1949-3053 (c) 2017 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information. This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/TSG.2017.2745491, IEEE Transactions on Smart Grid 1 Abstract--This paper investigates the application of Supervisory Control of Discrete Event System (SCDES) to the management and control of a Custom Power Park (CPP). The heterogeneous nature of upcoming devices and equipment in CPP require advanced control methods to ensure the integrity and reliability under different operational states. A solution to achieve appropriate controllability, while avoiding complexity, is to sub-divide the control problem based on event-triggered dynamics where the occurrence of specific events could change the state of the system. This idea is employed to formulate the problem of coordination of the devices in a CPP and develop a systematic method to design a supervisory control based on the theory of SCDES. Three modular supervisors are synthesized using the TCT software and simulated using the Simulink. The proposed methodology could be applied to several control problems in microgrids. Index Terms—Smart Grids, Discrete Event Systems, Supervisory Control, Custom Power park I. INTRODUCTION HE upcoming challenges faced by power delivery networks have questioned the capability of the traditional network’s topologies and technologies to deliver economical, green, reliable, and high-quality power to customers. The ever-increasing load demand, environmental concerns (such as reducing carbon emission), penetration of distributed renewable sources and energy storage units with non- deterministic loads such as electric vehicles have urged the network operators to invest in new technologies to augment the power network to operate more efficiently. Advanced communication and information technologies, as well as advanced equipment with a specific level of intelligence, are utilized to form an intelligent network which could accommodate a variety of generations, enhance the quality of delivered power, interact with customers and carry self- healing capabilities [1]. The topology of the future power network aims at distributed autonomous subnetworks. For instance, a cluster of loads and distributed generation resources in a specific geographical area will form a microgrid that is controlled by the local controller and could operate in grid-connected or Manuscript received on April 27; revised June 12; accepted August 22, 2017. A. Kharrazi and V. Sreeram are with The School of Electrical, Electronics and Computer Engineering, The University of Western Australia (email: ali.kharrazi@research.uwa.edu.au ; victor.sreeram@uwa.edu.au). Y. Mishra is with School Electrical Engineering and Computer Science, Queensland University of Technology, Brisbane, Australia (e-mail: yateendra.mishra@qut.edu.au). island mode while interacting with the grid or other microgrids [2]. This distributed topology will enhance the resilience of the power grid [3], facilitate the integration of distributed energy resources (DER), especially renewable sources, accommodate energy storage units and enhance the quality of power delivered. The design of the topology and technologies used in microgrids depends on a specific objective, e.g. a custom power park (CPP) topology aims to supply highly sensitive loads with high quality uninterruptible power. Since the management of smart grids and microgrids is dealing with a heterogeneous system with a large number of devices, new mechanisms are required to handle the energy management and control problems. One solution to simplify the complexity of the problem is to split the system into smaller subsystems based on the characteristics and behavior of the devices. For example, many operations in power systems could be modelled in form of a finite state machine (FSM) where the occurrence of events will change the state of the system. This behavior can be demonstrated through the methodology of the discrete-event system (DES). The dynamics of the power system could be divided into time- triggered and event-triggered dynamics. Instead of using a central complex hybrid control method to deal with both dynamics, a DES model of the system is obtained by observing the event-triggered dynamics and states of the system. It is then possible to synthesize a hierarchical control structure where local controllers regulate the continuous dynamics of individual devices and a supervisory control produces high-level commands to transit the system into desired states based on predefined control pattern. This systematic control method is viable in managing many operations in smart grids. For example, primary control of microgrids, economic signalling, demand response management and many decision makings that exhibit discrete event nature could be managed in the context of supervisory control of discrete event system (SCDES) [4]. Although the main applications of DES are in manufacturing processes and communication networks, it can be also useful in power systems for energy management, control, modeling, monitoring and diagnosis applications. Authors in [5] proposed DES framework for a sample 14-bus 40-line transmission network, where each transmission line is modelled as a two-states DES with two events (states: line in service, line out of service; events: line tripping, line restoring). A supervisory control is then proposed to manage the restoration process of tripped lines so that the highest security level is obtained. Similarly, authors in [6] and [7] Discrete-Event Systems Supervisory Control for a Custom Power Park Ali Kharrazi, Student Member, IEEE, Yateendra Mishra, Member, IEEE, and Victor Sreeram, Sr. Member, IEEE T