AI 中文总结
本研究将卷积神经网络与Soft-Actor-Critic强化学习智能体结合，实现CERN PS纵向三重分裂的自动化优化，其2025年部署的自主控制器为CERN注入器复合体首批强化学习束流质量优化系统之一。
AI 中文摘要
欧洲核子中心质子同步加速器（CERN PS）中的纵向三重分裂是定义交付给大型强子对撞机（LHC）的25 ns束团间距的关键射频（rf）操作。我们提出了一种基于机器学习的该操作的自动化优化方法。在PS中执行多个 revolution 频率谐波的射频系统的连续操作，从PS Booster（PSB）注入的每个束团被分裂为12个具有理想相同纵向束流参数的束团。需要精确的射频电压和相位设置以最小化强度、纵向发射度和束团形状的逐束团变化。我们的设置结合了两个不同部分：卷积神经网络，其根据分裂过程中纵向束团轮廓的演化提供初始相位校正；以及两个顺序的Soft-Actor-Critic（SAC）强化学习智能体，用于优化腔体相位和电压。这些模型在由模拟不确定性（包括从测量评估的噪声）增强的束流纵向动力学（BLonD）跟踪模拟数据上进行训练，以实现训练并稳健迁移至实际机器。2022年的首次测试在平均少于10个优化步骤内达到目标分裂质量，匹配或优于手动调整。这促成了按需版本的操作部署，随后于2025年3月推出了完全自主控制器。该控制器自那时起已可用于运行，代表了CERN注入器复合体中首批用于束流质量优化的基于强化学习的系统之一。
英文摘要
The longitudinal triple splitting in the CERN Proton Synchrotron (PS) is a key rf manipulation defining the 25 ns bunch spacing delivered to the Large Hadron Collider (LHC). We present an automated optimization of this manipulation based on machine learning. Successive manipulations with rf systems at multiple harmonics of the revolution frequency are performed in the PS. Each bunch injected from the PS Booster (PSB) is split into twelve bunches with ideally identical longitudinal beam parameters. Precise rf voltage and phase settings are required to minimize bunch-by-bunch variations in intensity, longitudinal emittance, and bunch shape. Our setup combines two distinct parts: a convolutional neural network providing an initial phase correction from the evolution of longitudinal bunch profiles during the splitting process, and two sequential Soft-Actor-Critic (SAC) reinforcement-learning agents that refine cavity phases and voltages. The models are trained on data from Beam Longitudinal Dynamics (BLonD) tracking simulations augmented by simulated uncertainties, including noise evaluated from measurements, to enable training and robust transfer to the machine. First tests in 2022 reached target splitting quality in fewer than ten optimization steps on average, matching or outperforming manual adjustments. This led to operational deployment of an on-demand version, followed by a fully autonomous controller in March 2025. This controller has been available to operations since, representing one of the first reinforcement-learning-based systems for beam quality optimization deployed in the CERN injector complex.
Comments15 pages, 14 figures, pre-print
