Magnet Power Supply Research Articles

Research on a high power density mechanical-electrical-hydraulic regenerative suspension system for high-speed tracked vehicles

High power density energy regeneration is one of the effective solutions to solve the contradiction between improving the damping performance and energy consumption of active suspension. The hydraulic commutator is used to realize hydraulic rectification and hydraulic variable speed/pump/motor with few teeth difference gear pairs is used to match the speed, combined with permanent magnet motor power generation and power supply to put forward kilowatt level high power density mechanical-electrical-hydraulic regenerative suspension system for high-speed tracked vehicles. The mathematical model and fluid-solid-thermo-magnetic multiphysics coupling model are built to analyze the damping performance and regenerative characteristics of the system under passive and semi-active working conditions. The simulation results show that the damping force of the system increases with the increase of the road excitation amplitude and the semi-active control can be realized by adjusting the duty cycle with the PWM control rectifier module. The high power density mechanical-electrical-hydraulic regenerative suspension system can realize kilowatt level energy regeneration, and the regenerative efficiency is more than 50% under low-frequency excitation. The temperature rise of the system is low during operation, which is helpful to improve the reliability and service life.

Real-time artificial intelligence for accelerator control: A study at the Fermilab Booster

We describe a method for precisely regulating the gradient magnet power supply (GMPS) at the Fermilab Booster accelerator complex using a neural network trained via reinforcement learning. We demonstrate preliminary results by training a surrogate machine-learning model on real accelerator data to emulate the GMPS, and using this surrogate model in turn to train the neural network for its regulation task. We additionally show how the neural networks to be deployed for control purposes may be compiled to execute on field-programmable gate arrays (FPGAs), and show the first machine-learning based control algorithm implemented on an FPGA for controls at the Fermilab accelerator complex. As there are no surprise latencies on an FPGA, this capability is important for operational stability in complicated environments such as an accelerator facility.

Research on Short Circuit Operation Control Strategy of a Superconducting Magnet Power Supply for CRAFT

Research on Short Circuit Operation Control Strategy of a Superconducting Magnet Power Supply for CRAFT

Efficiency, Power Loss, and Power Factor Measurement of Quadrupole Magnet Power Supplies at the Spallation Neutron Source

Efficiency, Power Loss, and Power Factor Measurement of Quadrupole Magnet Power Supplies at the Spallation Neutron Source

Developing Robust Digital Twins and Reinforcement Learning for Accelerator Control Systems at the Fermilab Booster

We describe the offline machine learning (ML) development for an effort to precisely regulate the Gradient Magnet Power Supply (GMPS) at the Fermilab Booster accelerator complex via a Field-Programmable Gate Array (FPGA). As part of this effort, we created a digital twin of the Booster-GMPS control system by training a Long Short-Term Memory (LSTM) to capture its full dynamics. We outline the path we took to carefully validate our digital twin before deploying it as a reinforcement learning (RL) environment. Additionally, we demonstrate the use of a Deep Q-Network (DQN) policy model with the capability to regulate the GMPS against realistic time-varying perturbations.

BOOSTR: A Dataset for Accelerator Control Systems

The Booster Operation Optimization Sequential Time-series for Regression (BOOSTR) dataset was created to provide a cycle-by-cycle time series of readings and settings from instruments and controllable devices of the Booster, Fermilab’s Rapid-Cycling Synchrotron (RCS) operating at 15 Hz. BOOSTR provides a time series from 55 device readings and settings that pertain most directly to the high-precision regulation of the Booster’s gradient magnet power supply (GMPS). To our knowledge, this is one of the first well-documented datasets of accelerator device parameters made publicly available. We are releasing it in the hopes that it can be used to demonstrate aspects of artificial intelligence for advanced control systems, such as reinforcement learning and autonomous anomaly detection.

A high-power and high-stability power supply design

BEPCII is a high-performance collider with a design energy of 2.1 GeV. In order to obtain more physical experiments and synchrotron radiation applications, BEPCII decided to carry out energy upgrading to 2.5 GeV. Insufficient energy of special magnet power supplies in collision zone is a key factor restricting BEPCII's operation under 2.5 GeV energy. Therefore, three high-power and high-stability power supplies are designed for BEPCII energy upgrading. The designed power supplies output part adopts high-frequency chopper modules parallel connected and adopts digital-analog hybrid control methods. The voltage loop regulator uses high-speed analog control, and current loop regulator uses digital control which is based on FPGA. The experimental data show that the output current stability of the developed power supplies is lower than 50 ppm. The output current resolution is lower than 70 ppm. The developed power supplies can well meet the design requirements of BEPCII energy upgrading.

