On-the-fly optimization of synchrotron beamlines using machine learning

Abstract

Autonomous methods to align beamlines can decrease the amount of time spent on diagnostics, and also uncover better global optima leading to better beam quality. The alignment of these beamlines is a high-dimensional, expensive-to-sample optimization problem involving the simultaneous treatment of many optical elements with correlated and nonlinear dynamics. Bayesian optimization is a strategy of efficient global optimization that has proved successful in similar regimes in a wide variety of beamline alignment applications, though it has typically been implemented for particular beamlines and optimization tasks. In this paper, we present a basic formulation of Bayesian inference and Gaussian process models as they relate to multiobjective Bayesian optimization, as well as the practical challenges presented by beamline alignment. We show that the same general implementation of Bayesian optimization with special consideration for beamline alignment can quickly learn the dynamics of particular beamlines in an online fashion through hyperparameter fitting with no prior information. We present the implementation of a concise software framework for beamline alignment and test it on four different optimization problems for experiments at x-ray beamlines of the National Synchrotron Light Source II and the Advanced Light Source and an electron beam at the Accelerator Test Facility, along with benchmarking on a simulated digital twin. We discuss new applications of the framework, and the potential for a unified approach to beamline alignment at synchrotron facilities.

Latent Bayesian optimization for the autonomous alignment of synchrotron beamlines

“…This allows the Blop agent to both command and control the beamline, leading to an easier implementation. 12 Bluesky has been mainly developed by NSLS-II, with a growing international collaboration at multiple facilities where it is used and expanded. The adoption of a single standard for experimental control and analysis across many facilities allows us to apply the same automated alignment tools with relatively little effort.…”

“…They also represent the first step toward a fully autonomous beamline [5]. Some attempts at beamline alignment apply methods like genetic and differential evolution [6][7][8][9], attempt to match beamline data to an online model [10,11], or use families of commonly-used optimization algorithms [12]. These approaches are limited in that they give no guarantee of convergence to a global optimum.…”

“…See https://blueskyproject.io/12 Though Blop is not limited to Bluesky facilities and can be made to only "command" the experiments using only its "ask" and "tell" methods.…”

AI-driven real-time optics control system to achieve aberration-free coherent wavefronts at fourth-generation synchrotron radiation and free electron laser beamlines

“…Considering all factors above, given the intricacy of fourth-generation beamlines and their stringent performance criteria, manual techniques are often ineffective and frequently unfeasible. Instead, AI-powered auto-alignment for synchrotron beamlines has grown interest within the scientific community, explored through ML approaches [12,13] and optimization Algorithms [14]. The primary goal of these AI systems is to drastically reduce alignment time and to achieve and maintain a coherent focal spot size that adapts to the evolving sample conditions.…”

AutoFocus: AI-driven alignment of nanofocusing X-ray mirror systems

“…Given the complex and demanding nature of the fourth-generation synchrotron beamlines, manual approaches are considered inefficient and, in many cases, nearly impossible. Instead, emerging AI-driven auto-alignment control methods using ML [15,16] and optimization algorithms [17] are being developed. They aim to reduce alignment time, allow dynamic adjustments to the coherent focal spot size, and conserve valuable experimental time, marking a promising direction for the next generation of synchrotron radiation facility operations.…”