arXiv subjects

Alexander Scheinker

Publications and source records attributed to Alexander Scheinker.

At least 37 records · Page 2Linked to original sources

arXiv2024

cDVAE: Multimodal Generative Conditional Diffusion Guided by Variational Autoencoder Latent Embedding for Virtual 6D Phase Space Diagnostics

Imaging the 6D phase space of a beam in a particle accelerator in a single shot is currently impossible. Single shot beam measurements only exist for certain 2D beam projections and these methods are destructive. A virtual diagnostic that can generate an accurate prediction of a beam's 6D phase space would be incredibly useful for precisely controlling the beam. In this work, a generative conditional diffusion-based approach to creating a virtual diagnostic of all 15 unique 2D projections of a beam's 6D phase space is developed. The diffusion process is guided by a combination of scalar parameters and images that are converted to low-dimensional latent vector representation by a variational autoencoder (VAE). We demonstrate that conditional diffusion guided by VAE (cDVAE) can accurately reconstruct all 15 of the unique 2D projections of a charge particle beam's 6 phase space for the HiRES compact accelerator.

physics.acc-ph↗

arXiv2024

Conditional Guided Generative Diffusion for Particle Accelerator Beam Diagnostics

Advanced accelerator-based light sources such as free electron lasers (FEL) accelerate highly relativistic electron beams to generate incredibly short (10s of femtoseconds) coherent flashes of light for dynamic imaging, whose brightness exceeds that of traditional synchrotron-based light sources by orders of magnitude. FEL operation requires precise control of the shape and energy of the extremely short electron bunches whose characteristics directly translate into the properties of the produced light. Control of short intense beams is difficult due to beam characteristics drifting with time and complex collective effects such as space charge and coherent synchrotron radiation. Detailed diagnostics of beam properties are therefore essential for precise beam control. Such measurements typically rely on a destructive approach based on a combination of a transverse deflecting resonant cavity followed by a dipole magnet in order to measure a beam's 2D time vs energy longitudinal phase-space distribution. In this paper, we develop a non-invasive virtual diagnostic of an electron beam's longitudinal phase space at megapixel resolution (1024 x 1024) based on a generative conditional diffusion model. We demonstrate the model's generative ability on experimental data from the European X-ray FEL.

physics.acc-ph↗

arXiv2024

Accelerator system parameter estimation using variational autoencoded latent regression

Particle accelerators are time-varying systems whose components are perturbed by external disturbances. Tuning accelerators can be a time-consuming process involving manual adjustment of multiple components, such as RF cavities, to minimize beam loss due to time-varying drifts. The high dimensionality of the system ($\sim$100 amplitude and phase RF settings in the LANSCE accelerator) makes it difficult to achieve optimal operation. The time-varying drifts and the dimensionality make system parameter estimation a challenging optimization problem. In this work, we propose a Variational Autoencoded Latent Regression (VALeR) model for robust estimation of system parameters using 2D unique projections of a charged particle beam's 6D phase space. In VALeR, VAE projects the phase space projections into a lower-dimensional latent space, and a dense neural network maps the latent space onto the space of system parameters. The trained network can predict system parameters for unseen phase space projections. Furthermore, VALeR can generate new projections by randomly sampling the latent space of VAE and also estimate the corresponding system parameters.

physics.acc-ph↗

arXiv2024

Towards latent space evolution of spatiotemporal dynamics of six-dimensional phase space of charged particle beams

Addressing the charged particle beam diagnostics in accelerators poses a formidable challenge, demanding high-fidelity simulations in limited computational time. Machine learning (ML) based surrogate models have emerged as a promising tool for non-invasive charged particle beam diagnostics. Trained ML models can make predictions much faster than computationally expensive physics simulations. In this work, we have proposed a temporally structured variational autoencoder model to autoregressively forecast the spatiotemporal dynamics of the 15 unique 2D projections of 6D phase space of charged particle beam as it travels through the LANSCE linear accelerator. In the model, VAE embeds the phase space projections into a lower dimensional latent space. A long-short-term memory network then learns the temporal correlations in the latent space. The trained network can evolve the phase space projections across further modules provided the first few modules as inputs. The model predicts all the projections across different modules with low mean squared error and high structural similarity index.

physics.acc-ph↗

arXiv2024

Practical Safe Extremum Seeking with Assignable Rate of Attractivity to the Safe Set

