CURENT Large-scale Testbed (LBT)
The CURENT Large-scale Testbed (LTB), is the first of its kind to provide a virtual electric power grid for researchers to experiment with closed-loop controls and algorithms. Research and application ideas can be quickly and seamlessly integrated for verification in this virtual power system. More information is available on the CURENT LTB website.
Products
LTB includes four core packages, ANDES, AMS, AGVis, and DiME. These packages provide dynamic simulation, market simulation, geo-visualization, and data messaging environment, respectively. These packages can be used independently for specific research purposes or in combination for closed-loop studies. To streamline the process for federal use, we also provide a repository LTB that includes all four packages as submodules.
ANDES
This software could be of interest to you if you are working on DAE modeling, simulation, and control for power systems. It has features that may be useful if you are applying deep (reinforcement) learning to such systems.
ANDES is by far easier to use for developing differential-algebraic equation (DAE) based models for power system dynamic simulation than other tools such as PSAT, Dome and PST, while maintaining high numerical efficiency.
ANDES comes with a rich set of commercial-grade dynamic models with all details implemented, including limiters, saturation, and zeroing out time constants.
ANDES provides a descriptive modeling framework in a scripting environment. Modeling DAE-based devices is as simple as describing the mathematical equations. Numerical code will be automatically generated for fast simulation.
Features:
- a rich library of transfer functions and discontinuous components (including limiters, deadbands, and saturation functions) available for model prototyping and system analysis.
- industry-grade second-generation renewable models (solar PV, type 3 and type 4 wind), distributed PV and energy storage model.
- routines including Newton method for power flow calculation, implicit trapezoidal method for time-domain simulation, and full eigenvalue analysis.
- developed with performance in mind. While written in Python, ANDES can finish a 20-second transient simulation of a 2000-bus system in a few seconds on a typical desktop computer.
- out-of-the-box PSS/E raw and dyr data support for available models. Once a model is developed, inputs from a dyr file can be immediately supported.
AMS
With the built-in interface with dynamic simulation engine, ANDES, AMS enables Dynamics Interfaced Stability Constrained Production Cost and Market Operation Modeling.
AMS produces credible scheduling results and competitive performance. The following results show the comparison of DCOPF between AMS and other tools.
Cost [$] | AMS | MATPOWER | pandapower |
---|---|---|---|
PEGASE 1354-Bus | 1,173,590.63 | 1,173,590.63 | 1,173,590.63 |
PEGASE 2869-Bus | 2,338,915.61 | 2,338,915.61 | 2,338,915.61 |
GOC 4020-Bus | 793,634.11 | 793,634.11 | 793,634.11 |
EPIGRIDS 5658-Bus | 1,195,466.12 | 1,195,466.12 | 1,195,466.12 |
EPIGRIDS 7336-Bus | 1,855,870.94 | 1,855,870.94 | 1,855,870.94 |
AGVis
AGVis is a geo-visualization tool that facilitates the visualization of large-scale real-time power system simulation.
AGVis can also visualize the dynamic results of the Western Electric Coordinating Council (WECC) power system:
DiME
DiME enables high-concurrency, high-volume real-time data exchange in large-scale power system simulations using a shared workspace.
Furthermore, DiME is compatible with multiple programming languages including Python, Matlab, and JavaScript.
Interactions between server and clients in DiME: