 |
Center for Nonlinear Studies |
It is well recognized that distributed energy sources such as solar and wind are intermittent, that their presence will shift the operation of power system from the current mode of regulated utilities to a competitive generation provision, and that a high penetration level of these sources demands new regimes of measurement and estimation, communication, control, protection, security, and operation. On the other hand, contemporary sensing and communication networks enable real-time collection and subscription of geographically-distributed information and such information can be used to significantly enhance the performance of electric power systems at the levels of generation, transmission and distribution. Through a shared sensing/communication network, distributed generation can now be controlled to operate autonomously and robustly as micro grids, and operation of these micro grids can exhibit cooperative behaviors to enhance voltage stability of distribution networks. By incorporating dynamic game and cooperative control algorithms, pooled generation/consumption of micro-grids can be automatically optimized to enhance both energy dispatch and transient stability of the overall power system. The talk illustrates analysis and design tools for the so-called cooperative networked systems among which information exchanges are local and intermittent, their changing patterns are not known a priori, and may have significant latencies. Canonical forms and designs of distributed cooperative control are introduced to ensure cooperative stability and design cooperative control for networked linear and nonlinear systems. Formulations and distributed algorithms of estimation, optimization and dynamic games are also illustrated. Based on these methodologies and tools, innovative algorithms of distributed control and optimization can be implemented to enable robust, intelligent and efficient operations for power systems with distributed and intermittent power generation sources. As an illustration, the following three-layered control-optimization-control structure is discussed: (i) A cooperative control algorithm that enables micro-grids to form autonomously and to evolve as distributed generation changes over time; (ii) Each of the self-evolving micro-grids negotiates with the main grid to determine the best operating conditions by following the incentive (and limit) specified by the main grid and by maximizing the group energy output; and (iii) In the event of a major disturbance or fault, distributed generations and their inverter-based controls provide transient controls that maintain voltage stability of distribution networks; (iv) Dynamic game algorithms enable multi-level optimization that improves both energy dispatch and transient stability of the overall power system. Sample results from the on-going DoE SEGIS and NSF projects will be presented to illustrate their effectiveness.
Host: Misha Chertkov, chertkov@lanl.gov, 665-8119