Menu
Mechatronics Research Lab Publications
List still being populated due to webpage maintenance
2019
4.
Kevin Vanslette; Arwa Alanqari; Zeyad Al-awwad; Kamal Youcef-Toumi
Vectorized Uncertainty Propagation and Input Probability Sensitivity Analysis Journal Article
In: 2019.
Abstract | Links | BibTeX | Tags: Algorithms, Computational Intelligence, Experimentation, Uncertainty estimation and calibration for modeling
@article{MRL_AFM_Vectorized_Uncertainty_Input_Probability,
title = {Vectorized Uncertainty Propagation and Input Probability Sensitivity Analysis},
author = {Kevin Vanslette and Arwa Alanqari and Zeyad Al-awwad and Kamal Youcef-Toumi},
url = {https://arxiv.org/abs/1908.11246},
doi = {1908.11246v1},
year = {2019},
date = {2019-06-04},
publisher = {King Abdulaziz City for Science and Technology},
abstract = {In this article we construct a theoretical and computational process for assessing Input Probability Sensitivity Analysis (IPSA) using a Graphics Processing Unit (GPU) enabled technique called Vectorized Uncertainty
Propagation (VUP). VUP propagates probability distributions through
a parametric computational model in a way that’s computational time
complexity grows sublinearly in the number of distinct propagated input probability distributions. VUP can therefore be used to efficiently
implement IPSA, which estimates a model’s probabilistic sensitivity to
measurement and parametric uncertainty over each relevant measurement
location. Theory and simulation illustrate the effectiveness of these methods.},
keywords = {Algorithms, Computational Intelligence, Experimentation, Uncertainty estimation and calibration for modeling},
pubstate = {published},
tppubtype = {article}
}
In this article we construct a theoretical and computational process for assessing Input Probability Sensitivity Analysis (IPSA) using a Graphics Processing Unit (GPU) enabled technique called Vectorized Uncertainty
Propagation (VUP). VUP propagates probability distributions through
a parametric computational model in a way that’s computational time
complexity grows sublinearly in the number of distinct propagated input probability distributions. VUP can therefore be used to efficiently
implement IPSA, which estimates a model’s probabilistic sensitivity to
measurement and parametric uncertainty over each relevant measurement
location. Theory and simulation illustrate the effectiveness of these methods.
Propagation (VUP). VUP propagates probability distributions through
a parametric computational model in a way that’s computational time
complexity grows sublinearly in the number of distinct propagated input probability distributions. VUP can therefore be used to efficiently
implement IPSA, which estimates a model’s probabilistic sensitivity to
measurement and parametric uncertainty over each relevant measurement
location. Theory and simulation illustrate the effectiveness of these methods.
2015
3.
Bo Jiang; Amro M Farid; Kamal Youcef-Toumi
A comparison of day-ahead wholesale market: Social welfare vs industrial demand side management Proceedings Article
In: 2015 IEEE International Conference on Industrial Technology (ICIT), pp. 2742-2749, IEEE IEEE, 2015, ISBN: 978-1-4799-7800-7.
