Projects per year
Abstract
As a promising solution to reduce greenhouse gas emission and alleviate energy shortage concerns, it is expected that the distributed energy resources (DERs) will be largely deployed in modern power systems. The deployment of DERs accelerates the transformation of the passive distribution network into the active distribution network (ADN) and could offer various benefits to the operation of the distribution network, such as the infrastructure upgrade deferral, and power quality and operational reliability improvement. However, the uncoordinated DER operation has adverse impacts and poses great challenges to the operation of the ADN. In order to accommodate the large-scale DERs while ensuring secure and efficient management of ADNs, it is important to leverage the benefits of DERs while mitigating the adverse impacts. Congestion could occur in the ADN due to the uncoordinated DER operation. Electrical vehicles (EVs) and heat pumps (HPs) as two popular DERs at the demand side can shift power consumption in response to electricity prices, which may cause congestion at those hours with low electricity prices. One focus of this thesis is to
study market-based congestion management methods for ADNs. Integrating different types of congestion management methods is believed to be a promising way to achieve a more efficient congestion management solution. In this context, this thesis proposes a comprehensive congestion management scheme to integrate the dynamic tariff (DT) (price-based method), network reconfiguration, and re-profiling product (incentive-based method) to alleviate day-ahead congestion. Moreover, a relaxed reconfiguration-based DT optimization model is formulated, which can ensure a feasible DT solution and set a threshold to DTs. Despite the proposed integration of the DT and incentive-based method improves the congestion management effectiveness, the integration is implemented in a simple way through independent optimizations of the DT and re-profiling product. In addition,
the integration may be limited due to potential conflicts with the current distribution system operator (DSO) regulations. Therefore, this thesis further proposes a coordination scheme of the DT method and incentive-based method, in which the DSO is in a profit-neutral position through the integrated optimization of the two methods. Additionally, the DT method may be ineffective because the energy plans made at the aggregator side may not respect network constraints due to forecast errors. Therefore, this thesis proposes a robust DT method to deal with the uncertainty, which ensures that the network constraints are satisfied even under the worst uncertainty realization scenario. As important as day-ahead congestion management, real-time congestion should be addressed as well. This thesis proposes a two-tier real-time congestion management method based on the flexible demand swap and transactive control. Real-time congestion can be alleviated efficiently by the flexible demand swap while considering the end-user’s rebound conditions and its willingness to change day-ahead schedules through a transactive control-based interaction mechanism. In the case of contingency, after fault location and isolation, service restoration is carried out to rapidly restore as many outage areas as possible through network reconfiguration and microgrid (MG) formation and through dispatching DERs. In
the existing literature, most of the service restoration models are solved by centralized algorithms based on centralized infrastructure, which is faced with a few drawbacks. Firstly, it may suffer from a heavy computation burden with the increasing
size and complexity of the ADN. Secondly, the centralized infrastructure is costly and has a single-point failure risk. Thirdly, it is challenging for the centralized infrastructure to adapt to system changes, e.g., the integration of DERs. Finally, it compromises privacy information protection. Therefore, another focus of the thesis is to design service restoration methods that can overcome the above-mentioned drawbacks. A hierarchical service restoration method is proposed based on the alternating direction method of multipliers (ADMM), in which the service restoration problem is solved in a hierarchical manner to reduce the computation burden for large-scale ADNs. Nevertheless, it still needs a central controller to send out control commands in a centralized manner. Therefore, this thesis further proposes a distributed service restoration method based on the ADMM, in which the service restoration problem is decomposed and solved in a distributed manner. The distributed method is more resilient to controller failures and can provide privacy information protection. However, these proposed methods are mainly based on the power fed by the upstream transmission network and post fault reconfiguration without considering the uncertainties of DERs and load consumption. Therefore, the thesis proposes a distributed risk-limiting service restoration method for the ADN with networked MGs, in which each MG individually uses its local DERs to restore loads without violating risk-limiting constraints of the whole network. In summary, several market-based congestion management methods and service restoration methods are proposed for the secure and efficient operation and management of ADNs.
