Font Size: a A A

A Parallel Algorithm For Solving The Probabilistic Security Constrained Optimal Power Flow

Posted on:2019-12-22Degree:MasterType:Thesis
Country:ChinaCandidate:Q L ZhangFull Text:PDF
GTID:2370330572966298Subject:Computer technology
Abstract/Summary:PDF Full Text Request
Modern power network is becoming more and more complex,and the probability of failure is increasing.Considering many Probabilistic Security Constrained Optimal Power Flow(P-SCOPF)problems with anticipated failure states,the solution scale is huge,and the corresponding computing memory requirement is large and the computing process is very time-consuming.For large-scale P-SCOPF problems,traditional serial computing seems to have little effort.The effective way to solve this problem is to use parallel optimization in parallel environment.Firstly,this paper conforms to the research upsurge of parallel optimization calculation,uncertainty dissipation of power system and static safe operation of power grid.In order to describe the influence of uncertainties of operation parameters of power system such as new green energy generation on power system,the uncertainties are considered by probability method and cited.A new P-SCOPF model is established by putting it into SCOPF model.After the line fault probability is obtained by using the predictive model of transmission line accident probability in this paper,the model constraints are linearized to pretreat and deform.Aiming at the determined model,a parallel optimization algorithm framework based on Alternating Direction Method of Multipliers(ADMM)is used to solve the model.The P-SCOPF problem is decoupled into two independent sub-problems after introducing relaxation variables(pre-accident and post-accident).The scale of each sub-problem is similar.Each computing node is uniformly adjusted by dual variables,and then each computing node is computed in parallel to realize the P-SCOPF problem.Fast parallel solution.Finally,the algorithm is simulated by using the IEEE 14,30,57 and 118-bus systems,and the experimental results are analyzed in terms of convergence,computing speed and parallel effect.Secondly,in order to improve the calculation results of ADMM algorithm,three improved algorithms,cooperative ADMM,Peaceman-Rachford splitting algorithm and regularized multiplier method,are adopted.Taking IEEE30,57 and 118-bus systems as examples,the convergence,computational speed and parallelism of the three improved algorithms are analyzed and compared.
Keywords/Search Tags:Transmission line, probabilistic fault, Probabilistic Security Constrained Optimal Power Flow, ADMM, Parallel
PDF Full Text Request
Related items