Projects per year
Rapidly increasing share of intermittent renewable energy production poses a great challenge of the management and operation of the modern power systems. Deployment of a large number of flexible demands, such as electrical vehicles (EVs) and heat pumps (HPs), is believed to be a promising solution for handling the challenge. Equipped with batteries and hot water storage systems, EVs and HPs are able to shift the consumption according to the production level of renewable energy. However, most of today’s distribution networks are not able to accommodate such large number of flexible demands if coordination is not exercised. Congestion can occur on distribution networks if the EVs and HPs consume power simultaneously. This thesis is dedicated to handle the congestion problems on distribution networks when there is high penetration of distributed energy resources (DERs), including EVs and HPs. Market-based congestion management methods are the focus of the thesis. They handle the potential congestion at the energy planning stage; therefore, the aggregators can optimally plan the energy consumption and have the least impact on the customers. After reviewing and identifying the shortcomings of the existing methods, the thesis fully studies and improves the dynamic tariff (DT) method, and proposes two new market-based congestion management methods, namely the dynamic subsidy (DS) method and the flexible demand swap method. The thesis improves the DT method from four aspects. Firstly, the formulation of the DT method has been improved. Based on the locational marginal pricing (LMP) concept, the DT method has been proposed in several previous works for congestion management in a decentralized manner. However, linear programming models are not suitable for determining DT due to the multiple-response issue (one price set can have multiple flexible demand responses from aggregators). The thesis proposes a quadratic programming model for the DT method which can avoid the multiple-response issue and make the DT method an efficient decentralized congestion management method. Secondly, the combination of the DT method and direct control methods is studied and the feeder reconfiguration based DT method is proposed for more efficient congestion management and loss reduction on distribution networks. Thirdly, the stochastic nature of flexible demands is studied and a method for uncertainty management of the DT method is proposed. The probability of congestion events is controlled to be under a certain level through the modified DT method, where the behavior and parameters of the flexible demands have a given probability distribution. At last, a convex relaxation based AC optimal power flow (OPF) model is proposed for determining DT where voltage constraints are included. Moreover, a sufficient condition for exact convex relaxation is proposed and validated. The condition is that there is no reverse power flow, or only active or reactive reverse power flow on the distribution network. After the study of the DT method, the thesis proposes the DS method for day-ahead congestion management, which is conceptually opposite to the DT method; however, it doesn’t discriminate the customers. Finally, the thesis proposes the flexible demand swap method for real-time congestion management, which handles the residual congestion after the day-ahead market and the congestion caused by forecast errors and contingent events. As such, a series of market-based methods, including DT, DS and flexible demand swap, are formed systematically in this thesis for handling congestion more comprehensively and efficiently.
|Publisher||Technical University of Denmark, Department of Electrical Engineering|
|Number of pages||188|
|Publication status||Published - 2017|
FingerprintDive into the research topics of 'Congestion management of electric distribution networks through market based methods'. Together they form a unique fingerprint.
- 1 Finished
Congestion Management of Distribution Networks with High Penetration of Distributed Energy Resources (DERs)
Huang, S., Wu, Q., Nielsen, A. H., Træholt, C., Hobbs, B. F. & Repo, S. P.
01/10/2013 → 08/02/2017