Toward Safe Dose Delivery in Plasma Medicine using Projected Neural Network-based Fast Approximate NMPC

Bonzanini, Angelo D.; Paulson, Joel A.; Graves, David B.; Mesbah, Ali

doi:10.1016/j.ifacol.2020.12.1208

Cited by 19 publications

(22 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This can especially be the case when sophisticated process models are used for MPC, or when the objective is to control the fast process dynamics that would hinge on fast measurement sampling frequencies on the order of KHz to MHz or even possibly faster. ML has proven useful for developing so-called approximate MPC approaches that learn a cheap-to-evaluate, explicit expression for the MPC law using data generated from offline solution of an MPC problem [141]; see Figure 1. A variety of function approximators, ranging from polynomials to deep neural networks, have shown promise for approximating optimization-based control laws with surrogates that can be evaluated on fast sampling times.…”

Section: ML Control Theorymentioning

confidence: 99%

“…To guarantee satisfaction of safety-critical constraints of plasma processes in the presence of approximation errors and system uncertainties, the control inputs computed by the neural network can be projected onto a safe input set that is constructed using the notion of robust invariant sets. Safe neural network-based controllers can play a pivotal role for control of fast-sampling plasma processes using resource-limited embedded control hardware [141]. A feedforward fully connected neural network approach has been implemented to measure the electron FIG.…”

Section: ML Control Theorymentioning

confidence: 99%

See 1 more Smart Citation

2022 Review of Data-Driven Plasma Science

Archibald¹,

Asif²,

Becker³

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

Data science and technology offer transformative tools and methods to science. This review article highlights latest development and progress in the interdisciplinary field of data-driven plasma science (DDPS). A large amount of data and machine learning algorithms go hand in hand. Most plasma data, whether experimental, observational or computational, are generated or collected by machines today. It is now becoming impractical for humans to analyze all the data manually. Therefore, it is imperative to train machines to analyze and interpret (eventually) such data as intelligently as humans but far more efficiently in quantity. Despite the recent impressive progress in applications of data science to plasma science and technology, the emerging field of DDPS is still in its infancy. Fueled by some of the most challenging problems such as fusion energy, plasmaprocessing of materials, and fundamental understanding of the universe through observable plasma phenomena, it is expected that DDPS continues to benefit significantly from the interdisciplinary marriage between plasma science and data science into the foreseeable future. CONTENTSI. Introduction 6 II. Fundamental Data Science 6 A. Introduction 6 B. Data Reduction/Compression 7 C. Dimensional reduction and sparse modeling 8 D. ML-enhanced modeling and simulation 9 E. ML Hardware and integration with models 10 F. Workflow Automation 11 G. Uncertainty Quantification 12 H. Visualization and Data Understanding 13 I. ML Control Theory 15 III. Basic Plasma Physics and Laboratory Experiments A. Introduction B. Spectroscopy, imaging and tomography C. Sparse measurement and noise D. Synthetic instruments and data E. Experimental data visualization F. High-rep rate laser experiments G. Charged particle beams H. Control and Optimisation of Plasma Accelerator Experiments I. Dusty and complex plasmas J. Physics and machine learning K. Challenges and outlook 5 IV. Magnetic Confinement Fusion 36 A. Introduction 36 B. Data-Driven Physics Models 36 C. Optimizing experimental workflows with data-driven methods 37 D. Diagnostics and Fusion Data Streams 38 E. Prediction of Tokamak Disruption 39 F. Surrogate models of fusion plasma 40 G. Magnetic Fusion Energy Data Challenges and Solutions 41 H. Data Science for extreme scale simulation 43 I. Challenges and outlook 44 V. Inertial confinement fusion and high-energy-density physics 45 A. Introduction 45 B. Representation learning for multimodal data 45 C. Transfer learning for simulation and experimen 47 D. Uncertainty quantification and Bayesian inference 47 E. High-performance computing and simulation acceleration 49 F. Design exploration and optimization 49 G. Self-driving experimental facilities 51 H. Challenges and outlook 52 VI. Space and astronomical plasmas 52 A. Introduction 52 B. Space and ground instruments 53 C. Space weather prediction 55 D. Transfer learning to improve historic data 56 E. Surrogate models of fluid closures using machine learning 56 F. Magnetic reconnection 58 G. Challenges and outlook 60 VII. Plasma tec...

