Generator Parameter Estimation by Q-Learning Based on PMU Measurements

Khazeiynasab, Seyyed Rashid; Qi, Junjian; Batarseh, Issa

doi:10.1109/isgt49243.2021.9372221

Cited by 11 publications

(8 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Based on the engineering experience [12], the steady-state mismatch can be determined strongly by the turbine governor model. Some parameters such as T 6 and T 5 only shift the total waveform of the generator outputs.…”

Section: Lstm Based Parameter Estimationmentioning

confidence: 99%

“…It is possible that for different events, the algorithm may bring different parameter packs. In addition to engineering judgment and experience that can help choose reasonable parameters [12], techniques can also be developed based on the available multiple events to help select the best parameters. For example, one can consider one event to estimate the parameters and use the other events to cross-validate the estimated parameters.…”

Section: Higher-dimensional Casementioning

confidence: 99%

“…The proposed method in [11] generates extensive simulation data to train a multi-output convolutional neural network to estimate the small number of generator parameters. The proposed qlearning method [12], deep q-learning-based method [13], and soft actor-critic deep learning-based method [14] are extremely dependent on the hyper parameters and reward functions. It is challenging, especially for high-dimensional cases, to design a suitable reward function, to find the global optimum as the estimated parameters.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

A Parallel Multi-Modal LSTM for Power System Model Calibration

Khazeiynasab¹

2022

Preprint

Self Cite

View full text Add to dashboard Cite

<div>Renewable energies, smart loads, energy storage, and new market behavior are adding new sources of uncertainty to power systems. Therefore, planning in real-time and developing high-quality models is crucial to adapt to uncertainties. Model validation based on actual measurements is necessary for obtaining accurate representations of power systems dynamics with system uncertainties. This paper presents a new measurement-based method to calibrate the parameters of a synchronous generator by deep learning method based on the long-short term memory (LSTM) network. First, critical parameters are determined regarding the active/reactive behavior of the generator. Then, a parallel multimodal LSTM (PM-LSTM) is designed with flexible input time steps to capture important features of temporal patterns from time-series measurements. The extracted features are then fed into a dense layer to capture the joint representation</div><div>of inputs. The simulations conducted for a hydro generator under different events show that the proposed method can estimate the model parameters accurately and efficiently.</div>

show abstract

Section: Lstm Based Parameter Estimationmentioning

confidence: 99%

Section: Higher-dimensional Casementioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

A Parallel Multi-Modal LSTM for Power System Model Calibration

Khazeiynasab¹

2022

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…It has recently been applied to the power systems calibration problem. We categorize these methods to CNN based methods [11], [12], DRL based approaches [13]- [16] and deep unsupervised approaches [17].…”

Section: Introductionmentioning

confidence: 99%

A Deep Learning-Based Framework for Parameters Calibration of Power Plant Models Using Event Playback Approach

Matar

Elmoudi

et al. 2022

IEEE Access

View full text Add to dashboard Cite

Accurate and reliable calibration of power plant models in the power system is crucial for maintaining grid stability and preventing power outages. This becomes even more crucial as the power grids become more dynamic and stochastic with the ever-increasing penetration of renewable energy sources, mid-size generators, smart loads, and energy storage. Current practice such as staged-test is costly and timeconsuming. Alternatively, recent PMU-based approaches using online measurements without interfering with generators operation have been introduced to provide a scalable and low-cost to meet North American Electric Reliability Corporation (NERC) standards. This research paper proposes a PMU-based framework to validate and calibrate model parameters by a new iterative deep learning approach. The results show that the proposed approach can accurately calibrate power plantmodels from a single disturbance event without the urgent need of prior knowledge of the model's initial parameters. The proposed method has been validated on real-world and simulated scenarios on a relatively large number of calibrated parameters with an average error of 1.43%.

show abstract

“…Since the model parameters depend on the operating conditions, the offline methods can not estimate the exact values of the parameters. In the online methods, real-time data or simulation-based data are obtained and used to identify the parameters of the system [9].…”

Section: Introductionmentioning

confidence: 99%

Measurement-Based Parameter Identification of DC-DC Converters with Adaptive Approximate Bayesian Computation

Khazeiynasab¹,

Batarseh²

2021

Preprint

Self Cite

View full text Add to dashboard Cite

The recent advances in power plants and energy resources have extended the applications of DC-DC converters in the power systems (especially in the context of DC microgrids). Parameter identification can extract the parameters of the converters and generate accurate discrete simulation models. In this paper, we propose a measurement-based converter parameter calibration method by an adaptive Approximate Bayesian Computation with sequential Monte Carlo sampler (ABC SMC), which estimates the parameters related to passive and parasitic components. At first, we propose to find suitable prior distribution for the parameter which we don't know the prior information about them. With having prior distributions, we can use the ABC SMC to find the exact values of the parameters of the converter. We chose the distance function carefully and based on the simulations we assigned the best method for the threshold sequencing. For improving the computationally of the algorithm, we propose an adaptive weight that helps the algorithm to find the optimal values with fewer simulations. The effectiveness of the proposed method is validated for a DC-DC buck converter. The results show that the proposed approach can accurately and efficiently estimate the posterior distributions of the buck parameters subject to gross errors in the prior distributions of the parameters. The proposed algorithm can also be applied to other parameter identifications and optimization applications such as rectifiers, filters, or power supplies, among others.

show abstract

Generator Parameter Estimation by Q-Learning Based on PMU Measurements

Cited by 11 publications

References 19 publications

A Parallel Multi-Modal LSTM for Power System Model Calibration

A Parallel Multi-Modal LSTM for Power System Model Calibration

A Deep Learning-Based Framework for Parameters Calibration of Power Plant Models Using Event Playback Approach

Measurement-Based Parameter Identification of DC-DC Converters with Adaptive Approximate Bayesian Computation

Contact Info

Product

Resources

About