In this work, a Preisach-recurrent neural network model is proposed to predict the dynamic hysteresis in ARMCO pure iron, an important soft magnetic material in particle accelerator magnets. A recurrent neural network coupled with Preisach play operators is proposed, along with a novel validation method for the identification of the model’s parameters. The proposed model is found to predict the magnetic flux density of ARMCO pure iron with a Normalised Root Mean Square Error (NRMSE) better than 0.7%, when trained with just six different hysteresis loops. The model is evaluated using ramp-rates not used in the training procedure, which shows the ability of the model to predict data which has not been measured. The results demonstrate that the Preisach model based on a recurrent neural network can accurately describe ferromagnetic dynamic hysteresis when trained with a limited amount of data, showing the model’s potential in the field of materials science.

The precise knowledge of the magnetic field produced by dipole magnets is critical to the operation of a synchrotron. Real-time measurement systems may be required, especially in the case of iron-dominated electromagnets with strong non-linear effects, to acquire the magnetic field and feed it back to various users. This work concerns the design and implementation of a new measurement system of this kind currently being deployed throughout the European Organization for Nuclear Research (CERN) accelerator complex. We first discuss the measurement principle, the general system architecture and the technology employed, focusing in particular on the most critical and specialized components developed, that is, the field marker trigger generator and the magnetic flux integrator. We then present the results of a detailed metrological characterization of the integrator, including the aspects of drift estimation and correction, as well as the absolute gain calibration and frequency response. We finally discuss the latency of the whole acquisition chain and present an outline of future work to improve the capabilities of the system.

Sensing coils are inductive sensors commonly used to measure magnetic fields, such as those generated by electromagnets used in many kinds of industrial and scientific applications. Inductive sensors rely on integrating the output voltage at the coil’s terminals in order to obtain flux linkage, which may suffer from the magnification of low-frequency noise resulting in a drifting integrated signal. This article presents a method for the cancellation of integrator drift. The method is based on a first-order linear Kalman filter combining the data from the coil and a second sensor. Two case studies are presented. In the first one, the second sensor is a Hall probe, which senses the magnetic field directly. In a second case study, the magnet’s excitation current was used instead to provide a first-order approximation of the field. Experimental tests show that both approaches can reduce the measured field drift by three orders of magnitude. The Hall probe option guarantees, in addition, one order of magnitude better absolute accuracy than by using the excitation current.

Rotating-coil measurement systems are widely used to measure the multipolar fields of particle accelerator magnets. This paper presents a rotating-coil measurement system that aims at providing a complete data set for the characterization of quadrupole magnets with small bore diameters (26 mm). The PCB magnetometer design represents a challenging goal for this type of transducer. It is characterized by an aspect ratio 30% higher than the state of the art, imposed by the reduced dimension of the external radius of the rotating shaft and the necessity of covering the entire magnet effective length (500 mm or higher). The system design required a novel design for the mechanical asset, also considering the innovation represented by the commercial carbon fiber tube, housing the PCB magnetometer. Moreover, the measurement system is based primarily on standard and commercially available components, with simplified control and post-processing software applications. The system and its components are cross-calibrated using a stretched-wire system and another rotating-coil system. The measurement precision is established in a measurement campaign performed on a quadrupole magnet characterized by an inner bore diameter of 45 mm.

A full-fledged neural network modeling, based on a Multi-layered Nonlinear Autoregressive Exogenous Neural Network (NARX) architecture, is proposed for quasi-static and dynamic hysteresis loops, one of the most challenging topics for computational magnetism. This modeling approach overcomes drawbacks in attaining better than percent-level accuracy of classical and recent approaches for accelerator magnets, that combine hybridization of standard hysteretic models and neural network architectures. By means of an incremental procedure, different Deep Neural Network Architectures are selected, fine-tuned and tested in order to predict magnetic hysteresis in the context of electromagnets. Tests and results show that the proposed NARX architecture best fits the measured magnetic field behavior of a reference quadrupole at CERN. In particular, the proposed modeling framework leads to a percent error below 0.02% for the magnetic field prediction, thus outperforming state of the art approaches and paving a very promising way for future real time applications.