We present Assignably Safe Extremum Seeking (ASfES), an algorithm designed to minimize a measured objective function while maintaining a measured metric of safety (a control barrier function or CBF) be positive in a practical sense. We ensure that for trajectories with safe initial conditions, the violation of safety can be made arbitrarily small with appropriately chosen design constants. We also guarantee an assignable ``attractivity'' rate: from unsafe initial conditions, the trajectories approach the safe set, in the sense of the measured CBF, at a rate no slower than a user-assigned rate. Similarly, from safe initial conditions, the trajectories approach the unsafe set, in the sense of the CBF, no faster than the assigned attractivity rate. The feature of assignable attractivity is not present in the semiglobal version of safe extremum seeking, where the semiglobality of convergence is achieved by slowing the adaptation. We also demonstrate local convergence of the parameter to a neighborhood of the minimum of the objective function constrained to the safe set. The ASfES algorithm and analysis are multivariable, but we also extend the algorithm to a Newton-Based ASfES scheme (NB-ASfES) which we show is only useful in the scalar case. The proven properties of the designs are illustrated through simulation examples.

math.OC↗

arXiv2024

A conditional latent autoregressive recurrent model for generation and forecasting of beam dynamics in particle accelerators

Particle accelerators are complex systems that focus, guide, and accelerate intense charged particle beams to high energy. Beam diagnostics present a challenging problem due to limited non-destructive measurements, computationally demanding simulations, and inherent uncertainties in the system. We propose a two-step unsupervised deep learning framework named as Conditional Latent Autoregressive Recurrent Model (CLARM) for learning the spatiotemporal dynamics of charged particles in accelerators. CLARM consists of a Conditional Variational Autoencoder (CVAE) transforming six-dimensional phase space into a lower-dimensional latent distribution and a Long Short-Term Memory (LSTM) network capturing temporal dynamics in an autoregressive manner. The CLARM can generate projections at various accelerator modules by sampling and decoding the latent space representation. The model also forecasts future states (downstream locations) of charged particles from past states (upstream locations). The results demonstrate that the generative and forecasting ability of the proposed approach is promising when tested against a variety of evaluation metrics.

physics.acc-ph↗

arXiv2023

Semi-Global Practical Extremum Seeking with Practical Safety

We introduce a type of safe extremum seeking (ES) controller, which minimizes an unknown objective function while also maintaining practical positivity of an unknown barrier function. We show semi-global practical asymptotic stability of our algorithm and present an analogous notion of practical safety. The dynamics of the controller are inspired by the quadratic program (QP) based safety filter designs which, in the literature, are more commonly used in cases where the barrier function is known. Conditions on the barrier and objective function are explored showing that non convex problems can be solved. A Lyapunov argument is proposed to achieve the main results of the paper. Finally, an example is given of the algorithm which solves the constrained optimization problem.

math.OC↗

arXiv2023

Experimental Safe Extremum Seeking for Accelerators

We demonstrate the recent designs of Safe Extremum Seeking (Safe ES) on the 1 kilometer-long charged particle accelerator at the Los Alamos Neutron Science Center (LANSCE). Safe ES is a modification of ES which, in addition to minimizing an analytically unknown cost, also employs a safety filter based on an analytically unknown control barrier function (CBF) safety metric. Accelerator tuning is necessitated by the accelerators being large, with many drifting parameters due to thermal effects and degradation. At the same time, safe operation (the maintenance of state constraints) is crucial, as damage brings astronomical costs, both financially and in operation downtime. Our measured (but analytically unknown) safety metric is the beam current. We perform multivariable Safe ES on three accelerator applications, in which we adapt 4, 6, and 3 magnet strength parameters, respectively. Two of the three applications are for validated simulation models of beamlines at LANSCE: the first for the Proton Radiography (pRad) beamline of 800 MeV protons for spot size tuning; the second on a high performance code, HPSim, for tuning the low energy beam transport (LEBT) region of of 750 keV protons. The third is an experimental tuning of the steering magnets in the LEBT at LANSCE.

