Complete Material Cost #3,000
Order for Complete Material now
ABSTRACT
Load forecasting has been an issue in electrical power supply and consumption, especially in a developing country like Nigeria where there is always necessity for electrical power expansion due to rapid development. For decades, the problems of improving the accuracy of load forecasts have been an important topic of research. In this work, an artificial neural network approach for short term load forecasting of Bori, in Khana Local Government Area is attempted. Time, temperature and similar previous day load are used for its forecast. ANN role base are prepared which are eventually used for the short term load forecasting. MATLAB SIMULINK software is used here in this work for system designing and simulation for the short-term load forecasting, load data from the specific area load control centre is considered. This helps in reduces the generation cost and increases stability of power systems Key words: ANN, Matlab Simulink, Forecasting.
TABLE OF CONTENTS
Cover page………………………………………………………………………………i
TITLE PAGE II
CERTIFICATION iii
DEDICATION iv
ACKNOWLEDGEMENT v
ABSTRACT vi
TABLE OF CONTENTS vii
CHAPTER ONE 1
1.0 INTRODUCTION 1
1.1 BACKGROUND OF THE STUDY 2
1.2 STATEMENT OF THE PROBLEM 3
1.3 RESEARCH AIM AND OBJECTIVES 3
1.4 SIGNIFICANCE OF THE PROJECT 3
1.5 ASSUMPTION OF THE STUDY 4
1.6 SCOPE OF THE STUDY 5
1.7 LIMITATIONS OF THE STUDY 5
CHAPTER TWO 6
2.0 LITERATURE REVIEW 6
2.1 TYPES OF LOAD FORECASTING 7
2.2 SHORT-TERM LOAD FORECASTING METHODS 7
2.3 IMPLEMENTATION OF BACK PROPAGATION ALGORITHM 12
2.4 MEDIUM TERM LOAD FORECASTING METHOD 15
2.5 LONG-TERM LOAD FORECAST 16
2.6 IMPORTANCE OF LOAD FORECASTING 17
2.7 ADVANTAGES OF LOAD FORECASTING 17
2.8 DISADVANTAGES OF LOAD FORECASTING 18
2.9 FACTORS THAT INFLUENCE ELECTRIC LOAD 19
CHAPTER THREE 25
RESEARCH METHODOLOGY AND ANALYSIS 25
3.1 INTRODUCTION 25
3.2 ANALYSIS OF THE PROPOSED SYSTEM 25
3.3 FORWARD PROPAGATION 26
3.4 STLF MODELING AND PROCEDURE 27
3.5 NEURON MODEL 28
3.6 NETWORK ARCHITECTURE 29
3.7 GENERAL ANN TRAINING 30
3.8 NEURAL NETWORK TRAINING 31
3.9 USING NEURAL NETWORK MODEL 32
CHAPTER FOUR 41
SIMULATION RESULTS AND DISCUSSION 41
4.1 VALIDATION AND DATA TESTING 41
4.2 SIMULATION INTERFACE OF BORI ELECTRICITY DEMAND PREDICTION WITH MATLAB;………………………………………………41
4.3 RESULT PRESENTATION OF BORI ELECTRICITY DEMAND PREDICTION 46
4.5 GRAPHICAL REPRESENTATION OF HOURLY FORECASTED LOAD………………………………………………………………………58
4.6 RESULT ANALYSIS 64
CHAPTER FIVE 66
CONCLUSION AND RECOMMENDATION 66
5.1 CONCLUSION 66
5.2 RECOMMENDATION 66
REFERENCE 67
APPENDIX I 70
APPENDIX II 74
PROGAM CODE FOR ANN 74
CHAPTER ONE
1.0 INTRODUCTION
Load forecasting has become in recent years one of the major areas of research in electrical engineering. Load forecasting is however a difficult task. First, because the load series is complex and exhibits several levels of seasonality. Second, the load at a given hour is dependent not only on the load at the previous day, but also on the load at the same hour on the previous day and previous week, and because there are many important exogenous variables that must be considered (Hippert et al 2001). Accurate models of power load forecasting are asserted to the operation and planning of a utility company. Load forecasting can be defined as the process of knowing what may happen to system in the next coming time periods. Load forecasts are extremely important in Bori for energy supplies, institutions and other participate in electric energy generation, transmission, distribution and markets.
