Statistical Analysis and Forecasting of Damping in the Nordic Power System

Abstract

This paper presents an application of multiple linear regression (MLR) to extract significant correlations between damping of electromechanical modes and system operating conditions and to forecast future damping values, based on existing day-ahead market forecasts for power flows and generation. The presented analysis uses measurements from the Nordic power system. First, a static MLR model is developed to explain the variability of the damping of the 0.35-Hz inter-area mode in the Nordic system. Together with the static model, a dynamic MLR model is used for forecasting the damping 24 hours ahead, using day-ahead market forecasts. Test results indicate the proposed methods are able to correctly predict about 90% of the low damped operating conditions observed during a year, if day-ahead market forecasts are accurate. These results suggest that the methods could be used to issue early warnings about future operating conditions with low damping.Index Terms-Damping forecast, electromechanical oscillations, multiple linear regression (MLR), phasor measurement unit (PMU) data.

A review on short‐term load forecasting models for micro‐grid application

“…Heteroscedasticity occurs whenever the confidence intervals of a prediction are different for each time. 2 In the STLF environment, different 'Auto-regressive conditional heteroscedasticity (ARCH)' approaches are as follows. The 'Generalized autoregressive conditional heteroscedasticity (GARCH)' is often used in a single variable as well as multi-variable scenario as in TSA [59].To predict energy usage and gasoline consumption in China, ANN was put to use to integrate 'seasonal generalized autoregressive conditional heteroscedasticity (SEGARCH)' and 'Winter model with an exponential form of generalized autoregressive conditional heteroscedasticity (WARCH)', two regression models, this approach has better forecasting results.…”

Statistical analysis on the bandwidth of a dual frequency textile antenna

“…The model adequacy assessment is argued in detail here before the discussion of ANOVA results as the results make sense only if model passes the adequacy assessment. To verify that it is sufficient to include only linear terms for the data, we performed the four tests for model adequacy checking which confirms the linear relation between the variable has also been performed as described in [29,30]. The model assessment of the data is done with the aid of residual plots.…”

“…In the above equation Y i is the observed value and Y i ′ is the predicted value. The relation derived for the lower and upper bandwidths is given by (30) and 31…”

“…The correlation coefficient for the upper bandwidth is calculated as 0.9165. To validate the relation given by (30) and (31), bandwidth is measured for specific values of bending, shear and stretching. The comparison graphs of the lower and upper bandwidths predicted through the regression equations and measured for validation is given in Figs.…”

Power System Oscillation Mode Prediction Based on the Lasso Method

“…Between these two categories, the former approach dominates the other. For instance, linear regression was implemented in [7], [8] to identify the damping sensitivity at an operation point, while weighted linear regression was employed in [9] to find critical variables in the identification procedure. Apart from the regression-type methods, in [10] the problem was formulated as a pattern classification problem, and decision trees were used to predict well or poorly damped oscillations using power flow data.…”