BESⅢ实验数据并行处理的研究

北京谱仪Ⅲ(BESⅢ)是北京正负电子对撞机Ⅱ(BEPCⅡ)上的一台大型通用磁谱仪,用于τ-粲物理实验研究。BESⅢ离线数据处理系统将实验获取的探测器原始数据转换为适合于物理分析的数据。主要研究了BESⅢ实验数据的并行计算处理。针对目前计算机的多核CPU硬件结构,采用多线程编程技术,实现了BESⅢ离线数据处理事例级别的多线程并行计算。性能测试表明多线程并行处理能够有效地提高运算速度,同时减少内存消耗。     

Real time algorithms for digital pulse processing applied to gamma-ray and hard X-ray spectroscopy

The duration of a single plasma discharge in the next generation of fusion experiments will be much longer than in the present devices. Storing all raw data acquired in each discharge will be more difficult and the high rates achieved by the new digitizers are already contributing to storage overload. This gap can be mitigated by real time (RT) analysis and compression, using devices such as field programmable gate arrays (FPGAs) capable to transfer and process data on the fly. However, to ensure a correct RT analysis, the FPGA algorithm must be adapted to the signal to be acquired. Since minor changes in signal shape may require significant algorithm modifications, it is important to know in advance the signal attributes. For that reason, the availability of more than one RT algorithm, especially during commissioning of new sub-systems and during campaigns with relevant changes in diagnostic conditions, is advantageous. This paper presents an implementation using two RT algorithms processing simultaneously, developed for the gamma-ray and hard X-ray diagnostics of the Joint European Torus (JET). Both algorithms perform pulse height analysis with pile-up rejection. While the first algorithm is suitable for Gaussian shaped pulses, the second is suitable for exponential signals. The algorithms are selectable by the user, during discharge configuration. Tests with radioactive sources made in JET are presented.     

γ相机交互式图像分析软件系统

本文概略介绍了我们研制的GCCS—89γ相机交互式图像分析系统。它包括系统管理、数据管理、设备管理和临床应用软件等几个部分。由于充分使用了模块化设计方法,系统有较大的灵活性,非常有利于系统的修改和维护。本系统临床程序十分丰富,可以满足不同医院的要求。     

同一物理量观测值的处理

在核数据评价和测量中,经常遇到同一物理量观测值的处理。当这些数据在统计学上一致时,很容易用统计方法处理,但存在系统误差时,处理就很困难。本文提出了一个普遍适用的方法。根据这个方法,由给出的系统误差的协方差矩阵,可计算平均值和它的误差。加权平均值、算术平均值、内误差、外误差和算术平均误差只是这个方法的特殊情况。给出的实例表明,这个方法是合理的。     

Design of FPGA based high-speed data acquisition and real-time data processing system on J-TEXT tokamak

Tokamak experiment requires high-speed data acquisition and processing systems. In traditional data acquisition system, the sampling rate, channel numbers and processing speed are limited by bus throughput and CPU speed. This paper presents a data acquisition and processing system based on FPGA. The data can be processed in real-time before it is passed to the CPU. It provides processing ability for more channels with higher sampling rates than the traditional data acquisition system while ensuring deterministic real-time performance. A working prototype is developed for the newly built polarimeter–interferometer diagnostic system on the Joint Texas Experimental Tokamak (J-TEXT). It provides 16 channels with 120 MHz maximum sampling rate and 16 bit resolution. The onboard FPGA is able to calculate the plasma electron density and Faraday rotation angel. A RAID 5 storage device is adopted providing 700 MB/s read–write speed to buffer the data to the hard disk continuously for better performance.     

