I AN IMPROVED DC LOAD FLOW TECHNIQUE FOR RELIABILITY STUDIES OF POWER SYSTEMS A dissertation submitted to THE UNIVERSITY OF MORATUWA for 1>i I ' % H i /H 1 THE DEGREE OF MASTER IN ENGINEERING by. P E R M A N E N T R t i ! • L K L AG E M f TC cfc R E M . - V t . u HsCM TH£- L I B R A R Y . ALEXANDER S. MALEMBEKA ELECTRICAL ENGINEERING DEPARTMENT UNIVERSITY OF MORATUWA SRI LANKA JUNE,1983 3952V 7 ^ - 7 (i) CONTENTS Page ACKNOWLEDGEMENT (ii) 1.0 INTRODUCTION , 1 2.0 BRIEF BACKGROUND AND HISTORICAL DEVELOPMENT OF POWER SYSTEM RELIABILITY STUDIES 5 3.0 METHODS FOR QUANTITATIVE EVALUATION OF RELIABILITY OF POWER SYSTEMS 8 3.1 AVERAGE INTERRUPTION RATE 8 3.2 FREQUENCY AND DURATION 9 3.3 MARKOV PROCESS 16 3.4 APPROXIMATE 19 3.5 CONTINGENCY ANALYSIS 2.1 3.6 COMPOSITE SYSTEM RELIABILITY .22 4.0 TECHNIQUES FOR LOAD FLOW ANALYSIS 29 4.1 AC FAST DECOUPLED LOAD FLOW 31 4.2 DC LOAD FLOW FOR USE WITH A NETWORK ANALYSER 35 4.3 IMPROVED DC LOAD FLOW FOR USE WITH A DIGITAL COMPUTER 40 4.4 STOCHASTIC LOAD FLOW 45 5.0 CALCULATION OF RELIABILITY INDICES 61 5.1 PROBABILITY OF FAILURE OF CONSUMER LOAD POINTS 61 5.2 SENSITIVITY FACTORS 61 5.3 EFFECT OF. GENERATOR OUTAGE ON THE FAILURE OF CONSUMER POINTS 65 5.4 OVERALL SYSTEM PERFORMANCE 65 5.5 SERVICE FAILURE PROBABILITY 65 5.6 EXPECTED UNSERVED ENERGY 67 6.0 R E S U L T S AND C O N C L U S I O N S 6.1 A P P L I C A T I O N OF THE METHOD AND R E S U L T S 6.2 C O N C L U S I O N S A P P E N D I X 1 A P P E N D I X 2 A P P E N D I X 3 R E F E R E N C E S (ii) ACKNOWLEDGEMENT The author is highly indebted to Dr. O.P. Kulshreshtha, now the UNESCO Chief Technical Advisor to the Open University of Sri Lanka who in 1975 made arrangements for the award of the UNESCO Fellowship while he was working in the United Republic of Tanzania. He wishes to express his gratitude to Professor S. Karunaratne of the University of Moratuwa for providing the facilities in the Electrical Engineering Department and for his tireless assistance in problems relating to Electrical Engineering in general. The author is also highly indebted to his Supervisor, Dr. A.S. Induruwa for his invaluable assistance and encouragement without which this work would have not been realised. He wishes to thank Dr. J.R. Lucas of the University of Moratuwa for his help in computer programming and Drs. H. Sriyananda and M.P. Dias for reading the manuscript and for their suggestions in presenting the material in this dissertation. Much thanks go to the staff members of the Computer Centre at the University of Moratuwa, Open University and University of Colombo for the good service provided by them. The author takes this opportunity to extend his sincere appreciation to the United Nations Educational, Scientific and Cultural organization for granting him the Fellowship and to the United Nations Development Programme in Sri Lanka for administering the fellowship. Finally he wishes to thank his employer, the Principal Secretary of the Ministry of National Education of the United Republic of Tanzania for granting him study leave to undertake this course of study. (iii) LIST OF .SYMBOLS » P - deterministic vectors which distribute the system load to individual buses. H - susceptance of series element of transmission line (i,k) D - number of days of study d. - mean duration of a contingency state e - mean duration of system load being in a peak state * = V e J 9 i • i 1 1 - complex voltage at node I E* - complex conjugate of voltage E^ erf - the error function .. ex]j - exponent *i current through line (i,k) at node i i,k - node identification j - complex operator -• Lcj - critical load of a contingency network state L system load being in a peak state L 0 - system load being in a low state ^ - failure rate of a component - repair rate of a component e ik - voltage angle difference between nodes i and k, (0 - 0 ) i k 0 1 - voltage angle at node i (reference to slack node) (iv) U - duration of normal weather N_. - contingency state of a transmission network W - normal state of a transmission network P - real power flow in line (i,k) at node i Pr - probability Plr"' - the probability of overload of line mn due to line pq mn V i being out of service f} - reactive power flow in line (i,k) at node i 0 - modified reactive power in line (i.k) m q - reactive power generated in line (i,k) P - random variable for system load r ~ series resistance of line (i,k) R - sensitivity factor J'i. - complex power flow in line (i,k) at node i, (S. = P. - jo ) 1 1 1 1 voltage magnitude a^ . node i x - series reactance of line (i,k) i_ K. x - inverse of the series susceptance of line (i,k), (x = 1) sh y^ k - shunt susceptance of line (i,ki z . . - series impedance of line (i,k) , z. = (r + jx ) ik ik ik ik