International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395 -0056 Volume: 04 Issue: 05 | May -2017 www.irjet.net p-ISSN: 2395-0072 © 2017, IRJET | Impact Factor value: 5.181 | ISO 9001:2008 Certified Journal | Page 3458 Short-Term Load Forecasting Using Kalman Filter M. M. Dixit 1 , P.R. Chavan 2 1 PG Student, Electrical Engineering Department, Y.T.I.E.T, Karjat Y.T.I.E.T. Bhivpuri, Karjat, Maharashtra, India 2 Head of Electrical Engineering Department, Y.T.I.E.T, Karjat Y.T.I.E.T. Bhivpuri, Karjat, Maharashtra, India ---------------------------------------------------------------------***------------------------------------------------------------------- Abstract - Short-term load forecasting (STLF) aims towards prediction of electricity loads for a period of minutes, hours, days or weeks. Accurate load forecasting will lead to appropriate scheduling and planning with optimize energy cost. The geographical location, population, social factors, and weather factors have different effects on load patterns. The models adopted for STLF mainly are of time series and casual models. The time series models include the methods based on Kalman filtering approach. In this paper hourly based load forecasting will be carried out by Kalman filter model. A 24- hour municipal load is being considered for the analysis. Key Words: Short-term load forecasting, Kalman filter, Municipal Load 1. INTRODUCTION In recent years, with the opening of electricity markets, electrical power system load forecasting plays an important role for electrical power operation. Accurate load forecast will lead to appropriate operation and planning for the power system, thus achieving a lower operating cost and higher reliability of electricity supply. Short-term load forecasting (STLF) of electric power is important in operation scheduling, economic dispatch, unit commitment, energy transactions and fuel purchasing [1, 2]. Short-term load forecasting aims towards prediction of electricity loads for a period of minutes, hours, days or weeks. The quality of short-term load forecasts with lead time ranging from one hour to several days ahead has significant impact on the efficiency of any power utility [3]. In the developing countries like India the power sector is often unable to meet peak demands. It seems essential that the scheduling of generation is to be planned carefully since one has to work within stringent limits. Hence, suitable strategies are necessary for generation control and load management. For this purpose, short-term load forecasting has to be carried out as accurately as possible. The objectives of STLF are [4]:  To derive the scheduling function that determines the most economic load dispatch with operational constraints and policies, environmental and equipment limitations.  To ensure the security of the power system at any time point.  To provide system dispatchers with timely information. The models adopted for STLF mainly belong to two classes: time series (univariate) models, modeling electric load as a function of only its past recorded values; casual models, modelling the electric load as a function of exogenous variables such as weather and social factors. The time series models include the methods based on Kalman Filtering approach[5, 6]. Owing to the importance of STLF, research in this area in the last years has resulted in the development of numerous forecasting methods [7]. These methods are mainly classified into two categories: classical approaches and artificial intelligence (AI) based techniques. Classical approaches are based on various statistical modeling methods. These approaches forecast future values of the load by using a mathematical combination of previous values of the load and other variable such as weather data. Classical STLF approaches use regression exponential smoothing, Box- Jenkins, autoregressive integrated moving average (ARIMA) models and Kalman filters. Recently several research groups have studied the use of artificial neural networks (ANNs) models and Fuzzy neural networks