Test bed for real-time image acquisition and processing systems based on FlexRIO,CameraLink, and EPICS

Image diagnostics are becoming standard ones in nuclear fusion. At present, images are typically analyzed off-line. However, real-time processing is occasionally required (for instance, hot-spot detection or pattern recognition tasks), which will be the objective for the next generation of fusion devices. In this paper, a test bed for image generation, acquisition, and real-time processing is presented. The proposed solution is built using a Camera Link simulator, a Camera Link frame-grabber, a PXIe chassis, and offers software interface with EPICS. The Camera Link simulator (PCIe card PCIe8 DVa C-Link from Engineering Design Team) generates simulated image data (for example, from video-movies stored in fusion databases) using a Camera Link interface to mimic the frame sequences produced with diagnostic cameras. The Camera Link frame-grabber (FlexRIO Solution from National Instruments) includes a field programmable gate array (FPGA) for image acquisition using a Camera Link interface; the FPGA allows for the codification of ad-hoc image processing algorithms using LabVIEW/FPGA software. The frame grabber is integrated in a PXIe chassis with system architecture similar to that of the ITER Fast Controllers, and the frame grabber provides a software interface with EPICS to program all of its functionalities, capture the images, and perform the required image processing. The use of these four elements allows for the implementation of a test bed system that permits the development and validation of real-time image processing techniques in an architecture that is fully compatible with that of the ITER Fast Controllers. This paper provides a specific example of a pattern search in a movie, its real-time implementation, and a performance analysis of the entire platform.     

Integrated data analysis on the electron temperature profile of HL-2A with the Bayesian probability inference method

Data analysis on tokamak plasmas is mainly based on various diagnostic systems, which are usually modularized and independent of each other. This leads to a large amount of data not being fully and effectively exploited so that it is not conducive to revealing the deep physical mechanism. In this work, Bayesian probability inference with machine learning methods have been applied to the electron cyclotron emission and Thomson scattering diagnostic systems on HL-2A/2M, and the effects of integrated data analysis (IDA) on the electron temperature of HL-2A with Bayesian probability inference are demonstrated. A program is developed to infer the whole electron temperature profile with a confidence interval, and the program can be applied in online analysis. The IDA results show that the full profile of the electron temperature can be obtained and the diagnostic information is more comprehensive and abundant with IDA. The inference models for electron temperature analysis are established and the developed programs will serve as an experimental data analysis tool for HL-2A/2M in the near future.     

基于Si-PIN探测器的电流型计数式α粒子测量

为解决基于Si-PIN探测器的传统计数型粒子测量系统在测量低强度脉冲辐射场时存在的计数率低和不能直接测量粒子时间信息的问题,提出了电流型计数测量方法。其原理是对电流型Si-PIN探测器在粒子入射时输出的脉冲信号进行线性放大和高速数字化,并使用数字信号处理算法进行数据分析以获得粒子测量信息。利用电流型Si-PIN探测器、电流型前置放大器和高带宽数字示波器组建了电流型计数测量系统,并在241 Am-244 Cm和239 Pu源上开展了α粒子测量实验。实现了脉冲幅度谱和粒子入射时间的联合测量,同时利用FIR及IIR型数字滤波器改善了能谱测量的能量分辨率。系统的最大计数率达2×106 s-1,适用于低强度脉冲辐射场测量。     

Evaluating parallel architectures for two real-time applicationswith 100 kHz repetition rate [hadron collider data]

In the context of research and development activities for future hadron colliders, competitive implementations of real-time algorithms for feature extraction have been made on various forms of commercial pipelined and parallel architectures. The algorithms used for benchmarking serve for decision making and are of relative complexity; they are required to run with a repetition rate of 100 kHz on data sets of kilobyte size. Results are reported and discussed in detail. Among the commercially available architectures, pipelined image processing systems can compete with custom-designed architectures. General-purpose processors with systolic mesh connectivity can also be used. Massively parallel systems of the SIMD type (many processors executing the same program on different data) are less suitable in the presently marketed form     

Experiment archive,analysis, and visualization at the National Ignition Facility

The National Ignition Facility (NIF) at the Lawrence Livermore National Laboratory is the world's most energetic laser, providing a scientific research center to study inertial confinement fusion and matter at extreme energy densities and pressures. A target shot involves over 30 specialized diagnostics measuring critical x-ray, optical and nuclear phenomena to quantify ignition results for comparison with computational models. The Shot Analysis and Visualization System (SAVI) acquires and analyzes target diagnostic data for display within a time-budget of 30 min. Laser and target diagnostic data are automatically loaded into the NIF archive database through clustered software data collection agents. The SAVI Analysis Engine distributes signal and image processing tasks to a Linux cluster where computation is performed. Intermediate results are archived at each step of the analysis pipeline. Data is archived with metadata and pedigree. Experiment results are visualized through a web-based user interface in interactive dashboards tailored to single or multiple shot perspectives. The SAVI system integrates open-source software, commercial workflow tools, relational database and messaging technologies into a service-oriented and distributed software architecture that is highly parallel, scalable, and flexible. The architecture and functionality of the SAVI system will be presented along with examples.     

