Change search
Link to record
Permanent link

Direct link
BETA
Publications (3 of 3) Show all publications
Yuan, Z., Hesamzadeh, M. R., Cui, Y. & Bertling Tjernberg, L. (2016). Applying High Performance Computing to Probabilistic Convex Optimal Power Flow. In: 2016 INTERNATIONAL CONFERENCE ON PROBABILISTIC METHODS APPLIED TO POWER SYSTEMS (PMAPS): . Paper presented at International Conference on Probabilistic Methods Applied to Power Systems (PMAPS), OCT 16-20, 2016, Beijing, PEOPLES R CHINA. IEEE
Open this publication in new window or tab >>Applying High Performance Computing to Probabilistic Convex Optimal Power Flow
2016 (English)In: 2016 INTERNATIONAL CONFERENCE ON PROBABILISTIC METHODS APPLIED TO POWER SYSTEMS (PMAPS), IEEE, 2016Conference paper, Published paper (Refereed)
Abstract [en]

The issue of applying high performance computing (HPC) techniques to computation-intensive probabilistic optimal power flow has not been well discussed in literature. In this paper, the probabilistic convex AC OPF based on second order cone programming (P-SOCPF) is formulated. The application of P-SOCPF is demonstrated by accounting uncertainties of loads. To estimate the distributions of nodal prices calculated from PSOCPF, two point estimation method (2PEM) is deployed. By comparing with Monte Carlo (MC) method, the accuracy of 2PEM is proved numerically. The computation efficiency of 2PEM outperforms MC significantly. In the context of large scale estimation, we propose to apply high performance computing (HPC) to P-SOCPF. The HPC accelerated P-SOCPF is implemented in GAMS grid computing environment. A flexible parallel management algorithm is designed to assign execution threads to different CPUs and then collect completed solutions. Numerical results from IEEE 118-bus and modified 1354pegase case network demonstrate that grid computing is effective means to speed up large scale P-SOCPF computation. The speed up of P-SOCPF computation is approximately equal to the number of cores in the computation node.

Place, publisher, year, edition, pages
IEEE, 2016
Keywords
Probabilistic Convex AC OPF, Grid Computing, Two Point Estimation, Nodal Price, Uncertainty of Load
National Category
Electrical Engineering, Electronic Engineering, Information Engineering
Identifiers
urn:nbn:se:kth:diva-202493 (URN)10.1109/PMAPS.2016.7764116 (DOI)000392327900071 ()2-s2.0-85015197150 (Scopus ID)978-1-5090-1970-0 (ISBN)
Conference
International Conference on Probabilistic Methods Applied to Power Systems (PMAPS), OCT 16-20, 2016, Beijing, PEOPLES R CHINA
Note

QC 20170301

Available from: 2017-03-01 Created: 2017-03-01 Last updated: 2017-05-19Bibliographically approved
Yuan, Z. & Hesamzadeh, M. R. (2016). Implementing zonal pricing in distribution network: The concept of pricing equivalence. In: IEEE Power and Energy Society General Meeting: . Paper presented at 2016 IEEE Power and Energy Society General Meeting, PESGM 2016, 17 July 2016 through 21 July 2016. IEEE
Open this publication in new window or tab >>Implementing zonal pricing in distribution network: The concept of pricing equivalence
2016 (English)In: IEEE Power and Energy Society General Meeting, IEEE, 2016Conference paper, Published paper (Refereed)
Abstract [en]

Distribution locational marginal pricing (DLMP) is critical market mechanism to boost services from distributed energy resources (DER). This paper propose to design zonal pricing in distribution network according to the concept of pricing equivalence (PE). The rules of the zonal pricing are derived. We prove that equivalent load shift from demand response can be achieved by zonal pricing if pricing equivalence is deployed. Convex AC optimal power flow (OPF) is used to calculate zonal prices. The benefits of convex AC OPF are more accurate energy pricing and global optimization target. The responsive load with passive load controllers are modeled and solved in GAMS platform. Different zonal pricing approaches (PE, reference node and average of nodal prices) are compared. IEEE 14-bus network and two IEEE 13-node networks are connected to be an illustrative test case offering numerical results. The results show that zonal pricing designed according to PE can achieve the same load shift effects and quite close consumer payments as nodal pricing. PE outperform other zonal pricing approaches prominently in congested network situations.

Place, publisher, year, edition, pages
IEEE, 2016
Keywords
Convex AC OPF, Demand response, Distribution network, Pricing equivalence, Zonal pricing, Electric load flow, Electric power distribution, Energy resources, Global optimization, Ac optimal power flows, Congested networks, Distributed energy resource, Locational marginal pricing, Numerical results, Costs
National Category
Communication Systems
Identifiers
urn:nbn:se:kth:diva-202151 (URN)10.1109/PESGM.2016.7741478 (DOI)000399937901121 ()2-s2.0-85002339096 (Scopus ID)9781509041688 (ISBN)
Conference
2016 IEEE Power and Energy Society General Meeting, PESGM 2016, 17 July 2016 through 21 July 2016
Note

QC 20170313

Available from: 2017-03-13 Created: 2017-03-13 Last updated: 2017-06-15Bibliographically approved
Yuan, Z. & Reza Hesamzadeh, M. A Modified Benders Decomposition Algorithm to Solve Second-Order Cone AC Optimal Power Flow. IEEE Transactions on Smart Grid
Open this publication in new window or tab >>A Modified Benders Decomposition Algorithm to Solve Second-Order Cone AC Optimal Power Flow
(English)In: IEEE Transactions on Smart Grid, ISSN 1949-3053, E-ISSN 1949-3061Article in journal (Refereed) In press
Abstract [en]

This paper proposes to speed up solving large-scale second-order cone AC optimal power flow (SOC-ACOPF) problem by decomposition and parallelization. Firstly, we use spectral factorization to partition large power network to multiple subnetworks connected by tie-lines. Then a modified Benders decomposition algorithm (M-BDA) is proposed to solve the SOC-ACOPF problem iteratively. Taking the total power output of each subnetwork as the complicating variable, we formulate the SOC-ACOPF problem of tie-lines as the master problem and the SOC-ACOPF problems of the subnetworks as the subproblems in the proposed M-BDA. The feasibility and optimality (preserving the original optimal solution of the SOC-ACOPF model) of the proposed M-BDA are analytically and numerically proved. A GAMS grid computing framework is designed to compute the formulated subproblems of M-BDA in parallel. The numerical results show that the proposed M-BDA can solve large-scale SOC-ACOPF problem efficiently. Accelerated M-BDA by parallel computing converges within few iterations. The computational efficiency (reducing computation CPU time and computer RAM requirement) can be improved by increasing the number of partitioned subnetworks.

Keywords
Optimal power flow, Network partitioning, Modified Benders decomposition, Feasibility and optimality proof, Parallel computing.
National Category
Electrical Engineering, Electronic Engineering, Information Engineering
Research subject
Electrical Engineering
Identifiers
urn:nbn:se:kth:diva-227042 (URN)
Note

QC 20180607

Available from: 2018-05-02 Created: 2018-05-02 Last updated: 2018-06-07Bibliographically approved
Organisations
Identifiers
ORCID iD: ORCID iD iconorcid.org/0000-0003-0471-9066

Search in DiVA

Show all publications