# Load Forecasting using Linear Regression Analysis in Time series model for RGUKT, R.K. Valley Campus HT Feeder

DOI : 10.17577/IJERTV6IS050443

Text Only Version

#### Load Forecasting using Linear Regression Analysis in Time series model for RGUKT, R.K. Valley Campus HT Feeder

M. Dinesh Reddy, Lecturer, RGUKT, R.K. Valley, N. Vishali,

Professor, JNTUA CEP, Pulivendula, Andhra Pradesh, India.

Index Terms: Load forecasting, Linear regression, RGUKT, HT Feeder.

INTRODUCTION

Linear Regression: Linear regression is a statistical technique used for finding a relation between two or more variables. If the relation is found between two variables, it is called simple linear regression. If the relation is found for more variables, it is called multi variable linear regression. After finding relation between the variables, it is assumed that the parameters are varying with same relation. Hence the same relation is applied to the forthcoming parameters. Which will give the values of dependent variable value for the corresponding forthcoming independent variable. Linear regression is quite simple method to fit the curve and find the coefficients. The model takes the form y= mx+c. Where m is the slope of the curve and c is the intercept. The parameter x is the independent variable. f(x) is dependent variable. The task is to find the coefficients m and c with the help of available data of x and y.

Performance Evaluation: There are different measures to evaluate the performance of the fore casted result.

Least Squared Error(LSE): It is measured by adding squares of error between actual value and forecasted value.

LSE=

Where Yi is Actual output at ith instant.

Xi is the value of forecasted variable at ith instant. N is number of independent variable instances.

Mean Squared Error(MSE): It is measured by find mean of squares of errors at each and every point.

MSE=

Where Yi is value of Actual dependent variable at ith instant of independent variable.

Xi is the value of forecasted variable at ith instant of . N is number of independent variable instances.

Root Mean Squared Error(RMSE): It is measured by square root of Mean of Square of difference between fore casted variable value and actual output.

RMSE=

Where Yi is value of Actual dependent variable at ith instant of independent variable.

Xi is the value of forecasted variable at ith instant of . N is number of independent variable instances.

Mean Absolute Percentage Error(MAPE): It is also known as Mean Absolute Percentage Deviation (MAPD). It is one of the most accurate and most popular measure of finding the performance.

MAPE=

Where Xi is the value of fore casted variable at ith instant . Yi is the value of Actual output at ith instant

N is number of independent variable instances.

Time Series linear Regression: In the normal linear regression there will be two variables specifically known. But in Time Series Linear Regression time is taken as one variable. That is dependent variable is taken strictly in equal intervals of time. So we have only one variable known specifically. These cases will be known as Time series linear regression.

RESULTS AND CONCLUSION

The data has been collected from RGUKT R.K. Valley substation for the month of January in 2017. The data is collected in the equal intervals of 30 minutes everyday. As the known dependent variable is only Load data, the technique applied is time series Linear Regression. For the data collected Time series linear regression has been applied. As the Load forecasting is heuristic in nature, for different cases different coefficients will be there. For RGUKT, R.K. Valley campus coefficients have been found as 2.197 (slope) and 7008.44(intercept). The error performances obtained are LSE=132.4821, MSE= 6.0219, RMSE= 2.453, MAPE= 0.029.

The graph showing Actual load and fore casted load is shown in figure-1

Figure-1

REFERENCES:

1. Gurley, Numerical Methods Lecture 5 – Curve Fitting Techniques Pages 89-102.

2. https://en.wikipedia.org/wiki/Time_series

3. Mosad Alkhathami, Introduction to Electric Load Forecasting Methods, Columbia International Publishing Journal of Advanced Electrical and Computer Engineering (2015) Vol. 2 No. 1 pp. 1-12.

4. Eugene A. Feinberg, Applied Mathematics for Restructured Electric Power Systems, Chapter-12, Load Forecasting, ISBN 978-0-387-23471-7, Pages 269-285.

5. Kumar Padmanabh, Load Forecasting in India at Distribution Transformer considering Economic Dynamics, Proceedings of Intl. Conference on Advances in Computing, Communications and Informatics (ICACCI), Sept. 21-24, 2016, Jaipur, India.

6. Filipe Rodrigues, Carlos Cardeira, J.M.F.Calado, The daily and hourly energy consumption and load forecasting using artificial neural network method: a case study using a set of

93 households in Portugal, The Mediterranean Green Energy Forum 2014.

7. Wei-Chiang Hong, Electric load forecasting by support vector model, Applied Mathematical Modelling 33 (2009) 24442454.

8. Geoffrey E. Hinton and Simon Osindero, A Fast Learning Algorithm for Deep Belief Nets, Neural Computation 18, 15271554 (2006).

9. Ghulam Mohi Ud Din and Angelos K. Marnerides, Short Term Power Load Forecasting Using Deep Neral Networks, Proceedings of 2017 International Conference on Computing, Networking and Communications (ICNC): Green Computing, Networking, and Communications, Pages: 594 598.

10. Heng Shi, Minghao Xu, Ran Li, Deep Learning for Household Load Forecasting A Novel Pooling Deep RNN, IEEE Transactions on Smart Grid, 2017, Volume PP, Issue 99, Page 1-1.