Embedded Files
Yijun has devoted his efforts and passion into developing uncertainty-related techniques for the modern power system. These techniques aim to improve the efficiency and accuracy in power system planning, guarantee the reliability of power system operations and enhance their resilience to cascading failures leading to a blackout.
General Research Framework
In my research of uncertainty in power system, I not only focus on power system applications, but also its algorithm development.
In my research of uncertainty in power system, I not only focus on power system applications, but also its algorithm development.
Furthermore, I aim to build a systematic framework in solving a general uncertainty-related problems.
Furthermore, I aim to build a systematic framework in solving a general uncertainty-related problems.
Now, the framework includes uncertainty modeling, forward uncertainty quantification (UQ), inverse uncertainty quantification and decision-making under uncertainty.
Now, the framework includes uncertainty modeling, forward uncertainty quantification (UQ), inverse uncertainty quantification and decision-making under uncertainty.
Uncertainty Modeling
An accurate uncertainty modeling is the prerequisite for a forward uncertainty quantification and its associated decision-making under uncertainty problem.
An accurate uncertainty modeling is the prerequisite for a forward uncertainty quantification and its associated decision-making under uncertainty problem.
The high-dimension uncertainty modeling is indeed not an easy task considering the non-parametric, non-Gaussian, asymmetric probability distributions. Not to mention the complicated spatial-temporal dependence relationship.
The high-dimension uncertainty modeling is indeed not an easy task considering the non-parametric, non-Gaussian, asymmetric probability distributions. Not to mention the complicated spatial-temporal dependence relationship.
Special techniques, such as dimension reduction, can sometimes enable the high-dimensional uncertainty modeling move to a low-dimensional latent space.
Special techniques, such as dimension reduction, can sometimes enable the high-dimensional uncertainty modeling move to a low-dimensional latent space.
Forward Uncertainty Quantification
Forward uncertainty quantification is the core of the uncertainty analysis. It tries to determine how likely certain outcomes are if some aspects of the system are not exactly known. This problem is vital in the modern power systems for their inherently stochastic in nature.
Forward uncertainty quantification is the core of the uncertainty analysis. It tries to determine how likely certain outcomes are if some aspects of the system are not exactly known. This problem is vital in the modern power systems for their inherently stochastic in nature.
This is mainly due to the stochastic load variations over time, the intermittency of renewable energy resources and the intrinsic random outages of its components, such as the generating units, lines, and transformers, to cite a few.
This is mainly due to the stochastic load variations over time, the intermittency of renewable energy resources and the intrinsic random outages of its components, such as the generating units, lines, and transformers, to cite a few.
Therefore, the cost-effective probabilistic analysis tools for the system operation, planning and control are developed by us for the applications like probabilistic power flow, probabilistic load margin, stochastic economic dispatch forecast, statistic dynamic simulations, risk assessment of rare events, etc.
Therefore, the cost-effective probabilistic analysis tools for the system operation, planning and control are developed by us for the applications like probabilistic power flow, probabilistic load margin, stochastic economic dispatch forecast, statistic dynamic simulations, risk assessment of rare events, etc.
Inverse Uncertainty Quantification
Inverse uncertainty quantification is in general more complicated than the forward uncertainty quantification problem.
Inverse uncertainty quantification is in general more complicated than the forward uncertainty quantification problem.
Using the measurement data, the inverse uncertainty quantification can estimate the unknown states, parameters or even model discrepancy of the power systems.
Using the measurement data, the inverse uncertainty quantification can estimate the unknown states, parameters or even model discrepancy of the power systems.
Using uncertainty quantification techniques, we have developed multiple techniques in the power system dynamic state and parameter estimations.
Using uncertainty quantification techniques, we have developed multiple techniques in the power system dynamic state and parameter estimations.
Decision-Making Under Uncertainty
Decision-making under uncertainty is based on uncertainty modeling and forward uncertainty quantification.
Decision-making under uncertainty is based on uncertainty modeling and forward uncertainty quantification.
A proper UQ-informed decision-making under uncertainty scheme usually achieves a more conservative decision compared with a traditional deterministic decision-making scheme while obtaining a less conservative decision compared with a fully robust scheme.
A proper UQ-informed decision-making under uncertainty scheme usually achieves a more conservative decision compared with a traditional deterministic decision-making scheme while obtaining a less conservative decision compared with a fully robust scheme.
To achieve this, we have developed both individual and joint chance-constrained optimal power flow techniques using the advanced UQ techniques.
To achieve this, we have developed both individual and joint chance-constrained optimal power flow techniques using the advanced UQ techniques.
Page updated
Google Sites
Report abuse