Application of the hybrid genetic particle swarm algorithm to design the linear quadratic regulator controller for the accelerator power supply

The purpose of this paper is to study a new method to improve the performance of the magnet power supply in the experimental ring of HIRFL-CSR. A hybrid genetic particle swarm optimization algorithm is introduced, and the algorithm is applied to the optimal design of the LQR controller of pulse width modulated power supply. The fitness function of hybrid genetic particle swarm optimization is a multi-objective function, which combined the current and voltage, so that the dynamic performance of the closed-loop system can be better. The hybrid genetic particle swarm algorithm is applied to determine LQR controlling matrices Q and R. The simulation results show that adoption of this method leads to good transient responses, and the computational time is shorter than in the traditional trial and error methods. The results presented in this paper show that the proposed method is robust, efficient and feasible, and the dynamic and static performance of the accelerator PWM power supply has been considerably improved.

Research on power supply of booster based on PWM rectifier

The magnet power supply system of HEPS mainly includes the linear magnet power supply, the booster power supply, the transportation line magnet power supply and the storage ring magnet power supply. According to the physical design requirements, the output current of the booster magnet power supply is a current with a DC bias period of 1 Hz, and the waveform can be a sine wave or a trapezoidal wave (Jiang et al. Sci Sin Phys Mech Astron 44(10):1075–1094, 2014). The dipole magnet of the booster should be powered in series, the inductance value is relatively large (about 160 mH), and the power of the power supply reaches several hundred kilowatts. The three-phase PWM rectifier controls the orderly turn-on of power electronic devices and takes the input voltage phase of the power grid as the command current phase to make the input current waveform follow the power grid voltage, so as to achieve the purpose of improving the power factor. In this paper, the PWM rectifier is modeled and simulated, and based on the simulation, an experimental prototype is built. The experimental results of the power supply prototype of the PWM rectifier show that the PWM rectifier can achieve high power factor operation on the grid side (Akagi and Sato, in: Control and performance of a flywheel energy storage system based on a doubly-fed induction generator–motor for power conditioning, 1999). The PWM rectifier can be used in the power supply of the booster and has a very good application prospect.

Design of control algorithm for accelerator magnet power supply with large time constant load

High Energy Photon Source-Test Facility (HEPS-TF) is a pre-research project for the construction of high energy synchrotron radiation source in the 12th five-year plan period. The purpose is to research the key technology and develop the key equipment of high energy synchrotron radiation source. Superconducting 3W1 magnet is the first self-developed superconducting wiggler magnet in China, and it is also one of the key research topics of HEPS. The author has completed a new digital closed-loop control algorithm for the superconducting 3W1 magnet with large load time constant and the nonlinear characteristics of inductance increasing with current, namely three-branch structure algorithm. In the face of the rapid development of high energy accelerator technology, the application of intelligent technology has become an inevitable development trend in the field of accelerator magnet power supply technology. Although the digital control of accelerator magnet power supply has been widely used, it is the first time to apply the new closed-loop control algorithm to realize fast adjustment and precision tracking in accelerator superconducting magnet power supply in China. According to the nonlinear characteristics of inductance and output current of superconducting magnet, a new digital closed-loop control algorithm for the load of superconducting magnet power supply with large time constant is proposed. This algorithm is quite different from the traditional algorithm and can attain the independent tracking and adjustment of the control target. Finally, by testing the ripple, error and stability of superconducting 3W1 magnet power supply, the correctness, practicability and reliability of power supply system as well as the digital control algorithm are verified. The results provide a new idea for the control of accelerator magnet power supply.

The design of power supply for HF MRI superconducting magnet

High-field (HF) magnet resonance imaging (MRI) is designed as a large aperture human body imaging system. The superconducting magnet provides magnetic field B0 for HF MRI. The power supply is a key component, provides excitation and demagnetization functions for superconducting magnet. According to the requirements, the superconducting magnet power supply needs to be designed and developed with small volume and high reliability and bidirectional energy flow. A combination topology is proposed, including a pulse-width-modulation (PWM) rectifier, a dual-active-bridge (DAB), and a H-bridge. In this paper, the principle of the combination topology is introduced briefly. Meanwhile, the analysis and test results of a PWM rectifier based on hybrid active-neutral-point-clamped (ANPC) three-level are given.

