TY - JOUR
T1 - A Transformation-Based Multi-Area Dynamic Economic Dispatch Approach for Preserving Information Privacy of Individual Areas
AU - Wu, Lei
N1 - Publisher Copyright:
© 2010-2012 IEEE.
PY - 2019/1
Y1 - 2019/1
N2 - As the growing interconnection of regional electricity grids and the large-scale integration of renewable energy demand for effective coordination among multiple areas, the multi-area dynamic economic dispatch (MA-DED) problem has been studied for exploring coordinated optimal operations to achieve overall energy security and economic efficiency. However, due to political and technical challenges, a centralized solution to MA-DED is unlikely to be practically feasible. In turn, decomposition schemes have been applied for solving MA-DED in a distributed manner so as to avoid disclosing commercially sensitive information of individual areas to others. In recognizing computational challenges of distributed approaches, this paper discusses a transformation-based MA-DED approach, which masks each area's private information in the objective and constraints of MA-DED by privately generated and held random matrices. Strategies of using additional slack variables and redundant constraints are also discussed to further disguise system scale of each area from being disclosed. More importantly, optimal solutions to the original MA-DED problem, including tie-line power flows, dispatches of generators and loads, and locational marginal prices, can be retrieved from solutions of the transformed problem. Numerical studies show effectiveness of the proposed approach for guaranteeing the same optimal MA-DED solution, while preserving information privacy of individual areas. Moreover, the proposed approach presents significant computational benefits over distributed approaches, especially for multi-area systems with larger numbers of tie-lines and more complicated connection topologies of areas.
AB - As the growing interconnection of regional electricity grids and the large-scale integration of renewable energy demand for effective coordination among multiple areas, the multi-area dynamic economic dispatch (MA-DED) problem has been studied for exploring coordinated optimal operations to achieve overall energy security and economic efficiency. However, due to political and technical challenges, a centralized solution to MA-DED is unlikely to be practically feasible. In turn, decomposition schemes have been applied for solving MA-DED in a distributed manner so as to avoid disclosing commercially sensitive information of individual areas to others. In recognizing computational challenges of distributed approaches, this paper discusses a transformation-based MA-DED approach, which masks each area's private information in the objective and constraints of MA-DED by privately generated and held random matrices. Strategies of using additional slack variables and redundant constraints are also discussed to further disguise system scale of each area from being disclosed. More importantly, optimal solutions to the original MA-DED problem, including tie-line power flows, dispatches of generators and loads, and locational marginal prices, can be retrieved from solutions of the transformed problem. Numerical studies show effectiveness of the proposed approach for guaranteeing the same optimal MA-DED solution, while preserving information privacy of individual areas. Moreover, the proposed approach presents significant computational benefits over distributed approaches, especially for multi-area systems with larger numbers of tie-lines and more complicated connection topologies of areas.
KW - Dynamic economic dispatch
KW - information privacy
KW - multi-area coordination
UR - http://www.scopus.com/inward/record.url?scp=85030261615&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85030261615&partnerID=8YFLogxK
U2 - 10.1109/TSG.2017.2751479
DO - 10.1109/TSG.2017.2751479
M3 - Article
AN - SCOPUS:85030261615
SN - 1949-3053
VL - 10
SP - 722
EP - 731
JO - IEEE Transactions on Smart Grid
JF - IEEE Transactions on Smart Grid
IS - 1
M1 - 8031400
ER -