虚拟仪器在核数据采集中的应用

介绍了一种利用虚拟仪器混合编程实现核数据采集与分析的可行方案。该方案采用并口进行数据传输，利用信号处理软件MATLAB对核谱数据进行分析，大大降低了谱分析软件设计中的难度。实现了信号的实时采集与分析功能。     

Development of advanced core noise monitoring system for BWRS

A BWR core noise monitoring system is developed for addressing core anomaly problems in future advanced core operation. In order to monitor in-core status from a limited number of signals, various up-to-date signal processing algorithms are introduced to compensate for a lack of information. These algorithms, such as independent component analysis, factor analysis and model based parameter estimation, are demonstrated to be effective through real plant data analysis to evaluate core and regional stability index, reactivity coefficients and core flow rate. Through these practices, we demonstrate that the core noise monitoring system is an effective general platform for providing a variety of monitoring tools to meet the requirements in future advanced core operation.     

ISTTOK plasma control with the tomography diagnostic

The presently available processing power in generic processing units (GPU) combined with state-of-the-art programmable logic devices enables the implementation of complex algorithms for plasma diagnostics in a real-time scenario.A tomography diagnostic based on three linear pin-hole cameras each with eight lines of sight has been developed for the ISTTOK tokamak. The plasma emissivity in a poloidal cross-section is computed locally on a sub-millisecond time scale, using a variant of the Fourier-Bessel algorithm. The output signals are then used for active plasma position control.The data acquisition and reconstruction system is based on ATCA technology and consists of one acquisition board with integrated FPGA capabilities and a dual-core Intel module running RTAI Linux.In this paper, the tomographic algorithm and some preliminary results of the real-time plasma position control are presented with a performance benchmarking against other available positioning diagnostics. The algorithm has shown to be accurate and the system has successfully controlled the plasma position during a plasma current reversal.     

辐射流体力学与中子输运并行数值模拟中的联接算法

复杂物理现象通常由多类复杂的物理过程紧耦合构成,其数值模拟也通常由适用不同物理过程的多类并行应用程序紧耦合完成。如何设计这些物理过程之间的联接算法,既要保证程序之间数据传递的高效,又要保证程序各自运行和总体模拟的高效,还要保证程序各自开发的独立,是一个值得研究的课题。文章基于广泛应用于高温高压物理研究中的辐射流体力学和中子输运多物理并行数值模拟,在非结构网格上,提出了两种联接算法:完全松散联接算法和两层紧耦合联接算法。前者侧重于实现程序各自运行的高效和开发的独立,后者在前者的基础上,还权衡了数据传递和总体模拟的高效。在两台并行机的数百个处理机上,通信复杂度分析和数值实验结果表明,两种算法均是有效的,可推广应用到其他多物理并行数值模拟。特别是,两层紧耦合联接算法是高效可扩展的,取得了近似最优的并行性能。     

1BP钚在线测量装置的研制及应用

为满足热实验在线监测的需要，根据热室工艺设备台架及热室的具体情况，设计加工安装了一套1BP钚在线测量装置，该在线测量装置主要由探测系统、光路系统、样品流通池系统和控制及数据处理系统组成。采用C语言开发了控制及数据处理程序，应用无线模块使仪器和计算机之间实现无线信号传输及远程控制和数据处理。采用Pu标准溶液刻度仪器，建立了1BP溶液中钚在线分析方法。工作曲线的线性范围为0.1～5.0 g/L，R²=0.999。用该测量装置测量0.5 g/L的Pu标准溶液，结果的相对标准偏差为0.67%。该测量装置的热实验应用结果表明，其可实时监测工艺运行状况，监测结果可用于指导各相液流配比的调节。     

Digital pulse processor for ion beam microprobe imaging

Capabilities of spectroscopic ion beam analysis (IBA) techniques that are available in ion microprobe facilities can be greatly improved by the use of digital pulse processing. We report here development of a digital multi parameter data acquisition system suitable for IBA imaging applications. Input signals from charge sensitive preamplifier are conditioned by using a simple circuit and digitized with fast ADCs. The digitally converted signals are processed in real time using FPGA. Implementation of several components of the system is presented.     

