A Study of Sparse Matrix Methods on New Hardware

2015 ◽  
Vol 3 (3) ◽  
pp. 1-19
Author(s):  
Athanasios Fevgas ◽  
Konstantis Daloukas ◽  
Panagiota Tsompanopoulou ◽  
Panayiotis Bozanis
Keyword(s):  
Power Systems ◽  
Linear Systems ◽  
Large Scale ◽  
Comprehensive Evaluation ◽  
Sparse Matrix ◽  
Electrical Power ◽  
Iterative Algorithms ◽  
Direct Methods ◽  
Computer Hardware ◽  
Computing Platforms

Modeling of numerous scientific and engineering problems, such as multi-physic problems and analysis of electrical power systems, amounts to the solution of large scale linear systems. The main characteristics of such systems are the large sparsity ratio and the large number of unknowns that can reach thousands or even millions of equations. As a result, efficient solution of sparse large-scale linear systems is of great importance in order to enable analysis of such problems. Direct and iterative algorithms are the prevalent methods for solution of linear systems. Advances in computer hardware provide new challenges and capabilities for sparse solvers. The authors present a comprehensive evaluation of some, state of the art, sparse methods (direct and iterative) using modern computing platforms, aiming to determine the performance boundaries of each solver on different hardware infrastructures. By identifying the potential performance bottlenecks of out-of-core direct methods, the authors present a series of optimizations that increase their efficiency on flash-based systems.

scholarly journals Predicting AC Optimal Power Flows: Combining Deep Learning and Lagrangian Dual Methods

2020 ◽  
Vol 34 (01) ◽  
pp. 630-637 ◽  
Author(s):  
Ferdinando Fioretto ◽  
Terrence W.K. Mak ◽  
Pascal Van Hentenryck
Keyword(s):  
Deep Learning ◽  
Power Systems ◽  
Power Flow ◽  
Optimal Power Flow ◽  
Large Scale ◽  
Renewable Energy Sources ◽  
Electrical Power ◽  
Practical Applications ◽  
Fundamental Building Block ◽  
Optimal Power

The Optimal Power Flow (OPF) problem is a fundamental building block for the optimization of electrical power systems. It is nonlinear and nonconvex and computes the generator setpoints for power and voltage, given a set of load demands. It is often solved repeatedly under various conditions, either in real-time or in large-scale studies. This need is further exacerbated by the increasing stochasticity of power systems due to renewable energy sources in front and behind the meter. To address these challenges, this paper presents a deep learning approach to the OPF. The learning model exploits the information available in the similar states of the system (which is commonly available in practical applications), as well as a dual Lagrangian method to satisfy the physical and engineering constraints present in the OPF. The proposed model is evaluated on a large collection of realistic medium-sized power systems. The experimental results show that its predictions are highly accurate with average errors as low as 0.2%. Additionally, the proposed approach is shown to improve the accuracy of the widely adopted linear DC approximation by at least two orders of magnitude.

scholarly journals Amesos2 and Belos: Direct and Iterative Solvers for Large Sparse Linear Systems

2012 ◽  
Vol 20 (3) ◽  
pp. 241-255 ◽  
Author(s):  
Eric Bavier ◽  
Mark Hoemmen ◽  
Sivasankaran Rajamanickam ◽  
Heidi Thornquist
Keyword(s):  
Linear Systems ◽  
Iterative Methods ◽  
Sparse Matrix ◽  
Sparse Matrices ◽  
Direct Methods ◽  
Iterative Solvers ◽  
Software Project ◽  
Sparse Linear Systems ◽  
Scalar Type ◽  
Mixed Precision

Solvers for large sparse linear systems come in two categories: direct and iterative. Amesos2, a package in the Trilinos software project, provides direct methods, and Belos, another Trilinos package, provides iterative methods. Amesos2 offers a common interface to many different sparse matrix factorization codes, and can handle any implementation of sparse matrices and vectors, via an easy-to-extend C++ traits interface. It can also factor matrices whose entries have arbitrary “Scalar” type, enabling extended-precision and mixed-precision algorithms. Belos includes many different iterative methods for solving large sparse linear systems and least-squares problems. Unlike competing iterative solver libraries, Belos completely decouples the algorithms from the implementations of the underlying linear algebra objects. This lets Belos exploit the latest hardware without changes to the code. Belos favors algorithms that solve higher-level problems, such as multiple simultaneous linear systems and sequences of related linear systems, faster than standard algorithms. The package also supports extended-precision and mixed-precision algorithms. Together, Amesos2 and Belos form a complete suite of sparse linear solvers.

