ReMIROR: Reference Modulation for Improved Response of Microgrid Resources

美国面临的重大能源挑战之一是到2050年以具有竞争力的成本将至少80%的可再生能源整合到电网中。虽然目前在技术上可行的可再生能源技术运行美国经济，但缺少的是一个灵活的电力系统，可以适应可再生能源的独特特性，比如电力电子接口对过流和过电压的敏感性。该项目将解决这一挑战，并通过快速准确的控制使可再生能源能够提供所需的系统性能，从而加速可再生能源的采用。这项工作的意义在于通过减少瞬变和随后提高资产利用率来消除过度设计的需要。此外，这项工作对工业产生了积极的影响，因为通过降低对控制器个人设计的敏感性，可以使用来自不同供应商的控制器。它还应有助于提高可再生能源的整合率，并有望产生积极的社会和环境影响。开发的控制器可以适应不同的电力系统配置，从小型孤立系统，如更电动的飞机，海军舰艇，公用事业微电网和军营，到直流分段系统。该项目的教育目标是通过(i)在高中为未被充分代表的少数群体举办讲习班，（ii）通过在本科和研究生课程中引入创新内容来加强课程，以及（iii）开发新的本科和研究生课程，建立实施/运营智能电网的人力能力。这一目标解决了国家对培养熟练的STEM专业人员的关注。项目组将制定方案，自主提高时变有限储备系统在各种运行条件下的资源设定点跟踪能力。这种系统的一个突出例子是微电网。微电网已经成为新兴智能电力系统的一个支持概念。微电网作为动态系统，具有快速、准确的控制器是必不可少的。当主机系统的工作点与原来的设计设想有较大的差异时，控制器的性能会下降。重新设计控制器需要计算资源和一个完整的、最新的系统模型，这两者在微电网中都不一定是现成的。此外，操作人员通常无法访问控制器的内部参数。该项目旨在通过监测响应和调节参考设定点以获得所需响应的策略来增强已经实施的控制器。具体而言，该项目将(i)建立基于离散事件系统理论的数学基础和稳定性研究，（ii）设计用于广义操作的算法，例如，在系统不平衡和噪声测量下，并提供高级功能，例如，由于故障和拓扑变化而产生的干扰的拒绝，以及（iii）进行实时模拟和实验验证。所提出的策略的显著特征将是(i)对系统变化的鲁棒性，（ii）不需要系统模型的知识，以及（iii）可扩展性和仅依赖于本地信号。这个项目将具有变革性，因为它有可能容纳来自不同供应商实现的不同控制设计。

One of the U.S. grand energy challenges is to enable integration of at least 80% renewable energy resources in the power grid at a competitive cost by 2050. While it is technically feasible to run the U.S. economy on renewable technologies available today, what is missing is a flexible power system that can accommodate the unique characteristics of renewable resources, such as their susceptibility to overcurrents and overvoltages due to power electronic interfaces. This project will address this challenge and accelerate the adoption of renewables by enabling them to deliver the required system performance via fast and accurate controls. The significance of this work is to eliminate the need for over-design by reducing transients and subsequently increasing asset utilization. Moreover, this work positively impacts industry because by reducing sensitivity to the individual design of controllers, controllers from different vendors can be employed. It should also contribute to enabling higher rates of integration of renewables, and is expected to have positive societal and environmental impact. The developed controllers can be adapted for different power system configurations, ranging from small isolated systems such as more-electric aircraft, naval ships, utility microgrids, and military camps, to DC-segmented systems. The educational goal of this project is to build human capacity for implementation/operation of the smart electricity grid through (i) workshops for underrepresented minority groups in high schools, (ii) curriculum enhancement through introducing innovative components in undergraduate and graduate courses, and (iii) developing new undergraduate and graduate courses. This goal addresses national concerns for producing skilled STEM professionals.The project team will develop schemes to autonomously improve set-point tracking capability of resources in a time-varying, limited-reserve system under various operating conditions. A prominent example of such systems is a microgrid. Microgrids have emerged as an enabling concept for the emerging smart power system. As dynamical systems, it is imperative for microgrids to have fast and accurate controllers. The performance of a controller deteriorates when the operating point of the host system varies significantly from that assumed in the original design. Redesigning controllers requires computational resources and a complete and up-to-date model of the system, neither of which are necessarily readily available in a microgrid. Moreover, the operator usually does not have access to the internal parameters of the controller. This project aims to augment the controllers that are already implemented with a strategy that monitors the response and modulates the reference set-point to attain the desired response. Specifically, this project will (i) build the mathematical foundation and stability studies based on discrete-event systems theory, (ii) design algorithms for generalized operation, e.g., under system imbalance and noisy measurements, and for providing advanced functions, e.g., rejection of disturbances due to faults and topological changes, and (iii) perform real-time simulation and experimental validation. The salient features of the proposed strategy will be (i) robustness to changes in the system, (ii) not requiring knowledge of the system model, and (iii) scalability and reliance only on local signals. This project will be transformative because it has the potential to accommodate different control designs from implemented by various vendors.