Discrete time optimal control and its applications in power systems

• Main Author: Zhang, Hehong
• Other Authors: Xiao Gaoxi
• Format: Thesis-Doctor of Philosophy
• Language: English
• Published: Nanyang Technological University 2020
• Subjects: Engineering::Electrical and electronic engineering
• Online Access: https://hdl.handle.net/10356/137191
• Institution: Nanyang Technological University

Almost all human systems and activities strongly depend on reliable electricity. However, with the expansion of interconnected power systems and the adoption of new cyber-physical technologies in smart grid, the challenges including occasional wide-area blackouts and stability issues have emerged. State estimation (SE) that serves to estimate the real-time operating system states is among the most fundamental and critical functions in enhancing the power system resilience, since it provides system operators with better situational awareness for decision-making. Motivated by the urgent need of accurate and efficient SE algorithms to help operators make appropriate decisions, my research has two focuses:Almost all human systems and activities strongly depend on reliable electricity. However, with the expansion of interconnected power systems and the adoption of new cyberphysical technologies in smart grid, the challenges including occasional wide-area blackouts and stability issues have emerged. State estimation (SE) that serves to estimate the real-time operating system states is among the most fundamental and critical functions in enhancing the power system resilience, since it provides system operators with better situational awareness for decision-making. Motivated by the urgent need of accurate and e cient SE algorithms to help operators make appropriate decisions, my research has two focuses:Almost all human systems and activities strongly depend on reliable electricity. However, with the expansion of interconnected power systems and the adoption of new cyberphysical technologies in smart grid, the challenges including occasional wide-area blackouts and stability issues have emerged. State estimation (SE) that serves to estimate the real-time operating system states is among the most fundamental and critical functions in enhancing the power system resilience, since it provides system operators with better situational awareness for decision-making. Motivated by the urgent need of accurate and efficient SE algorithms to help operators make appropriate decisions, my research has two focuses: i Design practical, accurate and efficient state estimators. ii Manage state estimations in power systems integrating PMUs measurements. The first part of the thesis focuses on construction and theoretical analysis of SE algorithms via discrete-time optimal control (DTOC). A closed-form mathematical derivation for the DTOC algorithm was fi rst proposed by Han, which has been widely used to design estimators, observers and noise-tolerant differentiators. I fi rst investigate the convergence of this DTOC algorithm to make its application technically sound. To cream o the best and filter out the impurities, a novel time optimal control algorithm with a simple structure is constructed with the method of the isochronic regions. Speci fically, a linearised criterion is adopted for determining the control signal sequence in this approach. The convergence of this new time optimal control algorithm is proved by demonstrating the convergence path of the state point sequence driven by the corresponding control signal sequence. Furthermore, a novel tracking differentiator (TD) based on the proposed DTOC algorithm is presented. Simulation and experiment results show that it performs well in signal-tracking, differentiation acquisition and helps reduce the computational resources needed. Then, I conduct state estimations in power systems. Speci fically, the field PMUs data are filtered using the proposed TD based on the DTOC algorithm. To test its effectiveness, I compare the system emergency control schemes' solutions and the corresponding control results during cascading failures based on the filtered data and the noise-free PMUs data, respectively. Not requiring complex power system modelling and historical data, the proposed TD is suitable for real-time synchrophasor estimation application especially when the states are corrupted by noises. The aforementioned SE algorithms can effectively reduce the computational resources needed and meet the requirements of state estimations in power systems. The proposed approaches may provide a new methodology to realize more accurate and efficient SE in power systems. The obtained results can provide system operators with better situational awareness to take prompt emergency control actions against failures, and in turn, improve the power systems' resilience.