Distance Relay Impedance Measuring Problems in Presence of Wind Farms

2016 ◽  
Vol 1 (3) ◽  
pp. 464
Author(s):  
Farhad Namdari ◽  
Fatemeh Soleimani ◽  
Esmaeel Rokrok
Keyword(s):  
Power Systems ◽  
Large Scale ◽  
Electrical Power ◽  
Wind Farms ◽  
Renewable Sources ◽  
Fundamental Frequencies ◽  
Fault Resistance ◽  
Distance Relays ◽  
Increasing Demand ◽  
The Impact

<p><em>Environmental concerns along with the increasing demand on electrical power, have led to power generation of renewable sources like wind. Connecting wind turbines in large scale powers with transmission network makes new challenges like the impact of these renewable sources on power system protection. This paper studies the impact of fault resistance and its location on voltage and current fundamental frequencies of faulted lines connected to DFIG based wind farms and it will be demonstrated that because of the large differences between these frequencies, impedance measuring of distance relays is inefficient. Hence in these power systems using conventional impedance measurements is not suitable anymore and new impedance measuring approaches are required in distance relays.</em></p>

NUMERICAL INVESTIGATION ON AC PROPERTIES IN HIGH TC SUPERCONDUCTING TAPES

2003 ◽  
Vol 17 (04n06) ◽  
pp. 528-533 ◽  
Author(s):  
RICCARDO TEBANO ◽  
RENATA MELE ◽  
VINCENZO BOFFA ◽  
FEDOR GÖMÖRY ◽  
FRANTISEK STRYCEK ◽  
...  
Keyword(s):  
Power Systems ◽  
Large Scale ◽  
Numerical Models ◽  
Electrical Power ◽  
Single Layer ◽  
Critical Issue ◽  
Ac Loss ◽  
Ac Losses ◽  
Superconducting Tapes ◽  
Operational Conditions

Reduction of AC losses for large-scale applications of superconductors is a critical issue. Therefore, the quantitative evaluation of AC losses is important for the development of superconductors and their applications to electrical power systems. The development of numerical models that simulate the electromagnetic phenomena inside superconductors allows to understand the electromagnetic behavior of superconductors and to evaluate the AC loss properties. Following an approach proposed by Brandt in several papers, a numerical model was developed in order to study the AC properties of superconducting tapes in different geometrical arrangements and with time dependent current and magnetic field. Here we show an example for simple single-layer model cables to show how this rather simple and versatile numerical approach allows optimizing configurations for actual operational conditions.

scholarly journals Analysis of cyber vulnerabilities of the emergency control and relay protection to assess the reliability and survivability of electrical power systems in the era of total digitalization

2020 ◽  
Vol 216 ◽  
pp. 01040
Author(s):  
Alexey Osak ◽  
Daniil Panasetsky ◽  
Elena Buzina
Keyword(s):  
Power Systems ◽  
Control Systems ◽  
Large Scale ◽  
Electrical Power ◽  
Relay Protection ◽  
Distribution Networks ◽  
Cyber Attacks ◽  
Small Scale ◽  
Electrical Power Systems ◽  
Emergency Control

Cyber threats pose an increasing threat to energy objects. It is essential to ensure the cybersecurity of automatic control systems, such as relay protection devices (RP), devices of regime control (RC) and emergency control (EC), automated control systems. At the same time, the issues of cybersecurity include not only the problem of hacker attacks, but also the whole complex of problems relating to adequate functioning of cybernetic systems in the power industry. The authors consider two of the most acute aspects of cybersecurity in the energy systems of the future in the era of total digitalization: large-scale prepared cyber attacks on the electrical power systems (EPS) as a whole and large-scale cyber attacks on distribution networks with small-scale generation facilities and active consumers.

PARALLEL DISTRIBUTED SOLVERS FOR LARGE STABLE GENERALIZED LYAPUNOV EQUATIONS

1999 ◽  
Vol 09 (01) ◽  
pp. 147-158 ◽  
Author(s):  
PETER BENNER ◽  
JOSÉ M. CLAVER ◽  
ENRIQUE S. QUINTANA-ORTI
Keyword(s):  
Large Scale ◽  
Iterative Algorithms ◽  
Direct Methods ◽  
Matrix Equations ◽  
Distributed Architectures ◽  
Matrix Sign Function ◽  
Sign Function ◽  
The Matrix ◽  
Qz Algorithm ◽  
Lyapunov Matrix

In this paper we study the solution of stable generalized Lyapunov matrix equations with large-scale, dense coefficient matrices. Our iterative algorithms, based on the matrix sign function, only require scalable matrix algebra kernels which are highly efficient on parallel distributed architectures. This approach avoids therefore the difficult parallelization of direct methods based on the QZ algorithm. The experimental analtsis reports a remarkable performance of our solvers on an IBM SP2 platform.

Sign in / Sign up

Export Citation Format

Share Document