Load forecasting plays an important role in power system planning and operation. Basic operating functions such as unit commitment, economic dispatch, fuel scheduling and unit maintenance, can be performed efficiently with an accurate forecast ( Alsayegh (2003), Senjyu (2002) and Baklrtzis (1996) ).
The need for accurate load forecasts will increase in the future because of the dramatic changes occurring in the structure of the utility industry due to deregulation and competition. This environment compels the utilities to operate at the highest possible efficiency, which, as indicated above, requires accurate load forecasts.
For decades the problem of improving the accuracy of load forecasts has been an important topic of research. Various types of load forecasting methodologies have their own advantages. Load forecasting can be performed using many techniques such as regression analysis, statistical methods, artificial neural networks, genetic algorithm, fuzzy logic etc.
1.1 BACKGROUND OF THE STUDY
Electrical energy is a variable in which many factors affect its capacity. Load prediction is a complex procedure because of the nature of the influencing factors—weather factors, seasonal factors and social-economic factors. The essence of load forecasting in Bori is to help an electric utility to make important decisions including decisions on purchasing and generating electric power, load switching and infrastructure development. Also, energy suppliers, financial institutions and other “players” in the electric energy generation and distribution markets can benefit from reliable load forecasting.
1.2 STATEMENT OF THE PROBLEM
This research work will assist the Public Power Supply in addressing the issue of power supply and load demand of Bori, in Rivers State through forecasting of hourly load consumption for weeks ahead within 33KV line that serve the area in focus.
1.3 RESEARCH AIM AND OBJECTIVES
This study is aimed at developing a neural network based application for forecasting the Electricity Demand in Bori, Rivers State with ANN using a short-term load forecast.
The specific objectives used to achieve the aim are as follows
- Neural Network with Back Propagation is employed in Learning Algorithm in training data.
- Simulating the supplied load data using a neural network based application to generate the optimal prediction.
1.4 SIGNIFICANCE OF THE PROJECT
There is a common belief of mutually interdependence between electricity demand growth and economic development of the area in consideration due to electricity’s vital role for research and academic development.Effective load prediction is usually the utility company’s delight in which they want to ensure the availability of supply of electricity, as well as providing the means of avoiding over- or under-utilization of generating capacity and making the best possible use of the capacity. Thus the research will help the utility companies to forecast future energy demand of any area since this research proves that Artificial Neural Network is very efficient and gives accurate load prediction. This will lead to the attraction of academicians and utility companies alike to identify reliable and profitable methods of predicting load demand.
1.5 ASSUMPTION OF THE STUDY
Due to the incessant power outage in Nigeria, it is difficult to have power supply of 24 hours a day for complete seven days which make up a week. For this reason, the following assumptions are made:
For every hour or day with power outage, the same power demand for previous week will be considered
The neural network is assume to be the best model for STLF due to many researchers approval
Maximum power demand for the weeks considered is assumed to be uniform for other previous weeks. Before it.
Optimum temperature value is adopted since most of the control rooms in the utility company still operate on analog without keeping records for temperature
1.6 SCOPE OF THE STUDY
This research work only covers the data gotten from PHEDC for Bori in Khana LGA, in Rivers State for the Forecasting of Electricity Demand of that area, in which two weeks data of hourly load demand from the utility company of that region was used in the course of this study. All the information pertaining to fundamental analysis input variable are limited to the data gotten from PHEDC. The Artificial Neural Network model (ANN) approach with Backward Propagation (BP) has been employed in this study using input variables in arriving at a better prediction.
1.7 LIMITATIONS OF THE STUDY
This research work was achieved through the use of Backward Propagation in ANN. Other forecasting method can be employed by other researchers to compare the result with that of ANN.
Complete Material Cost #3,000
Order for Complete Material now