Abstract | Links | BibTeX | Tags: Algorithms, intelligent systems, Physical System Modeling, Simulation, Uncertainty estimation and calibration for modeling
@inproceedings{MRL_UEC_Welfare_vs_DSM,
title = {A comparison of day-ahead wholesale market: Social welfare vs industrial demand side management},
author = {Bo Jiang and Amro M Farid and Kamal Youcef-Toumi},
url = {https://ieeexplore.ieee.org/abstract/document/7125502?section=abstract},
doi = {10.1109/ICIT.2015.7125502},
isbn = {978-1-4799-7800-7},
year = {2015},
date = {2015-06-18},
booktitle = {2015 IEEE International Conference on Industrial Technology (ICIT)},
pages = {2742-2749},
publisher = {IEEE},
organization = {IEEE},
abstract = {The intermittent nature of renewable energy has been discussed in the context of the operational challenges that it brings to electrical grid reliability. In contrast, Demand Side Management (DSM) with its ability to allow customers to adjust electricity consumption in response to market signals has often been recognized as an efficient way to mitigate the variable effects of renewable energy. However, the industrial & academic literature have taken divergent approaches to DSM implementation. Academic studies often implement demand side management on the basis of a social welfare maximization. Meanwhile, industrial implementations minimize total system costs where customers are compensated for load reductions from a predefined baseline of electricity consumption that would have occurred without DSM. This paper rigorously compares these two different approaches in a day-ahead wholesale market context using the same system configuration and mathematical formalism. The comparison showed that a proper reconciliation between the dispatchable demand utility function and the load reduction cost function lead to fundamentally the same stochastic netload mitigation and the two DSM models generate the same dispatch results under specific conditions. However, while the social welfare model uses a stochastic net load composed of two terms, the industrial DSM model uses a stochastic net load composed of three terms, and is thus more prone to error and more likely requires more control activity in subsequent layers of enterprise control.},
keywords = {Algorithms, intelligent systems, Physical System Modeling, Simulation, Uncertainty estimation and calibration for modeling},
pubstate = {published},
tppubtype = {inproceedings}
}
The intermittent nature of renewable energy has been discussed in the context of the operational challenges that it brings to electrical grid reliability. In contrast, Demand Side Management (DSM) with its ability to allow customers to adjust electricity consumption in response to market signals has often been recognized as an efficient way to mitigate the variable effects of renewable energy. However, the industrial & academic literature have taken divergent approaches to DSM implementation. Academic studies often implement demand side management on the basis of a social welfare maximization. Meanwhile, industrial implementations minimize total system costs where customers are compensated for load reductions from a predefined baseline of electricity consumption that would have occurred without DSM. This paper rigorously compares these two different approaches in a day-ahead wholesale market context using the same system configuration and mathematical formalism. The comparison showed that a proper reconciliation between the dispatchable demand utility function and the load reduction cost function lead to fundamentally the same stochastic netload mitigation and the two DSM models generate the same dispatch results under specific conditions. However, while the social welfare model uses a stochastic net load composed of two terms, the industrial DSM model uses a stochastic net load composed of three terms, and is thus more prone to error and more likely requires more control activity in subsequent layers of enterprise control.
2014
2.
B N Shapiro; R Adhikari; J Driggers; J Kissel; B Lantz; J Rollins; K Youcef-Toumi
Noise and control decoupling of Advanced LIGO suspensions Journal Article
In: Classical and Quantum Gravity, vol. 32, no. 1, pp. 015004, 2014.
Abstract | Links | BibTeX | Tags: Control Theory, Experimentation, intelligent systems, Physical System Modeling, Simulation, Uncertainty estimation and calibration for modeling, Visualization
@article{MRL_AFM_Noise_Control_Decoupling,
title = {Noise and control decoupling of Advanced LIGO suspensions},
author = {B N Shapiro and R Adhikari and J Driggers and J Kissel and B Lantz and J Rollins and K Youcef-Toumi},
url = {https://doi.org/10.1088/0264-9381/32/1/015004},
doi = {10.1088/0264-9381/32/1/015004},
year = {2014},
date = {2014-12-10},
journal = {Classical and Quantum Gravity},
volume = {32},
number = {1},
pages = {015004},
publisher = {IOP Publishing},
abstract = {Ground-based interferometric gravitational wave observatories such as Advanced LIGO must isolate their optics from ground vibrations with suspension systems to meet their stringent noise requirements. These suspensions typically have very high quality-factor resonances that require active damping. The sensor noise associated with this damping is a potential significant contributor to the sensitivity of these interferometers. This paper introduces a novel scheme for suspension damping that isolates much of this noise and permits greater amounts of damping. It also decouples the damping feedback design from the interferometer control. The scheme works by invoking a change from a local coordinate frame associated with each suspension, to a coordinate frame aligned with the interferometric readout. In this way, degrees of freedom invisible to the readout can employ effective, but noisy damping. The degree of freedom measured by the readout is then damped using low noise interferometer signals, eliminating the need to use the usual noisy sensors. Simulated and experimental results validate the concepts presented in this paper.},
keywords = {Control Theory, Experimentation, intelligent systems, Physical System Modeling, Simulation, Uncertainty estimation and calibration for modeling, Visualization},
pubstate = {published},
tppubtype = {article}
}
Ground-based interferometric gravitational wave observatories such as Advanced LIGO must isolate their optics from ground vibrations with suspension systems to meet their stringent noise requirements. These suspensions typically have very high quality-factor resonances that require active damping. The sensor noise associated with this damping is a potential significant contributor to the sensitivity of these interferometers. This paper introduces a novel scheme for suspension damping that isolates much of this noise and permits greater amounts of damping. It also decouples the damping feedback design from the interferometer control. The scheme works by invoking a change from a local coordinate frame associated with each suspension, to a coordinate frame aligned with the interferometric readout. In this way, degrees of freedom invisible to the readout can employ effective, but noisy damping. The degree of freedom measured by the readout is then damped using low noise interferometer signals, eliminating the need to use the usual noisy sensors. Simulated and experimental results validate the concepts presented in this paper.
