Machine Learning and Data-Driven Techniques for the Control of Smart Power Generation Systems: An Uncertainty Handling Perspective

Sun, Li; You, Fengqi

doi:10.1016/j.eng.2021.04.020

Cited by 115 publications

(33 citation statements)

References 114 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Electrical power generation from PV array and WT farm is highly uncertain and is affected by solar irradiance and wind speeds [41][42][43][44]. Accordingly, the accurate modeling of the generators of the PV panels and the wind turbine is essential.…”

Section: Modeling Of Renewable Energy Uncertaintymentioning

confidence: 99%

Solution of Probabilistic Optimal Power Flow Incorporating Renewable Energy Uncertainty Using a Novel Circle Search Algorithm

Shaheen

Ullah

Qais³

et al. 2022

Energies

View full text Add to dashboard Cite

Integrating renewable energy sources (RESs) into modern electric power systems offers various techno-economic benefits. However, the inconsistent power profile of RES influences the power flow of the entire distribution network, so it is crucial to optimize the power flow in order to achieve stable and reliable operation. Therefore, this paper proposes a newly developed circle search algorithm (CSA) for the optimal solution of the probabilistic optimal power flow (OPF). Our research began with the development and evaluation of the proposed CSA. Firstly, we solved the OPF problem to achieve minimum generation fuel costs; this used the classical OPF. Then, the newly developed CSA method was used to deal with the probabilistic power flow problem effectively. The impact of the intermittency of solar and wind energy sources on the total generation costs was investigated. Variations in the system’s demands are also considered in the probabilistic OPF problem scenarios. The proposed method was verified by applying it to the IEEE 57-bus and the 118-bus test systems. This study’s main contributions are to test the newly developed CSA on the OPF problem to consider stochastic models of the RESs, providing probabilistic modes to represent the RESs. The robustness and efficiency of the proposed CSA in solving the probabilistic OPF problem are evaluated by comparing it with other methods, such as Genetic Algorithm (GA), Particle Swarm Optimization (PSO), and the hybrid machine learning and transient search algorithm (ML-TSO) under the same parameters. The comparative results showed that the proposed CSA is robust and applicable; as evidence, an observable decrease was obtained in the costs of the conventional generators’ operation, due to the penetration of renewable energy sources into the studied networks.

show abstract

Section: Modeling Of Renewable Energy Uncertaintymentioning

confidence: 99%

Solution of Probabilistic Optimal Power Flow Incorporating Renewable Energy Uncertainty Using a Novel Circle Search Algorithm

Shaheen

Ullah

Qais³

et al. 2022

Energies

View full text Add to dashboard Cite

show abstract

“…To obtain the representative days, multiple types of clustering approaches were applied in the existing literature. − In this work, we propose a novel approach assisted by machine learning to improve the performance of the existing clustering-based approaches for obtaining representative days. The approach couples PCA with clustering techniques, as PCA can help to deal with the correlations of high-dimensional data in sustainable systems planning. − The clustering approaches include AHC, Gaussian mixture model (GMM), Dirichlet process mixture model (DPMM), and K-means clustering. − The clustering data are the 24-dimension hourly power loads for all days in a year, based on which we investigate the performances of using PCA coupled with each clustering approach and the performances using the existing clustering approaches without PCA. The clustering performances are evaluated by three metrics, namely, intra-cluster variance, inter-cluster variance, and the Calinski–Harabasz index.…”

Section: Multi-scale Bottom-up Electricity Transition Optimization Fr...mentioning

confidence: 99%

Toward Carbon-Neutral Electric Power Systems in the New York State: a Novel Multi-Scale Bottom-Up Optimization Framework Coupled with Machine Learning for Capacity Planning at Hourly Resolution

Zhao

You

2021

ACS Sustainable Chem. Eng.

Self Cite

View full text Add to dashboard Cite

In this work, we propose a novel multi-scale bottom-up optimization framework for the carbon-neutral transition planning of the electric power sector, which incorporates hourly time scale and electricity storage to address the reliability and energy balance issues of the future deep-decarbonized power systems. In addition to the technology and capacity information for each facility, the proposed framework also accounts for facility ages, which are usually omitted in the literature, without significantly increasing the computational demand. To reduce the computational requirement of simultaneously optimizing capacity planning and hourly systems operations over the next few decades, a reduced model is developed based on representative days, using a novel approach that integrates multiple machine learning techniques. Based on the optimal transition pathways, hourly operational simulations are conducted for every year within the planning horizon to obtain detailed optimization results. To illustrate the applicability of the proposed framework, a case study for the New York State is presented through two cases, with and without electricity storage capacity expansion. The proposed approach using principal component analysis coupled with K-means outcompetes multiple conventional approaches of using clustering techniques directly. The transition planning results show that the total generation capacity for the case with electricity capacity expansion is 39% higher than the other case, while the latter case has 200% more generation capacity from non-intermittent sources. Detailed hourly operational simulation results indicate that offshore wind, hydro, and utility solar are the primary power sources by 2040 for the case with electricity storage capacity expansion, while hydro, offshore wind, and nuclear are the main electricity sources for the other case.

show abstract

“…The smart particle is based on the convergence factor (CF) technique, which combines memory of particle positions, the second stage is for comparison, and finally the leader declaration, to find the best optimal solution. Furthermore, some researchers have worked on energy system management and design algorithms for the purpose of developing smart artificial intelligence [8][9][10][11][12][13].…”

Section: Introductionmentioning

confidence: 99%

A Multimodal Improved Particle Swarm Optimization for High Dimensional Problems in Electromagnetic Devices

et al. 2021

View full text Add to dashboard Cite

Particle Swarm Optimization (PSO) is a member of the swarm intelligence-based on a metaheuristic approach which is inspired by the natural deeds of bird flocking and fish schooling. In comparison to other traditional methods, the model of PSO is widely recognized as a simple algorithm and easy to implement. However, the traditional PSO’s have two primary issues: premature convergence and loss of diversity. These problems arise at the latter stages of the evolution process when dealing with high-dimensional, complex and electromagnetic inverse problems. To address these types of issues in the PSO approach, we proposed an Improved PSO (IPSO) which employs a dynamic control parameter as well as an adaptive mutation mechanism. The main proposal of the novel adaptive mutation operator is to prevent the diversity loss of the optimization process while the dynamic factor comprises the balance between exploration and exploitation in the search domain. The experimental outcomes achieved by solving complicated and extremely high-dimensional optimization problems were also validated on superconducting magnetic energy storage devices (SMES). According to numerical and experimental analysis, the IPSO delivers a better optimal solution than the other solutions described, particularly in the early computational evaluation of the generation.

show abstract

Machine Learning and Data-Driven Techniques for the Control of Smart Power Generation Systems: An Uncertainty Handling Perspective

Cited by 115 publications

References 114 publications

Solution of Probabilistic Optimal Power Flow Incorporating Renewable Energy Uncertainty Using a Novel Circle Search Algorithm

Solution of Probabilistic Optimal Power Flow Incorporating Renewable Energy Uncertainty Using a Novel Circle Search Algorithm

Toward Carbon-Neutral Electric Power Systems in the New York State: a Novel Multi-Scale Bottom-Up Optimization Framework Coupled with Machine Learning for Capacity Planning at Hourly Resolution

A Multimodal Improved Particle Swarm Optimization for High Dimensional Problems in Electromagnetic Devices

Contact Info

Product

Resources

About