The electric power system is a cyber-physical system with power flow in the physical system and information flow in the cyber. Simulation is crucial to understanding the dynamics and control of electric power systems yet the underlying communication system has historically been ignored in these studies. This paper aims at meeting the increasing needs to simulate the operations of a real power system including the physical system, the energy management system, the communication system, and the emerging wide-area measurement-based controls. This paper proposes a cyber-physical testbed design and implementation for verifying and demonstrating wide-area control methods based on streaming telemetry and phasor measurement unit data. The proposed decoupled architecture is composed of a differential algebraic equation based physical system simulator, a software-defined network, a scripting language environment for prototyping an EMS system and a control system, all of which are integrated over industry-standard communication protocols. The proposed testbed is implemented using open-source software packages managed by a Python dispatcher. Finally, demonstrations are presented to show two wide-area measurement-based controls - system separation control and hierarchical voltage control, in the implemented testbed.
Modern electric power systems rely on automated controls for secure and economic operation. The increasing level of renewable energy penetration through power electronics interfaces is bringing in faster dynamics and will require well-coordinated controls. Recently, synchrophasors have been deployed to obtain real-time data for situational awareness and fast, coordinated controls over a wide-area. Testing and demonstration of the effectiveness of wide-area measurement-based controls, therefore, become fundamental before actual implementations in real systems. Wide-area measurement-based control decisions are made based on global information which relies on the communication infrastructure. The conventional approach of modelling controllers using differential algebraic equations (DAE) has been dominant in power system simulation. However, this has many limitations in representing the measurement device characteristics and communication network behaviours. A more sophisticated simulation platform, namely, a testbed, needs to be designed and implemented to enable more testing capabilities.
A modern electric power system is intrinsically a cyber-physical system (CPS), which consists of the power loop as the physical part, and the communication loop as cyber. The physical part of a transmission system is an electric network of AC and DC transmission lines connected to generators, transformers, load, and power electronic devices. Electric power flows in the physical system following Kirchhoff’s Law. The cyber part consists of a communication system in which information is exchanged, an energy management system (EMS, also known as supervisory control and data acquisition system, SCADA) in which generation scheduling is performed, and a control system in which fast wide-area control signals are generated. Information flows in the cyber system following defined information exchange routes.
The physical system and the cyber system are bridged by measurement devices and actuators. Measurement devices, such as synchrophasors, take time-stamped measurements from the physical network and send them to EMS/control systems for decision-making. Actuators, such as flexible AC transmission system (FACTS), receive the control signals and adjust the physical power flow accordingly. Therefore, closed-loop monitoring and control of a cyber-physical electric power system can be realized in an integrated fashion.
The major challenges in building a cyber-physical testbed for electric power systems are two-fold. On the one hand, in the physical system simulator/emulator, the mathematical equations to describe the power system dynamics need to be inclusive and have sufficient detail for control studies. On the other hand, in the cyber system simulator/emulator, the architecture, and topology of the communication network needs to be modelled to simulate delay, package loss, and cyber security events. Extensive efforts are required to design and implement a cyber-physical testbed from scratch.
Open-source software for the research community has captured growing attentions in recent years. Open-source software is distributed in the form of source code and can be reused or modified under license terms. For example, the most widely used operating system, Linux, and the most widely used web server, Apache, are both open-source software. In the power community, the number of opensource packages has also been increasing. Packages such as MATPOWER, PSAT, OpenDSS, GridLab-D, GridDyn and Andes are widely used for research activities. In the communication community, packages such as NS-3, OpenFlow, and Mininet are also open-source for communication network simulation. These packages are well maintained and documented.
The main contribution of this paper is the detailed design and implementation of a cyber-physical testbed, known as the Large-Scale Testbed (LTB), for wide-area measurement-based control verification and demonstration using open-source software. This paper is organized as follows. Section 2 provides a review on related work and available open-source software. Section 3 discusses the design choices and design architecture. Section 4 elaborates on the implementation of modules and the synchronization. Section 5 shows two sample case studies and demonstrations and section 6 provides conclusions.
The interest in developing a cyber-physical simulation platform/testbed has been growing since the rapid development of the smart grid. Simulation has been a powerful method to study large engineered systems where it is difficult to perform actual tests; however, the traditional simulation method of the electric power system using standalone DAE are not sufficient for communication scenarios and cyber security assessments. The concept for designing the next-generation real-time control, communication, control and computation for large power systems has been proposed a decade ago [1]. Related works are categorized into the development of co
This content is AI-processed based on open access ArXiv data.