physics.acc-ph↗

arXiv2023

Virtual-Diagnostic-Based Time Stamping for Ultrafast Electron Diffraction

In this work, non-destructive virtual diagnostics are applied to retrieve the electron beam time of arrival and energy in a relativistic ultrafast electron diffraction (UED) beamline using independently-measured machine parameters. This technique has the potential to improve temporal resolution of pump and probe UED scans. Fluctuations in time of arrival have multiple components, including a shot-to-shot jitter and a long-term drift which can be separately addressed by closed loop feedback systems. A linear-regression-based model is used to fit the beam energy and time of arrival and is shown to be able to predict accurately behavior for both on long and short time scales. More advanced time-series analysis based on machine learning techniques can be applied to improve this prediction further.

physics.acc-ph↗

arXiv2023

Physics-constrained 3D Convolutional Neural Networks for Electrodynamics

We present a physics-constrained neural network (PCNN) approach to solving Maxwell's equations for the electromagnetic fields of intense relativistic charged particle beams. We create a 3D convolutional PCNN to map time-varying current and charge densities J(r,t) and p(r,t) to vector and scalar potentials A(r,t) and V(r,t) from which we generate electromagnetic fields according to Maxwell's equations: B=curl(A), E=-div(V)-dA/dt. Our PCNNs satisfy hard constraints, such as div(B)=0, by construction. Soft constraints push A and V towards satisfying the Lorenz gauge.

physics.acc-ph↗

arXiv2022

2022 Review of Data-Driven Plasma Science

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, plasma processing 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.

physics.plasm-ph↗

arXiv2022

Machine Learning in Nuclear Physics

Advances in machine learning methods provide tools that have broad applicability in scientific research. These techniques are being applied across the diversity of nuclear physics research topics, leading to advances that will facilitate scientific discoveries and societal applications. This Review gives a snapshot of nuclear physics research which has been transformed by machine learning techniques.

nucl-th↗

arXiv2022

Feedback and control systems for future linear colliders: White Paper for Snowmass 2021 Topical Group AF07-RF

Particle accelerators for high energy physics will generate TeV-scale particle beams in large, multi-Km size machines colliding high brightness beams at the interaction point [1-4]. The high luminosity in such machines is achieved by producing very small asymmetric beam size at the interaction point, with short durations to minimize beam-beam effects. Tuning energy, timing and position of the beam for optimal performance will require high-precision controls of amplitude and phase of high-frequency electromagnetic fields and real-time processing of complex algorithms. The stability of the colliding beams has a large impact on the collider's effective luminosity. Therefore, the technology readiness level of diagnostic and control systems will be a major consideration in the collider design. The technical requirements of such systems depend on the specifics of beam parameters, such as transverse and longitudinal dimensions, charge/pulse and beam pulse format, which are driven by the accelerating technology of choice. While feedback systems with single bunch position monitor resolution below 50 nm and latency <300 ns have been demonstrated in beam test facilities, many advanced collider concepts make use of higher repetition rates, brighter beams and higher accelerating frequencies, and will require better performance, up to 1-2 order of magnitude, demanding aggressive R&D to be able to deliver and maintain the targeted luminosity.

physics.acc-ph↗

arXiv2022

Adaptive Machine Learning for Time-Varying Systems: Towards 6D Phase Space Diagnostics of Short Intense Charged Particle Beams

As charged particle bunches become shorter and more intense, the effects of nonlinear intra-bunch collective interactions such as space charge forces and bunch-to-bunch influences such as wakefields and coherent synchrotron radiation also increase. Shorter more intense bunches are also more difficult to accurately image because their dimensions are beyond the resolution of existing diagnostics and they may be destructive to intercepting diagnostics. The limited availability of detailed diagnostics for intense high energy beams is a fundamental challenge for the accelerator community because both beams and accelerators are time-varying systems that change in unpredictable ways. The detailed 6D (x,y,z,px,py,pz) distributions of beams emerging from sources vary with time due to factors such as evolving photocathode laser intensity profiles and the quantum efficiency of photocathodes. Accelerator magnets, RF amplifiers, and control systems are perturbed by external disturbances, beam-loading effects, temperature variations, and misalignments. Although machine learning (ML) methods have grown in popularity in the accelerator community in recently years, they are fundamentally limited when it comes to time-varying systems for which most current approaches are to simply collect large new data sets and perform re-training, something which is not feasible for busy accelerator user facilities because detailed beam measurements usually interrupt operations. New adaptive machine learning (AML) methods designed for time-varying systems are needed to aid in the diagnostics and control of high-intensity, ultrashort beams by combining deep learning tools such as convolutional neural network-based encoder-decoder architectures, model-independent feedback, physics constraints, and online models with real time non-invasive beam data, to provide a detailed virtual view of intense bunch dynamics.

