多核处理器在中低压保护测控一体化装置中的应用

阐述了多核处理器在电力二次保护测控装置中的应用。结合多核处理器基本知识和电力二次设备的典型硬件系统方案,详细描述了某嵌入式多核处理器在新一代中低压保护测控一体化装置中的设计与实现,以及基于消息机制的多核处理器间数据传输共享的方法。最后,对多核处理器在电力二次设备中的应用趋势进行了展望。     

Parallel computing

Parallel processing is the harnessing of multiple processors to work on the same problem. The aim is to speed up the computational process, ideally by the number of processors used. Parallel processing is increasingly emerging as the key to very-high-speed computation. This article is an introductory overview emphasizing the use of parallel computers for computational engineering     

Graphical programming harnesses multicore PC technology

Multicore processing is generating considerable buzz with in the PC industry,largely because both Intel and AMD have released initial versions of their own multicore processors.These first multicore processors contain two cores,or computing engines,located in one physical processor-hence the name dual-core processors.     

一种新型快速瞬态响应VRM拓扑

微处理器的迅速发展,使其对功率管理的要求越来越高。作为CPU电源的电压调节模块(VRM),其瞬态响应已成为目前研究的热点问题。提出了一种新型快速瞬态响应VRM拓扑,分析了该拓扑的结构、工作原理和控制策略。通过一台1.8 V/15 A的样机实验,证明了该拓扑在提高瞬态响应方面的有效性。     

Transients computation for relay testing in real-time

This paper discusses characteristics of a new digital simulator for protection relay testing. The most demanding design requirement is computation of fault transients under the condition of real-time change of power system configuration due to relay operation. This problem is solved using EMTP computational techniques enhanced with novel numerical solutions for dynamic power system configuration change and nonlinear element modeling. An advanced computer architecture is utilized to achieve further optimization of the execution time for the transients computation code. The main advantage of this design is the use of conventional single processor computer architecture in combination with advanced digital signal processors (DSPs). This makes this simulator an off-the-shelf product with all the benefits of commercially available computers priced at a relatively low cost     

位流运算研究及应用

提出了运用数字电路技术直接处理一位S-D编码位流信号的新方法。首先，提出了处理位流编码的2个基本模块：位流加法器和乘法器；然后，利用基本模块可以方便组成积分器、滤波器等电路。研究了位流加法器和乘法器的基本原理，提出了可实现电路。采用位流信号直接处理器，可以避免在位流信号和多位信号之间的转换；其次，可以有效地节约信号的物理布线；最后，位流处理器具有比Nyquist采样率处理器更少的硬件电路资源。直流电机PI控制仿真实验表明：该电路可行、具有很高的效率。     

双口RAM在DSP与单片机数据通信中的应用

在多处理器的应用系统中,处理器之间的有效通信是系统能否协同工作的关键。多处理器之间的通信方法有很多种,以双口RAM作为共享存储器来实现2个处理器之间的通信具有实时性好、接口电路简单、数据传输量大等优点。文中介绍了双口RAM内部结构与功能,以及用它来实现DSP与单片机之间通信的接口电路与通信流程,此接口电路同样适合其他类型的多处理器间的通信。     

An application of neural network to dynamic dispatch using multiprocessors

This paper presents an application of neural networks to dynamic dispatch. The proposed method uses a neural network with appropriate noises and can give efficient initial neuron conditions which are specific to the problem. Therefore, convergence to a local minimum can be suppressed. The method is implemented on a transputer, that is one of the efficient parallel processors, and the appropriate number of processors is examined. It can develop optimal and feasible generator output trajectories quickly by applying forecasts of system load patterns to practical thermal generating unit systems     

Computing - Two good to be true - Multicore microprocessors may be great for doing lots of things at once, but what about doing one thing more quickly?