Power Supplies of Correction Magnets of Modern Accelerating Complexes

The Budker Institute of Nuclear Physics has been designing two accelerator facilities of the MegaScience class [1, 2]. It is planned to use the same bipolar power supplies sources for corrector magnets in both projects. These power supplies must provide an output current of 6 and 12 A and output voltage up to 100 V. Identical power-supply sources will be used both for slow orbit correction and for fast orbit feedback. As a result, the power-supply sources must provide the high stability of output current (at least 0.01%) and adequate bandwidth (at least 1k Hz). The project must allow for the fast replacement of a failed power-supply source.

Evaluation of the Backup Signal-processing System of the KSTAR Quench Detection System

Abstract The Quench Detection System (QDS) of the Korea Superconducting Tokamak Advanced Research (KSTAR) has been operated to prevent from overheating the superconducting coil system for 11 years. The QDS discriminates a normal voltage of ∼100 mV on NbTi or Nb3Sn cable-in-conduit conductors in the event of quench, while the Poloidal Field coils are applied with voltages of up to some kV by pulsed operation of the magnet power supply system. Three VME systems (VMEs) are used to compensate remaining induced voltage in the quench detection signals; however, products of the VMEs are probably discontinued on the global markets before the end of the KSTAR project. Accordingly a Backup Signal-processing System (BSS) is being developed and tested by mostly using generic and reliable commercial-off-the-shelf devices. The BSS is designed to replicate all the logical functions of the VMEs, and conducts quench detection by using FPGAs; whereas, the VMEs are using CPUs. Both the VMEs and the BSS simultaneously operate to detect quench, and quench alarms are voted by 1-out-of-2 logic to generate interlock signals in normal operation. A subset of the BSS integrated with the QDS has demonstrated expected performances in the KSTAR campaign 2018.

Research on protection system of resonant network in CSNS magnet power supplies

Chinese Spallation Neutron Source (CSNS) is one of the key projects of China, in which the main magnet power supplies adopt the structures of resonant networks (Wang et al. in Sci China Phys Mech Astron 54(Suppl 2):239–244, 2011). Since each network contains a lot of components, and the working state of all the components will affect the whole accelerator, it is necessary to monitor the operation of each part in the resonant network. In this protections system, a supervision to 314 switch signals and 118 analog signals is realized using a chip of field programmable gate array (FPGA). The details about the hardware and software designs are introduced in this paper. This protection system has been used in CSNS project, and the result proves that the RCS has been well monitored by this system. This system is a self-developed protection system according to the engineering needs of CSNS. It can realize the real-time monitoring and protection of the working state of resonance network by processing a large number of data using FPGA.

Application of Magnetic Coupling Resonant Wireless Power Supply in a Torque Online Telemetering System of a Rolling Mill

The torque of the main drive system is one of the most important force and energy parameters of the rolling mill, and the strain type torque online telemetry system is a practical method for measuring torque parameters. The strain gauges and transmitters on the rotating shaft are driven by a high-frequency induction power system. Their installation, debugging, and maintenance are cumbersome, and their very low power transmission efficiency (PTE) has become a challenging problem for online telemetry systems. In this paper, a magnetic coupling resonant wireless power supply method is utilized to replace the high-frequency induction power supply. Through theoretical and experimental research, it is concluded that the magnetically coupled resonant wireless power supply method can realize long-distance power supply, overcoming many shortcomings of the high-frequency induction power supply system. In the laboratory, a maximum PTE of 83.7% is obtained for a transmission distance of 50 mm. Under the influence of environmental factors common for the transmission shaft of a rolling mill, the PTE decreases by 34%, but normal operation of the system can be achieved by adjusting the compensation capacitance. The proposed system provides a guarantee for a long-term stable power supply on the measured axis of a rolling mill torque online telemetry system.

Reducing Fluctuations in Slow-Extraction Beam Spill Using Transit-Time-Dependent Tune Modulation

Slow extraction of beam from synchrotrons or storage rings as required by many fixed target experiments is performed by controlled excitation and feeding of a structural lattice resonance. Due to the sensitive nature of this resonant extraction process, the temporal structure of the extracted beam is modulated by the minuscule current fluctuations present on the quadrupole magnet power supplies. Such a modulation lead to pile-ups in detectors and a significant reduction in accumulated event statistics. This contribution proposes and experimentally demonstrates that by an introduction of further modulation on quadrupole currents with a specific amplitude and frequency, the inherent power supply fluctuations are mitigated leading to a smoothening of the beam temporal structure. The slow extraction beam dynamics associated with this method are explained along with the operational results.

