Using Vine Copulas to Generate Representative System States for Machine Learning

Konstantelos, Ioannis; Sun, Mingyang; Tindemans, Simon H.; Issad, Samir; Panciatici, Patrick; Štrbac, Goran

doi:10.1109/tpwrs.2018.2859367

Cited by 36 publications

(24 citation statements)

References 36 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Recently, given the importance of the database generation step, papers focusing mainly on building more effective databases for machine learning-based security assessment and control were published [21], [23], [24]. In [23], the authors propose an approach using Vine-Copulas to generate more representative states of power systems. They show on a testcase that the security classifier built with this approach is superior to the one build with data obtained from a classical approach.…”

Section: A Database Buildingmentioning

confidence: 99%

Recent Developments in Machine Learning for Energy Systems Reliability Management

2020

View full text Add to dashboard Cite

This paper reviews recent works applying machine learning techniques in the context of energy systems reliability assessment and control. We showcase both the progress achieved to date as well as the important future directions for further research, while providing an adequate background in the fields of reliability management and of machine learning. The objective is to foster the synergy between these two fields and speed up the practical adoption of machine learning techniques for energy systems reliability management. We focus on bulk electric power systems and use them as an example, but we argue that the methods, tools, etc. can be extended to other similar systems, such as distribution systems, micro-grids, and multi-energy systems.

show abstract

Section: A Database Buildingmentioning

confidence: 99%

Recent Developments in Machine Learning for Energy Systems Reliability Management

2020

View full text Add to dashboard Cite

show abstract

“…The C-vine copula model is based on two-dimensional copula theory to model the distribution function and conditional distribution function, then multiply the PDF of the two-dimensional copula to obtain the PDFs of multidimensional RVs [13], [16], [18]. The C-vine copula modeling process is shown in Fig.…”

Section: C-vine Copula Modelmentioning

confidence: 99%

“…Fig. 6 characterizes the dependencies among RVs in the form of a DAG, from which it can be seen that there is a complex dependence between the wind speeds (nodes 1-9), between the various solar irradiations (nodes [11][12][13][14][15][16][17], and between the loads (node 10). When the BN structure and the discrete probability values of the 17 nodes are known, the MLE method is used to learn the network parameters, and a conditional probability table between the nodes is obtained.…”

Section: Wind Speed Solar Irradiation and Load Probability Modmentioning

confidence: 99%

See 1 more Smart Citation

Probabilistic Computational Model for Correlated Wind Speed, Solar Irradiation, and Load Using Bayesian Network

Wang

Zou

2020

IEEE Access

View full text Add to dashboard Cite

This paper proposes a Bayesian network (BN) that can construct the nonlinear dependence among wind speed, solar irradiation, and load. The correlations of random variables (RVs) are analyzed using the Pearson correlation coefficient, Kendall rank correlation coefficient, and Spearman rank correlation coefficient. According to Bayesian theory, the Bayesian information criterion (BIC) and maximum likelihood estimation (MLE) methods are employed to determine the structure and parameters of a BN. Then the BN model of RVs is established. The constructed model is the joint probability distribution of RVs that can present the nonlinear dependence and marginal distribution (MD) of RVs without limitation. The testing samples of wind speed, solar irradiation, and load are generated using the autoregressive integrated moving average (ARMA) model. Then these samples are utilized to construct the probability model of a BN and C-vine copula, whose modeling accuracy and efficiency are compared, and the quality of their output synthetic samples are analyzed. In the modified IEEE 118-bus test system, two kinds of synthetic samples are used to calculate the probabilistic load flow (PLF), whose accuracy and efficiency based on the BN are tested. The validity of the BN model is verified.

show abstract

“…An alternative method to generate training data is to use a scenario generator 16 or vine copulas. 21 The creation of training data with these methods is especially useful if no time series data are available or to obtain additional data for future planning. In section5, we show prediction results when using the scenario generator.…”

Section: Training Datamentioning

confidence: 99%

Evaluating machine learning models for the fast identification of contingency cases

2020

View full text Add to dashboard Cite

Fast approximations of power flow results are beneficial in power system planning and live operation. In planning, millions of power flow calculations are necessary if multiple years, different control strategies, or contingency policies are to be considered. In live operation, grid operators must assess if grid states comply with contingency requirements in a short time. In this paper, we compare regression and classification methods to either predict multivariable results, for example, bus voltage magnitudes and line loadings, or binary classifications of time steps to identify critical loading situations. We test the methods on three realistic power systems based on time series in 15 and 5 minutes resolution of 1 year. We compare different machine learning models, such as multilayer perceptrons (MLPs), decision trees, k-nearest neighbors, gradient boosting, and evaluate the required training time and prediction times as well as the prediction errors. We additionally determine the amount of training data needed for each method and show results, including the approximation of untrained curtailment of generation. Regarding the compared methods, we identified the MLPs as most suitable for the task. The MLP-based models can predict critical situations with an accuracy of 97% to 98% and a very low number of false negative predictions of 0.0% to 0.64%. K E Y W O R D S contingency analysis, grid planning, machine learning, power flow, time series calculation 1 | INTRODUCTION Power flow (PF) results are the basis for power system planning and are needed in live operation to assess the system state. Quasistatic time series simulations allow evaluating asset loadings, voltage profiles, and contingency situations over a long period, for example, multiple years. This has several advantages in the planning process compared to single "worst-case" analysis including the calculation of grid losses or the integration of demand and generation flexibility. 1,2 However, the computational effort is very high. Millions of PF calculations are necessary if multiple years, different

show abstract

Using Vine Copulas to Generate Representative System States for Machine Learning

Cited by 36 publications

References 36 publications

Recent Developments in Machine Learning for Energy Systems Reliability Management

Recent Developments in Machine Learning for Energy Systems Reliability Management

Probabilistic Computational Model for Correlated Wind Speed, Solar Irradiation, and Load Using Bayesian Network

Evaluating machine learning models for the fast identification of contingency cases

Contact Info

Product

Resources

About