Microprocessor makers have come up with a design that combines two or more independent processors on one chip. Most PC programs, however, cannot take advantage of additional processors since they have been written to run on a single processor. Despite this, programs have still been getting more powerful because of improved processors. Organizing software so that blocks of code in one program can be spread across many processors and run at the same time with increased performance may address this issue. This approach, which has been carried over into transactional memory, gives an operating system more freedom to break a program down into blocks that can be more freely distributed across the processing resources available     

Asynchronous distributed state estimation based on a continuous‐time stochastic model

In this paper, the problem of state estimation in an asynchronous distributed multi‐sensor estimation (ADE) system is considered. In such an ADE system, the state of a plant of interest is estimated by a group of local estimators. Each local estimator based, for example, on a Kalman filter, performs fusion of data from its local sensor and other (remote) processors to compute possibly best state estimates. In performing data fusion, however, two important issues need to be addressed, namely, the problem of asynchronism of local processors and the one of unknown correlation between asynchronous data in local processors. Consequently, there are two main contributions proposed in this paper. The first is a method to deal with asynchronous discrete‐time data based on a continuous‐time stochastic plant model. The second contribution is an asynchronous distributed data‐fusion algorithm. Simulated experiments illustrate the effectiveness of the proposed ADE approach. Copyright © 2011 John Wiley & Sons, Ltd.     

High-order volterra series analysis using parallel computing

This paper addresses the determination of high-order Volterra transfer functions of non-linear multiport networks containing multidimensional non-linear elements. A new method is developed for utilizing parallel computing which is very efficient for approximately 10–20 processors. The utilization of each processor may be as high as 80%–95%. The developed program is very flexible as it is ANSI C++ compatible and may run on both single- and multiprocessor computers. Using about eight processors it is possible to analyse rather complicated non-linear circuits up to ninth order in a few hours. © 1997 by John Wiley & Sons, Ltd.     

A high‐speed MIMO FFT processor with full hardware utilization

The conventional way to design multi‐input‐multi‐output (MIMO) fast Fourier transform (FFT) processors for MIMO‐orthogonal frequency division multiplexing systems is to adopt a parallel architecture which uses as many single‐input‐single‐output FFT processors as the number of transmit/receive antennas. These MIMO FFT processors can provide high throughput, but they perform with low hardware utilization when there are not all input sequences available. In this paper, we propose a high‐speed MIMO FFT processor which can work efficiently with high throughput and full hardware utilization for variable 1 to 4 input sequences. Our MIMO FFT processor is designed by reordering and distributing data sequences to all data paths and is constructed by some novel modules. Being synthesized by using UMC 0.18‐μm process demonstrates that our 64‐point 4 × 4 FFT can achieve high throughput with full hardware utilization and perform correctly up to 62.25 MHz with low power consumption for variable 1 to 4 input sequences.     

基于双处理器结构的输电线路监测终端

鉴于当前输电线路监测设备终端未能完全适应电力传输质量监督管理需求,为了提高监测系统的稳定性,介绍了一种基于双微控制器(双MCU)结构的监测终端,其双MCU既能协同工作,又能各司其职,大大降低了系统的耦合度。当系统中一个MCU负责的部分功能异常时,另一个MCU不会受到影响,同时还能及时发现和自动排除故障。实验表明,当以稳定性为目标时,双MCU的结构具有明显的优势。     

双数字信号处理的电流继电保护和测控装置的研制

根据微机保护系统的快速、准确的发展趋势和DSP数字信号处理芯片的特点，设计并实现了一种双数字信号处理芯片结构的电流继电保护和测控装置，提高了数据处理速度。采用双端口随机存储器实现两处理器的通信。通过每周波多点数快速傅立叶算法计算，实现对配电线路实时高精度测量、快速谐波分析等功能。微机保护算法采用了具有数字滤波功能的24点离散傅立叶变换算法，消除高次谐波的干扰，提高了电流保护的选择性和可靠性。     

Concurrent boundary element computation

A major limitation of the boundary element method (BEM) for the solution of electrical potential problems is the long computational time required. However, a large portion of the calculations involved can be viewed as being parallel in nature and can therefore be computed concurrently. This paper makes an effort to increase the efficiency of the BEM process using transputer-based multiprocessor computing techniques. The algorithms developed may equally well be applied to any multiprocessor system. The application selected to demonstrate the technique is the solution of an electrostatic problem governed by a two-dimensional Laplace equation. A parallel algorithm for problem setup and field extraction using BEM is designed and implemented on a transputer array. Special attention is directed to the utilization of the parallel processors to achieve maximum efficiency. The analysis in this work concentrates on the communication strategies for passing data between processors as well as a consideration of the workload attributed to each processor. The parallel algorithms were implemented using 3L Parallel Fortran; however, the choice of topology for the overall BEM implementation was limited by the fact that certain parts of the algorithm could only utilize a pipeline configuration of processors. Comprehensive results for the parallel BEM algorithm are given and they are encouraging, indicating that parallel processing has much to offer when applied to the boundary element method.     

