Fast Real-time Linearized State Estimation in Active Distribution Systems

电力传输系统在高压（HV）变电站处享有各种测量，这使得电网的电气状况（即电压和功率流）的估计是可行的。然而，配电系统通常以单向电力流运行，另一方面，具有可预测的行为。从中央大型发电厂到可再生能源在配电系统内的高度渗透的模式转变改变了上述特征。配电系统不再是被动的可预测的电力消费者，而是主动电网，这要归功于主要基于间歇性风能或太阳能的分布式发电（DG）。可再生能源带来的波动性使人们对及时监测配电系统的能力以及采取具体行动应对大规模干扰和保证高水平复原力的能力感到关切。在这种情况下，状态估计是指使用从现有仪表/测量接收的数据来推断系统中不同位置处的电压和电流的过程。这个过程并不像听起来那么容易，特别是因为实际因素，如计量误差和通信延迟。尽管如此，有一个共识，即配电系统运营商（DSO）将很快非常需要的状态估计，以采取适当的行动，尽快违反系统条件。这个博士项目的目的是促进配电系统的状态估计过程-不需要新的投资或必然诉诸新技术。这将通过依靠先进的求解技术和利用电力系统的基本特性来实现。该项目的四个主要目标如下：1）在没有任何近似的情况下线性建模问题，2）即使在大规模配电网的情况下也能实时执行该过程，3）在保持网格方程线性的同时避免数据同步的需要，4）消除伪测量以提高精度。该项目及其产出将有助于英国的优势，并与EPSRC交付计划保持一致，以建立一个有弹性和相互联系的国家。配电系统的实时监测将提高电网的弹性，从而有助于EPSRC定义的“能源安全和效率”优先领域，以实现其“弹性国家”的产出。该项目还与EPSRC为能源网络研究领域设定的目标有关。它通过利用英国电网中现有的测量技术和基础设施，确保“满足国家需求的负担得起的解决方案”。

Power transmission systems enjoy various measurements at high-voltage (HV) substations, which make the estimation of electrical condition of the grid, i.e. voltages and power flows, viable. Distribution systems however have been conventionally operated with unidirectional power flows and on the other hand, had predictable behaviour. The paradigm shift from central large power plants to the high penetration of renewable energy sources within the distribution system has transformed the foregoing characteristics. Distribution systems are not passive predictable consumers of electricity anymore, rather active grids thanks to the distributed generation (DG) mainly based on intermittent wind or solar power. Volatility introduced by renewables has raised concerns regarding the ability to monitor distribution systems in time, and taking concrete actions to counteract large disturbances and guarantee a high level of resilience. In this context, state estimation refers to the process of using the data received from existing meters/measurements to infer voltages and currents at different locations in the system. This process is not as easy as it may sound, particularly because of practical factors such as metering errors and communication latencies. Nonetheless, there is a consensus that distribution systems operators (DSOs) will soon be greatly in need of state estimation in order to take proper action against violating system conditions as swiftly as possible. This PhD project is aimed at facilitating the state estimation process for distribution systems- with no need for new investment or necessarily resorting to new technologies. This will be achieved by relying on advanced solving techniques and taking advantage of fundamental characteristics of power systems. Four main objectives can be associated with the project as below: 1) Modeling the problem linearly without any approximations, 2) Carrying out the process in real-time even in the case of large-scale distribution grids, 3) Obviating the need for data synchronization while keeping the grid equations linear, 4) Incorporating pseudo-measurements to increase accuracy. This project and its outputs will contribute to UK strengths and are aligned with the EPSRC delivery plan to build a resilient and connected nation. Real-time monitoring of distribution systems will enhance the electricity grids' resilience, thereby contributing to the "Energy security and efficiency" priority area that EPSRC defines as important to deliver its output of a "Resilient Nation". The project also relates to the ambitions set out by EPSRC for Energy Networks research areas. It ensures "affordable solutions for national needs" by exploiting existing measurement technologies and infrastructure in the UK electricity grids.