show abstract

Section: ML Control Theorymentioning

confidence: 99%

Section: ML Control Theorymentioning

confidence: 99%

2022 Review of Data-Driven Plasma Science

Archibald¹,

Asif²,

Becker³

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…On the other hand, the LD 50 value of 34.67 mJ/cell is also an important reference datapoint when studying the lethal and/or sub-lethal effects of other cell lines using different types of CAPs. The published results concerning plasma dosage can be roughly divided into two types: one is the quantitative definition of plasma dosage [8], and the other skips the definition of plasma dosage but controls specific effects with the help of advanced algorithms such as machine learning [19,[23][24][25]. For the first type, the number densities of all the considered RONS should be precisely measured or calculated with great efforts; in particular, whenever the discharge parameters or even the environmental conditions are changed, all the measurements or calculations need to be re-conducted.…”

Section: Determination Of Ld 50mentioning

confidence: 99%

“…These have made it extremely difficult to propose an accurate and versatile definition for plasma dosage, considering all the possible mechanisms. More recently, in order to adjust and/or control the effective CAP dosages, some researchers begin to focus on machine learning and neural network [18,19], which further proves the complexity and difficulty of the plasma dosage definition. In our opinion, since the concept of "dosage" is frequently used in pharmaceuticals, the definition of plasma dosage can refer to the basic knowledge in pharmacy.…”

Section: Introductionmentioning

confidence: 99%

“…The value of LD 50 can serve as an indicator to help the users determine and control the plasma dosage used in their practical studies. Distinguished from the physics-and/or chemistry-based method to define or control the plasma dosage [8,18,19,[23][24][25], this biology-based approach has the advantages of being versatile to various CAP sources and easy to be conducted without considering the specific and complex interaction processes between the CAPs and the treated biological targets.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

A Novel Method for Estimating the Dosage of Cold Atmospheric Plasmas in Plasma Medical Applications

Zhao

et al. 2021

Applied Sciences

View full text Add to dashboard Cite

Cold atmospheric plasmas (CAPs) used in plasma medicine have shown great potential in various aspects including wound healing, dermatology, cancer therapy, etc. It is one of the important issues to determine the plasma dosage in plasma medicine because it dominates the specific plasma treatment results. However, the multi-process interactions between CAPs and biological materials make it rather challenging to give an accurate and versatile definition for plasma dosage. In this study, the ratio of the discharge energy to the number of the treated in vitro kidney cells (mJ/cell) was employed as the unit of the plasma dosage. Additionally, inspired by basic knowledge of pharmacy, the median lethal dose (LD50) was employed to help estimate the plasma dosage. The experimental results show that the value of LD50 using the newly designed CAP Bio-Med Platform for the kidney cells is 34.67 mJ/cell. This biology-based method has the advantages of easy operation, independence of specific CAP sources, and also independence of complex interactions between CAPs and the treated biological targets, and consequently, may provide a new direction to quantitatively define the plasma dosage in various plasma medical applications.

show abstract

Computationally efficient integrated design and predictive control of flexible energy systems using multi‐fidelity simulation‐based Bayesian optimization

Sorourifar

Choksi

Paulson

2021

Optim Control Appl Methods

Self Cite

View full text Add to dashboard Cite

We present a multi‐fidelity black‐box optimization approach for integrated design and control (IDC) of constrained nonlinear systems in the presence of uncertainty. The IDC framework is becoming increasingly important for the systematic design of next‐generation (flexible) manufacturing and energy systems. However, identifying optimal solutions to realistic IDC problems is intractable when (i) the dynamics occur on much shorter timescales than the system lifetime, (ii) the uncertainties are described by continuous random variables with high variance, and (iii) operational decisions involve a mixture of discrete and continuous variables. Instead of aggressively simplifying the problem to improve tractability, we develop a simulation‐based optimization procedure using high‐quality decision rules that map information that can be measured online to optimal control actions. In particular, we rely on the Bayesian optimization (BO) framework that has been shown to perform very well on noisy and expensive‐to‐evaluate objective functions. We also discuss how BO can be extended to take advantage of computationally cheaper low‐fidelity approximations to the high‐fidelity IDC cost function. Three major low‐fidelity approximation strategies are described in this work, which are related to the simplification of the system simulator, decision rule solution method, and time grid. Lastly, we demonstrate the advantages of multi‐fidelity BO on the design of a solar‐powered building heating/cooling system (with battery and grid support) under uncertain weather and demand conditions with hourly variation over a year‐long planning horizon.

show abstract

Toward Safe Dose Delivery in Plasma Medicine using Projected Neural Network-based Fast Approximate NMPC

Cited by 19 publications

References 22 publications

2022 Review of Data-Driven Plasma Science

2022 Review of Data-Driven Plasma Science

A Novel Method for Estimating the Dosage of Cold Atmospheric Plasmas in Plasma Medical Applications

Computationally efficient integrated design and predictive control of flexible energy systems using multi‐fidelity simulation‐based Bayesian optimization

Contact Info

Product

Resources

About