提高预处理共轭梯度法计算大型电网潮流时并行性能的方法

研究了如何提高预处理共轭梯度(preconditioned conjugate gradient,PCG)法计算大网络交流潮流时的并行性能,提出了一种新的并行处理方法——并行节点分配法,即将节点导纳矩阵和节点出力的数据以节点为单位分派给各个处理器,在各个处理器中完成余下的计算处理部分。算例分析表明,在使用PCG法进行大网络交流潮流计算时,并行节点分配法在并行性能上具有一定的优势,进而为PCG法在大网络交流潮流并行计算中的应用提供了可借鉴的经验。     

基于Modbus/TCP的火电厂实时数据集成及网络通信控制器研制

研制了应用于火电厂给煤系统的火电厂实时数据集成及网络通信控制器.使用Modbus/TCP统一标准协议实现数据的实时通信,将异构设备接口通信归一化.控制器采用双处理器并行通信模式及分任务处理结构,双处理器在握手信号控制下,通过8位并行数据线传输数据,随之进行Modbus/TCP数据封装,将TTL电平信号转换成网络信号,通过工业以太网送至上位监控中心.所研制的控制器解决了现场异构设备不同通信规约带来的数据集成和传输问题.该控制器现场运行稳定可靠,满足研制要求.     

A feasibility study of a direct-mapping parallel processing method to solve linear equations in load-flow calculation

With the growing size and complexity of power systems, system analysis—such as transients calculation—takes much time. Hence, fast calculation methods are required. Although parallel processing is a hopeful method, there have been difficulties in the parallel solution of linear equations which appear in power-flow calculations by the Newton-Raphson method. This paper aims at the fast calculation of the power-flow problem by means of parallel processing. In order to improve the suitability to the parallel solution of the differential equation in transients calculation, we assume the use of a direct-mapping parallel processing machine to map directly the network of a power system onto a network of processors. Under this assumption, we propose a new parallel-processing-oriented method in which the linear equation is solved by linear iterations between nodes with Aitken acceleration. We simulate the method on three model power systems and compare this Parallel Iterative Method (PIN) with a Parallel Direct Method (PDM) which uses the banded matrix according to the number of operations required. As a result, we can expect that PIM may solve linear equations faster than PDM with m processors, although the PIM might be inferior to the PDM with m × m processors, where m denotes the half-band width of the banded matrix.     

A transmission line fault‐location system using the wavelet transform

This paper describes the locating system of line‐to‐ground faults on a transmission line by wavelet transform. The possibility of the location with the surge generated by a fault has been theoretically proposed. In order to make the method practicable, we realize very fast processors. We design wavelet transform and location chips, and construct a very fast fault location system by processing the measured data in parallel. This system is realized by a computer with three FPGA processor boards on a PCI bus. The processors are controlled by UNIX and the system has the graphic user interface with X window system. © 2002 Wiley Periodicals, Inc. Electr Eng Jpn, 140(4): 27–37, 2002; Published online in Wiley InterScience ( www.interscience.wiley.com ). DOI 10.1002/eej.10040     

一种暂态稳定并行仿真的改进算法及其加速比分析

提出了一种基于因子路径树网络划分的暂态稳定空间并行仿真的改进算法。采用新的网络划分性能评价指标、任务划分后续调整策略进行任务划分,使各处理器间计算负荷的分配更为合理;同时,还提出了该算法在理想状况下加速比的计算公式,分析了影响并行仿真计算效率的各个因素,并以某3872节点系统为例介绍了一种估计实现最大加速比所需处理器数目的方法。在Cluster1350集群系统上的算例表明,所提改进算法提高了仿真效率。