinal properties. Co-culturing producer species with specific inducer microbes is an effective method to activate silent gene clusters and produce secondary metabolites. However, the mechanisms and means of induction in co-cultures are not well understood, which limits the diversity and yield of valuable compounds. This review summarizes the known physiological mechanisms of secondary metabolite production in inducer-producer consortia and discusses approaches to optimize their discovery and production.
BIOTECHNOLOGY ADVANCES
(2023)
Article
Automation & Control Systems
Licio Romao, Antonis Papachristodoulou, Kostas Margellos
Summary: We revisit the sampling and discarding approach used for quantifying the probability of constraint violation in convex scenario programs. We propose a removal scheme that consists of a cascade of optimization problems, which leads to less conservative bounds for the probability of constraint violation compared to existing methods. We also provide evidence of the tightness of the proposed bound by characterizing a class of optimization problems that achieve the given upper bound. The improved performance of the proposed methodology is illustrated through an example involving a resource sharing linear program.
IEEE TRANSACTIONS ON AUTOMATIC CONTROL
(2023)
Proceedings Paper
Automation & Control Systems
Han Wang, Kostas Margellos, Antonis Papachristodoulou
Summary: This paper introduces the application of the control barrier function approach in the synthesis of safe controllers and proposes a method to explicitly synthesize a safe control law for nonlinear control-affine systems with limited control ability. By transforming the online quadratic programming problem into an offline parameterized optimization problem, considering states as parameters, a piece-wise Lipschitz continuous function as an explicit safe controller can be obtained.
2022 IEEE 61ST CONFERENCE ON DECISION AND CONTROL (CDC)
(2022)
Proceedings Paper
Automation & Control Systems
Matthew Newton, Antonis Papachristodoulou
Summary: This paper presents a method using a Sum of Squares programming framework to directly compute the stability of non-linear systems with neural network controllers. By proposing higher order candidate Lyapunov functions and constraint multipliers, the method is able to better capture the non-linear dynamics of the system and the neural network.
2022 IEEE 61ST CONFERENCE ON DECISION AND CONTROL (CDC)
(2022)
Article
Automation & Control Systems
Emmanouil Alexis, Luca Cardelli, Antonis Papachristodoulou
Summary: Here, a highly tunable PID bio-controller is introduced and its performance is compared to PI regulation using numerical simulations.
IEEE CONTROL SYSTEMS LETTERS
(2022)
Article
Multidisciplinary Sciences
Aivar Sootla, Nicolas Delalez, Emmanouil Alexis, Arthur Norman, Harrison Steel, George H. Wadhams, Antonis Papachristodoulou
Summary: We present a new design framework, called 'dichotomous feedback', for implementing negative feedback regulation in synthetic biology. This approach utilizes the existing fluxes in the process to be controlled and takes advantage of the process's architecture to design the control law. This mechanism of signal sequestration is commonly found in biological systems and is potentially easier to achieve compared to other methods. By linking the strength of signal sequestration to the process output, our feedback regulation mechanism shapes the signal response, reduces noise, and improves system robustness.
JOURNAL OF THE ROYAL SOCIETY INTERFACE
(2022)
Article
Microbiology
Carolin C. M. Schulte, Vinoy K. Ramachandran, Antonis Papachristodoulou, Philip S. Poole
Summary: This study reconstructed a genome-scale metabolic model for Rhizobium leguminosarum and integrated various experimental data sets to investigate the metabolic characteristics of rhizosphere bacteria, undifferentiated bacteria inside root nodules, and nitrogen-fixing bacteroids. These findings provide a framework for future experimental studies and help advance efforts to engineer novel symbioses with cereal crops for sustainable agriculture.