physics.acc-ph↗

arXiv2021

Adaptive Machine Learning for Time-Varying Systems: Low Dimensional Latent Space Tuning

Machine learning (ML) tools such as encoder-decoder convolutional neural networks (CNN) can represent incredibly complex nonlinear functions which map between combinations of images and scalars. For example, CNNs can be used to map combinations of accelerator parameters and images which are 2D projections of the 6D phase space distributions of charged particle beams as they are transported between various particle accelerator locations. Despite their strengths, applying ML to time-varying systems, or systems with shifting distributions, is an open problem, especially for large systems for which collecting new data for re-training is impractical or interrupts operations. Particle accelerators are one example of large time-varying systems for which collecting detailed training data requires lengthy dedicated beam measurements which may no longer be available during regular operations. We present a recently developed method of adaptive ML for time-varying systems. Our approach is to map very high (N>100k) dimensional inputs (a combination of scalar parameters and images) into the low dimensional (N~2) latent space at the output of the encoder section of an encoder-decoder CNN. We then actively tune the low dimensional latent space-based representation of complex system dynamics by the addition of an adaptively tuned feedback vector directly before the decoder sections builds back up to our image-based high-dimensional phase space density representations. This method allows us to learn correlations within and to quickly tune the characteristics of incredibly high parameter systems and to track their evolution in real time based on feedback without massive new data sets for re-training.

cs.LG↗

arXiv2021

Adaptive Latent Space Tuning for Non-Stationary Distributions

Powerful deep learning tools, such as convolutional neural networks (CNN), are able to learn the input-output relationships of large complicated systems directly from data. Encoder-decoder deep CNNs are able to extract features directly from images, mix them with scalar inputs within a general low-dimensional latent space, and then generate new complex 2D outputs which represent complex physical phenomenon. One important challenge faced by deep learning methods is large non-stationary systems whose characteristics change quickly with time for which re-training is not feasible. In this paper we present a method for adaptive tuning of the low-dimensional latent space of deep encoder-decoder style CNNs based on real-time feedback to quickly compensate for unknown and fast distribution shifts. We demonstrate our approach for predicting the properties of a time-varying charged particle beam in a particle accelerator whose components (accelerating electric fields and focusing magnetic fields) are also quickly changing with time.

stat.ML↗

arXiv2021

Adaptive deep learning for time-varying systems with hidden parameters: Predicting changing input beam distributions of compact particle accelerators

Machine learning (ML) tools such as encoder-decoder deep convolutional neural networks (CNN) are able to extract relationships between inputs and outputs of large complex systems directly from raw data. For time-varying systems the predictive capabilities of ML tools degrade as the systems are no longer accurately represented by the data sets with which the ML models were trained. Re-training is possible, but only if the changes are slow and if new input-output training data measurements can be made online non-invasively. In this work we present an approach to deep learning for time-varying systems in which adaptive feedback based only on available system output measurements is applied to encoded low-dimensional dense layers of encoder-decoder type CNNs. We demonstrate our method in developing an inverse model of a complex charged particle accelerator system, mapping output beam measurements to input beam distributions while both the accelerator components and the unknown input beam distribution quickly vary with time. We demonstrate our results using experimental measurements of the input and output beam distributions of the HiRES ultra-fast electron diffraction (UED) microscopy beam line at Lawrence Berkeley National Laboratory. We show how our method can be used to aid both physics and ML-based surrogate online models to provide non-invasive beam diagnostics and we also demonstrate how our method can be used to automatically track the time varying quantum efficiency map of a particle accelerator's photocathode.

physics.acc-ph↗

arXiv2020

Adaptive 3D convolutional neural network-based reconstruction method for 3D coherent diffraction imaging

We present a novel adaptive machine-learning based approach for reconstructing three-dimensional (3D) crystals from coherent diffraction imaging (CDI). We represent the crystals using spherical harmonics (SH) and generate corresponding synthetic diffraction patterns. We utilize 3D convolutional neural networks (CNN) to learn a mapping between 3D diffraction volumes and the SH which describe the boundary of the physical volumes from which they were generated. We use the 3D CNN-predicted SH coefficients as the initial guesses which are then fine tuned using adaptive model independent feedback for improved accuracy.

physics.comp-ph↗