The purpose of optimal power flow (OPF) is to achieve the best outcome possible for a given power system in terms of power loss, generation, and cost while taking inequality and security constraints into consideration [1]. Economic dispatch is concerned with determining the generating levels at which the overall cost of generation is minimized for a specific level of demand. This is frequently employed in power exchanges and pools. The supply must match the demand in this situation, and the generators’ generation constraints also apply. On the other hand, the DC-OPF is an optimized power flow system that has been linearized. It takes transmission line congestion into account. It is utilized for nodal and zonal pricing in the power market. On the entire power flow equation, AC-OPF is based. This nonlinear issue considers active and reactive power flow, current voltage, and losses. AC-OPF is employed in electricity markets, to minimization of losses, optimize the voltage profile, and maintain system security.
One of the most extensively explored topics in power systems is the optimum power flow (OPF) problem, which is mostly investigated using the centralized technique in current practice [2]. Based on the current practice, the system operator as a central control entity gathers extensive information such as the generator’s physical characterization and cost structure, load forecast, and electrical network parameters, and solves the OPF problem in a centralized manner. Thereby, the current prevalent practice of solving the OPF problem is considered as a centralized optimization approach for OPF.
The decentralized OPF, on the other hand, reduces the level of computing for the optimal solution to the nodal level. The OPF problem is resolved in this case by the individual nodes, which only have local knowledge of the system, eliminating the need for a central or sub-area controller [3]. And, the nodes or we can say a cluster or an area of the network can communicate with its other adjacent nodes for reaching the condition of optimality.
The modern power system network is experiencing huge change with the recent advancement in technology and the concerns towards clean energy. Also, complexity arises due to the expansion of the power system network as time progresses. Due to this, control objects in a power system network can be diverse and their number can increase dramatically thereby making the interaction between control objects and the central controller more complex and dynamic [4]. This brings new challenges to the traditional centralized control framework. On the other hand, OPF is the heart of the power system network as it finds its application in unit commitment, economic dispatch of power, transmission line congestion management, and so on.
Computationally, the optimization has nonconvexities which makes the problems difficult to solve. With the system growing in scale and increasing complexity due to the addition of more control entities the optimization problem in terms of power flow becomes an additional challenge [1]. The centralized framework for power flow has various limitations or issues related to the communication bottleneck, confidentiality, and autonomy [3]. The decentralized framework for the OPF could address the issues regarding the centralized framework where the decentralized approach would break the global optimization problem into sub-problems thereby reducing the computational burden, communication bottleneck, increased confidentiality among the areas or nodes in the network, and finally will boost the autonomy.
References
[1] Cain, M. B., O’neill, R. P., & Castillo, A. (2012). History of optimal power flow and formulations. Federal Energy Regulatory Commission, 1, 1-36.
[2] Hug-Glanzmann, Gabriela, and Göran Andersson. “Decentralized optimal power flow control for overlapping areas in power systems.” IEEE Transactions on Power Systems 24, no. 1 (2009): 327-336
[3] Lu, Wentian, Mingbo Liu, Shunjiang Lin, and Licheng Li. “Fully decentralized optimal power flow of multi-area interconnected power systems based on distributed interior point method.” IEEE Transactions on Power Systems 33, no. 1 (2017): 901-910.
[4] Molzahn, Daniel K., Florian Dörfler, Henrik Sandberg, Steven H. Low, Sambuddha Chakrabarti, Ross Baldick, and Javad Lavaei. “A survey of distributed optimization and control algorithms for electric power systems.” IEEE Transactions on Smart Grid 8, no. 6 (2017): 2941-2962.