The Preliminary Design of Large-Scale High-Intensity DC and Power Frequency Magnetic Field Immunity Test Platform

The electromagnetic interference caused by the strong stray magnetic field generated by a tokamak device and its coil power supply will greatly threaten the reliable operation of electrical and electronic devices nearby, and a magnetic field immunity test is the most commonly used method to qualify the sensitive equipment and ensure the reliability of the system. Thus, building a high-intensity magnetic field immunity test platform is essential for the stable operation of a large tokamak device in the future. This article presents the preliminary design of a large-scale high-intensity magnetic field immunity test platform, including dc static test platform, dc transient test platform, and power frequency test platform. Referring to the actual magnetic field environment around the International Thermonuclear Experimental Reactor (ITER) tokamak device and its magnet power supply, the design parameters of the subplatforms are determined. Then the scheme design of the subplatforms and the integration design of the whole platform are briefly introduced. As a part of the large-scale superconductor rest facility (LSTF) in Chinese Academy of Science, Institute of Plasma Physics (ASIPP), the design takes full advantages of the other devices planned in the facility. This test platform will be the largest in the world and enable the test of various equipment under different magnetic field environments, and more importantly, play an important role in the reliable operation of China Fusion Engineering Test Reactor (CFETR) and future large tokamak devices.

Preliminary Design of Power Supply for Water-Cooled Resistive Magnet Based on Three-Level Buck Converters

Water-cooled resistive high-field magnets require a high power supply with an extremely high current stability, and a low current ripple and noise. The conventional resistive magnet power supply uses thyristor rectifiers, followed by a passive filter and an active filter to reduce the voltage ripple. This Paper describes a new circuit topology design of the resistive magnet power supply, which consists of uncontrolled rectifier bridges and three-level buck converters in series instead of the conventional thyristor rectifiers. The three-level buck converters work at a higher switching frequency, which can greatly reduce the cost, volume and weight of the filter. According to simulation and analysis results, it is seen that the three-level buck converter structure provides a lower voltage and current ripple than the conventional thyristor rectifiers, reduces the voltage stress of the power electronic devices. Beside, this new design can achieve a higher power factor than the power supply based on thyristor rectifiers.

Design and Implementation of a SoC Embedded Waveform Acquisition System for Surveillance and Diagnostics of Pulsed Magnet Power Supplies

Pulsed magnet power supply systems used for electron beam injection and extraction have been developed for the Taiwan Photon Source (TPS) and the Taiwan Light Source (TLS). Control systems for these supplies have been implemented with well-functioning operations interfaces. To accomplish higher reliable operation, advanced real-time diagnostic toolkits for pulsed magnet power supplies had to be developed. A SoC (System-on-Chip) embedded waveform acquisition system has been designed, implemented and applied to inspect pulsed magnet power supplies during routine operation. This waveform acquisition system not only operates at the 125 MS/s sample rate, 50 MHz bandwidth and 14 bits resolution, but supports also the EPICS software framework for complete system integration. The acquired current waveforms can be immediately extracted specific characteristics to examine the status of pulsed magnet power supplies. In this paper, we summarize the design, implementation and real-time data analysis of the SoC embedded waveform acquisition system for the pulsed magnet power supplies.

A precision ADC sampling system design

The high-energy photon source, which has been built in Huairou, Beijing, has high requirements on magnetic field dithering. Magnetic field dithering is mainly determined by the stability of the output current of the power supply. In order to ensure the stability of the output current of quadrupole magnet power supply, the power supply sampling control loop needs to be precisely designed. In this paper, a precision ADC sampling system based on internal temperature control is designed to carry out precise control of the sampling ADC. In this design, precise ADC chip is used to complete the precise sampling of the system. The precise sampling system contains a DAC system for high-speed settings. In order to verify the design of the system, high-precision quadrupole magnet power supply is used for measurement. The experimental results show that the temperature variation range of precision temperature control ADC system is ± 0.1 °C. By using the precise temperature control ADC system, the output current stability of the high-precision quadrupole magnet power supply is effectively improved.