Indian Journal of Electronics and Electrical Engineing (IJEEE) Vol.2.No.2 204pp 89-93. available at: www.goniv.com Pap Received :5-04-204 Pap Published:25-04-204 Pap Reviewed by:. John Arht 2. Hendry Goyal Editor : Prof. P.Muthukumar MODELING AND ANALYSIS OF THREE AREA THERMAL POWER SYSTEM USING CONVENTIONAL CONTROLLERS A.Ruby meena S. Senthil Kumar Electrical Engineing, Govnment college of Engineing Salem, India -6360. Email:rubymeena77@gmail.com Email:sengce2009@yahoo.in ABSTRACT This pap describes the automatic load frequency control of intconnected multi area pow system with conventional s. The study has been designed for a intconnected three area thmal system considing reheat turbine thmal system in each area. The comparison between a conventional Proportional Integral(PI), Proportional Divative (PD) and Proportional Integral Divative (PID) with a step load change given in all the three areas shows that the PID Controll can genate best dynamic pformance in tms of peak ov shoot, peak undshoot and settling time. The system simulation is realized by using MATLAB/SIMULINK software. Index Tms Automatic Load Frequency Control, Area Control Error, Frequency deviation, Tie Line Pow deviation, Proportional Integral Divative Controll. I. INTRODUCTION L oad frequency control is a vy important topic in pow system opation and control. The pow system is affected with sudden load change and seve fault conditions. This leads to system frequency deviations and scheduled Tie line pow intchange between the intconnected areas. So Automatic Load Frequency Control (ALFC) is needed in the intconnected pow system design to regulate the system frequency and Tieline pow intchange within the scheduled values. If these values deviate from their limits, they cause unwanted disturbances in the pow system []-[6]. For example, the frequency deviation will affect the pow system opation, security, reliability, efficiency, degrading load pformance, ov loading of transmission lines and trigging of protection devices. The turbine mechanical pow output depends on the steam injection into the turbine blades, which will then be convted to electrical pow by synchronous genator. Thefore, the frequency of current and voltage waveforms at the genator output mainly depends on steam injection to the turbine blades. So the frequency can be varied by varying the steam injection, which involves the adjustment of control valve at the steam flow pipe. Two types of control loops are available in the pow system are the Automatic Load Frequency Control (ALFC) and the Automatic Voltage Regulator (AVR). The ALFC which maintains the system frequency and tie line pow intchange within the limits. ALFC also has two control loops which are primary control and secondary control. Und normal opating conditions, the small frequency deviation can be corrected by primary control which includes the fly ball govnor and speed govning mechanism. The sudden load changes initially managed by the stored the kinetic engy in the flywheel. Then the speed govning mechanism with secondary control is responsible for the fine tuning of frequency deviation [7]-[8]. Conventional PI, PD, PID s can be used as secondary control [9]. By tuning the proportional, Integral and divative gains, the desired dynamic response of the pow system can be achieved with minimum Area Control Error. goniv Publications Page 89
II. MODELING OF THERMAL PLANT AND U = ( K p + ) E(s) CONVENTIONAL CONTROLLERS S (7) A. Modeling of a thmal pow plant The control signal produced in PID is The steam input to the turbine is ed using the speed govnor, when the is an imbalance occurs U = ( Kp+ + KdS) E(s) S between the genation and demand. This imbalance is (8) sensed by the govnor in tms of change in frequency (Δf). Based on the change in frequency, the speed III. AUTOMATIC LOAD FREQUENCY govnor controls the position of the control valve, and CONTROL OF THREE AREA SYSTEM increases or decreases the steam injection to the turbine The pow system with three control area blades. The steam input can also be ed using intconnected by tie line as shown in fig. is refence pow setting (ΔPref) of the govnor. The consided. Each area supplies its us pool and the tie speed govnor output (ΔPg) is given by equation (). line allows electric pow to flow between areas. Thefore, the load distribution in one of the areas Pg = Pr ef f () affects the frequencies of oth areas, as well as the R P ( ) Pr ( ) ( ) pow flows on tie line. Due to this a control system is g S = ef S F S (2) needed in each area to bring the system frequency and R tie line pow to its steady state values. Fig. 2 shows the The govnor is assumed to have a time constant of Simulink diagram of intconnected three area T g then the govnor output equation is system.the area control ror of i th area in a multi area system is given in equation (9). Pv( S) = Pg( S) (3) stg ACE The turbine is assumed to have a time constant of T i = P ij+ K iδ w. (9) t then the turbine output equation is Pt( S) = Pv( S) stt (4) The reheat turbine is consided in this pap, whose output (ΔP T ) is furnished in equation (5). KT r rs PT( S) = P t( S) (5) str