1979
1.
Aramazd Muzhikyan; Amro M Farid; Kamal Youcef-Toumi
An a priori analytical method for the determination of operating reserve requirements Journal Article
In: International Journal of Electrical Power & Energy Systems, vol. 86, pp. 1-17, 1979, ISSN: 0142-0615.
Abstract | Links | BibTeX | Tags: Algorithms, Computational Intelligence, intelligent systems, Physical System Modeling, Simulation, Uncertainty estimation and calibration for modeling
@article{MRL_AFM_Priori_Operating_Requirements,
title = {An a priori analytical method for the determination of operating reserve requirements},
author = {Aramazd Muzhikyan and Amro M Farid and Kamal Youcef-Toumi},
url = {https://www.sciencedirect.com/science/article/pii/S0142061515300089},
doi = {https://doi.org/10.1016/j.ijepes.2016.09.005},
issn = {0142-0615},
year = {1979},
date = {1979-04-01},
journal = {International Journal of Electrical Power & Energy Systems},
volume = {86},
pages = {1-17},
publisher = {ScienceDirect},
abstract = {Power balance is one of the key requirements for reliable power system operation. However, factors, such as net load variability and forecast errors, impose practical limitations on matching the scheduled generation and the real-time demand. Normally, potential power imbalances are mitigated by scheduling additional generation capacity called operating reserves. However, reserves are a costly commodity and their requirements should be accurately assessed to avoid unnecessary expense. Currently, the reserve requirements are determined using a posteriori methods based upon operator’s experience and established assumptions. While these assumptions are made out of a level of engineering practicality, they may not be formally true given the numerical evidence. This paper presents a formal mathematical framework for the a priori determination of three types of operating reserve requirements, namely load following, ramping and regulation. Validation of the methodology is performed by a set of extensive simulations that model the power system operations for different scenarios. This methodology is used to study the sensitivity of each type of reserve requirement to the net load and power system parameters.},
keywords = {Algorithms, Computational Intelligence, intelligent systems, Physical System Modeling, Simulation, Uncertainty estimation and calibration for modeling},
pubstate = {published},
tppubtype = {article}
}
Power balance is one of the key requirements for reliable power system operation. However, factors, such as net load variability and forecast errors, impose practical limitations on matching the scheduled generation and the real-time demand. Normally, potential power imbalances are mitigated by scheduling additional generation capacity called operating reserves. However, reserves are a costly commodity and their requirements should be accurately assessed to avoid unnecessary expense. Currently, the reserve requirements are determined using a posteriori methods based upon operator’s experience and established assumptions. While these assumptions are made out of a level of engineering practicality, they may not be formally true given the numerical evidence. This paper presents a formal mathematical framework for the a priori determination of three types of operating reserve requirements, namely load following, ramping and regulation. Validation of the methodology is performed by a set of extensive simulations that model the power system operations for different scenarios. This methodology is used to study the sensitivity of each type of reserve requirement to the net load and power system parameters.