基于DIANE的BESⅢ分布式离线数据处理

北京谱仪Ⅲ(BESⅢ)是北京正负电子对撞机Ⅱ(BEPCⅡ)上用于τ-粲物理研究的大型通用磁谱仪。BESⅢ实验数据经过离线处理系统重建后才能用于物理分析。论文主要研究了BESⅢ分布式离线数据处理。基于DIANE软件,建立了一个分布式系统用于数据的事例级分布式处理。该系统在漂移室刻度的应用表明,性能能够满足BESⅢ实验的需求。     

SNAP 10A Data Reduction

On April 3, 1965 the world's first nuclear reactor space power system, SNAP I0A, was launched and operated successfully for forty-three days in orbit. To reduce and analyze the flight test data, data handling procedures and data analysis computer programs representing the end results of three years of effort and improvement of similar data systems for SNAP 10A ground tests were used. This paper describes the evolution of data handling and reduction from the manual methods used for the first non-nuclear system test to later ground tests where data loggers of increasing degrees of sophistication were used to generate punched paper tape which was then computer processed, reduced, and analyzed, and finally to the completely electronic and computerized reduction of flight test telemetry data. The equipment used, handling procedures developed, and computer programs are briefly described. Improvements in each generation of data system resulting from experience and improved equipment are emphasized. As a result of the experience in semi-automatic data collection, reduction, and analysis gained in the SNAP 10A program recommendations for creation and improvement of similar data systems are given, and some of the more important pitfalls encountered are discussed. Finally, the more significant differences between flight and ground acquisition and reduction are noted.     

Management of complex data flows in the ASDEX Upgrade plasma control system

Establishing adequate technical and physical boundary conditions for a sustained nuclear fusion reaction is a challenging task. Phased feedback control and monitoring for heating, fuelling and magnetic shaping is mandatory, especially for fusion devices aiming at high performance plasmas. Technical and physical interrelations require close collaboration of many components in sequential as well as in parallel processing flows. Moreover, handling of asynchronous, off-normal events has become a key element of modern plasma performance optimisation and machine protection recipes.The manifoldness of plasma states and events, the variety of plant system operation states and the diversity in diagnostic data sampling rates can hardly be mastered with a rigid control scheme. Rather, an adaptive system topology in combination with sophisticated synchronisation and process scheduling mechanisms is suited for such an environment. Moreover, the system is subject to real-time control constraints: response times must be deterministic and adequately short.Therefore, the experimental tokamak device ASDEX Upgrade employs a discharge control system DCS, whose core has been designed to meet these requirements. In the paper we will compare the scheduling schemes for the parallelised realisation of a control workflow and show the advantage of a data-driven workflow over a managed workflow. The data-driven workflow as used in DCS is based on signals connecting process outputs and inputs. These are implemented as real-time streams of data samples. Consequently, real-time signal management forms the foundation of DCS. The paper explains the principal features such as tagged samples, signal groups, algorithmic blocks and processes as well as scheduling schemes which allow DCS control applications to be defined as self-contained modular building blocks glued together by a software framework.By virtue of this sound foundation, DCS is a mature but still evolving system for reliable, distributed control of an entire tokamak device coordinating and monitoring 20 diagnostic systems, 14 magnetic power supplies, 5 heating systems with a total power of more than 25 MW, 8 gas fuelling channels, a pellet injector and a killer gas gun.     

Automatically processing physical data from LHD experiments

Physical data produced by large helical device (LHD) experiments is supplied by the Kaiseki server, and registers more than 200 types of diagnostic data. Dependencies exist amongst the data; i.e., in many cases, the calculation of one data requires other data. Therefore, to obtain unregistered data, one needs to calculate not only the diagnostic data itself but also the dependent data; however, because the data is registered by different scientists, each scientist must separately calculate and register their respective data. To simplify this complicated procedure, we have developed an automatic calculation system called AutoAna. The calculation programs of AutoAna are distributed on a network, and the number of such programs can be easily increased dynamically. Our system is therefore scalable and ready for substantial increases in the size of the target data.