Simulation for Design of Smart Grid Controllers
The monitoring and control of the Smart Grid are expected to be characterized by the distribution of functions and by the dependence on communications. To support the design of such systems with simulation, new methods and tools have to be developed to take into account the interaction between power system and communication network since the earliest stages of the design process.
At ACS we have implemented various solutions to evaluate the interaction between communication network and power system. For what concerns the off line simulation, we developed a co-simulation tool, VPNET, that interfaces the Virtual Test Bed, a multiphysic simulation software developed by the University of South Carolina, with OPNET, a communication network simulation software.
For what concerns real time and Hardware in the Loop (HIL) testing, a custom solution, based on a general purpose PC
architecture and open-source software (WANem), was developed. WANem is a Wide Area Network Emulator, wich allows for the setup of a transparent application gateway. Considering the complexity of the scenario of interest, the ability to support the design of future smart grids controllers with simulation requires a formal approach, a framework that integrates analytical and numerical steps, and specifies which simulation tool should be used at which stage of the design process, and how. In this context, we propose a design method based on two loops, representing the iterative steps of the design process, each supported by suitable simulation platforms.
In Loop 1 the analytical method integrates the communication effects in the control design, with particular reference to stochastic phenomena such as delays. The main design objective of the analytical method stage is the definition of the overall control strategy, of the communication infrastructure topology and the definition of the information to exchange between the different local controllers. Applying Co-Simulation to the outcome of the analytical design yields validation, or triggers the looping back to analytical design for reconsideration.
Testing in a HiL environment the controlstrategy that the first loop has produced, allows the verification of the requirements on execution time, as well as mechanical and electrical compatibility of the different parts of the system. Also in this case design Loop 2 may produce final validation of the algorithm, and its implementation, or may trigger the need for another co-simulation step. We are using the proposed approached to design the distributed voltage controller of the micro-grid in figure.