study market-based congestion management methods for ADNs. Integrating different types of congestion management methods is believed to be a promising way to achieve a more efficient congestion management solution. In this context, this thesis proposes a comprehensive congestion management scheme to integrate the dynamic tariff (DT) (price-based method), network reconfiguration, and re-profiling product (incentive-based method) to alleviate day-ahead congestion. Moreover, a relaxed reconfiguration-based DT optimization model is formulated, which can ensure a feasible DT solution and set a threshold to DTs. Despite the proposed integration of the DT and incentive-based method improves the congestion management effectiveness, the integration is implemented in a simple way through independent optimizations of the DT and re-profiling product. In addition,
the integration may be limited due to potential conflicts with the current distribution system operator (DSO) regulations. Therefore, this thesis further proposes a coordination scheme of the DT method and incentive-based method, in which the DSO is in a profit-neutral position through the integrated optimization of the two methods. Additionally, the DT method may be ineffective because the energy plans made at the aggregator side may not respect network constraints due to forecast errors. Therefore, this thesis proposes a robust DT method to deal with the uncertainty, which ensures that the network constraints are satisfied even under the worst uncertainty realization scenario. As important as day-ahead congestion management, real-time congestion should be addressed as well. This thesis proposes a two-tier real-time congestion management method based on the flexible demand swap and transactive control. Real-time congestion can be alleviated efficiently by the flexible demand swap while considering the end-user’s rebound conditions and its willingness to change day-ahead schedules through a transactive control-based interaction mechanism. In the case of contingency, after fault location and isolation, service restoration is carried out to rapidly restore as many outage areas as possible through network reconfiguration and microgrid (MG) formation and through dispatching DERs. In
the existing literature, most of the service restoration models are solved by centralized algorithms based on centralized infrastructure, which is faced with a few drawbacks. Firstly, it may suffer from a heavy computation burden with the increasing
size and complexity of the ADN. Secondly, the centralized infrastructure is costly and has a single-point failure risk. Thirdly, it is challenging for the centralized infrastructure to adapt to system changes, e.g., the integration of DERs. Finally, it compromises privacy information protection. Therefore, another focus of the thesis is to design service restoration methods that can overcome the above-mentioned drawbacks. A hierarchical service restoration method is proposed based on the alternating direction method of multipliers (ADMM), in which the service restoration problem is solved in a hierarchical manner to reduce the computation burden for large-scale ADNs. Nevertheless, it still needs a central controller to send out control commands in a centralized manner. Therefore, this thesis further proposes a distributed service restoration method based on the ADMM, in which the service restoration problem is decomposed and solved in a distributed manner. The distributed method is more resilient to controller failures and can provide privacy information protection. However, these proposed methods are mainly based on the power fed by the upstream transmission network and post fault reconfiguration without considering the uncertainties of DERs and load consumption. Therefore, the thesis proposes a distributed risk-limiting service restoration method for the ADN with networked MGs, in which each MG individually uses its local DERs to restore loads without violating risk-limiting constraints of the whole network. In summary, several market-based congestion management methods and service restoration methods are proposed for the secure and efficient operation and management of ADNs.
Original language | English |
---|
Publisher | Technical University of Denmark |
---|---|
Number of pages | 246 |
Publication status | Published - 2020 |
Fingerprint
Dive into the research topics of 'Robust Congestion Management and Service Restoration for Active Distribution Networks'. Together they form a unique fingerprint.Projects
- 1 Finished
-
Robust Congestion Management and Service Restoration for Active Distribution Networks
Shen, F. (PhD Student), Liu, Y. (Examiner), Zhang, X.-P. (Examiner), Bindner, H. W. (Examiner), Wu, Q. (Main Supervisor), Nielsen, A. H. (Supervisor) & Xu, Y. (Supervisor)
Technical University of Denmark
01/09/2017 → 13/01/2021
Project: PhD