B. Modeling of conventional s Fig. Intconnected three area pow system. The Proportional Integral, Proportional Divative and Proportional Integral Divative are used in many control system are called as conventional s. The PD produces the control signal consisting of two tms one is proportinal to the ror signal and the oth is proportinal to the divative of the ror signal. The PD could add damping and but it maintains a constant steady state ror all ov the time. When a PI is added with a system the type and ord numb of the system is increaesd by one. Thus the steady state ror is improved by one ord and it can reduced to zo, if the system input is constant. But at the same time the rise time is decreased. To ovcome these problems many systems are ed with PID which has the best features of PI and PD s. The control signal produced in PD is U = ( Kp+ KdS) E(s) (6) The control signal produced in PI is Fig. 2 Simulink model for three area intconnected thmal pow system. goniv Publications Page 90
To obtain the optimum gain values of PD, PI and PID s Integral Of Squared Error Multiplied With Time (ITSE) paramet optimisation method is used. The pformance index J is to be minimised to get the optimum gain value is given in equation (0). The pformance index curve for PD, PI and PID s are shown in figure 3a,3b and 3c respectively.the optimal gain values is shown in table. 2 2 2 Pformance Index J = ( f + f 2 + f 3 ). t dt PI PD PID TABLE I OPTIMAL GAIN VALUES Pforman ce Index J Proportion al Gain Kp Integr al Gain (0) Divati ve Gain Kd 0.0023 0.263 0.206 -.79 3.5-2.4 0.003 0.284 0.020 8 0.04 Fig. 3c Pformance Index curve for PID V.SIMULATION RESULTS Frequency deviation of three area sytem with PD, PI and PID in area, area2 and area3 following a step load disturbance is shown in Fig 4a, 4b and 4c respectively. Tie line pow deviation of area 2,area23 and area3 using PD, PI and PID is shown in figures 5a,5b and 5c respectively. The peak ovshoot values, undshoot values, steady state ror and settling time values of the frequency deviation curve area, area2 and area3 of intconnected three area system detmines the system stability. The simulation results shows that the PID Controll can genate best dynamic pformance. Fig.3a. Pformance Index curve for PD. Fig.4aFrequency deviation of area with PI,PD,PID with step load disturbance Fig. 3b Pformance Index curve for PI. Fig. 4bFrequency deviation of area2 with PI,PD,PID with step load disturbance goniv Publications Page 9
Fig. 4cFrequency deviation of area3 with PI,PD,PID with step load disturbance Fig. 5c Tie line pow deviation of three area system using PID with step load Fig. 5a Tie line pow deviation of three area system using PD with step load Fig. 5b Tie line pow deviation of three area system using PI with step load VI. CONCLUSION In this study, Automatic Load Frequency Control of three area intconnected pow system with reheat turbine in each area is employed. The pformance of PD, PI and PID is shown in the simulation results in tms of Peak ovshoot, Peak undshoot and settling time. From the results, it is obsved that the PID has less settling time and less steady state for frequency deviation as compared to conventional PD and PI. Even though the PD has less peak ovshoot and peak undshoot the steady state ror is not minimized. He the genator rate constraints and govnor non linearity s are not taken for simplicity. Appendix System Paramets: T g =0.2sec, T t =0.3sec, K ps =20, T ps = 20 sec, K r =0.5, T r =0sec, R =5Hz/puMW, B =0.2083puMW/Hz. T g2 =0.sec, T t2 =0.4sec, K ps2 =00, T ps2 = 8 sec, K r2 = 0.5, T r2 =0sec, R 2 =2.4Hz/puMW, B 2 =0.425 pumw/hz. T g3 =0.sec, T t3 =0.5sec, K ps3 =20, T ps3 = 20 sec, K r3 = 0.5, T r3 =0sec, R 3 =4 Hz/puMW, B 3 =0.2583 pumw/hz. T g :Govnor time constant, T t : Turbine time constant, K ps : Pow system Gain, T ps : Pow system gain constant,k r : Reheat gain,t r :Reheat time constant,r: Regulation paamet, B: Frequency bias factor. REFERENCES [].A.J.Wood, B.F.Woolenbg, Pow Genation Opation and Control, John Wiely and Son s 984. [2].O.I.Elgd, Electric Engy System Theory-An Introduction, Mc Graw Hill Co.200. goniv Publications Page 92
[3].Hadi Saadat Pow System Analysis, Tata Mc Graw Hill 200. [4].M.Gopal Modn Control Theory, Wiely Eastn Ltd, 2 nd edition 993. [5].Jaleeli, N. VanSlyck, L.S. Ewart, D.N. Fink, L.H. Hoffmann, A.G. Undstanding automatic genation control IEEE Transactions on Pow Systemss,vol 7, No.3, Aug. 992. [6]. Vaibhav Donde, M.A.Pai,Ian.Hiskens Simulaton and optimization in an AGC system aft Degulation IEEE Transactions on Pow Systemss,vol 6, No.3, Aug. 200. [7].M.F.Hossian,T.Takahashi,M.G.Rabbani,M.R.I.Sheik h, M.S.Anow Fuzzy Proportional Integral Controll for an AGC in a single area Pow System. ICECE 2006. [8]. G.A.Chown, R.C.Hartman Design and expience with a Fuzzy Logic Controll for Automatic Genation Control IEEE Transactions on Pow Systemss,vol 3, No.3, Aug. 997. [9] Nanda, J., Mangala, A., Suri, S. Some new findings on Automatic genation control of an intconnected hydro thmal system with conventional s, IEEE Transactions on Engy Convsion,Vol. 2,No.pp.87-94, 2006. goniv Publications Page 93