Internatonal Journal of Scentfc Research Engneerng & Technology (IJSRET), ISSN 2278 0882 692 Power loss and Relablty optmzaton n Dstrbuton System wth Network Reconfguraton and Capactor placement A.Manasa 1, D. Rav Kumar 2, N. Krshna Kumar 3 1 (Department of EEE,VNR VJIET, Hyderabad Emal: mansa.ankam@gmal.com) 2, 3 (Department of EEE, VNR VJIET, Hyderabad Emal: ravkumar_d@vnrvjet.n, nkkpsg@gmal.com/krshnakumar_n@ vnrvjet.n) ABSTRACT Network Feeder Reconfguraton s the process of alterng the topologcal structure of dstrbuton feeders by changng the open/closed status of the te and sectonalzng swtches. In ths paper, the Network Reconfguraton problem wll be formulated as mult objectve optmzaton problem wth equalty and nequalty constrants. The proposed soluton s based on optmzaton algorthms known as Network Reconfguraton Algorthms and Cutset Approach. The proposed Algorthms have been mplemented on IEEE 33-Bus Radal Dstrbuton System usng MATLAB/ETAP. The comparson of varous Network Reconfguraton Algorthms s also consdered wth respect to power loss and Relablty Indces. Optmal Placement of Capactor has been consdered for loss reducton. Keywords Radal Dstrbuton System, Dstrbuton Load Flow, Network Reconfguraton, Voltage Stablty Index, Performance Indces, Optmal Capactor Placement. I. INTRODUCTION The networks are reconfgured to reduce system real power losses and accomplsh load balancng to releve network overloads. The Voltage Stablty of the Electrcal Dstrbuton Systems (EDS) can be enhanced f the loads are rescheduled by network reconfguraton [1], whch also allows smoothng the peak demands, mproves the voltage profle and Relablty. Relablty evaluaton can be used to evaluate past performance and predct future performance of the EDS. It also dentfes the problematc components n the system that can mpact Relablty. It can also help to predct the relablty performance of the system after any expanson and quantfy the mpact of addng new components to the system. Relablty evaluaton technques developed by [2] are appled n EDS plannng studes and operaton. A method whch apples an artfcal Bee Colony algorthm (ABC) for determnng the sectonalsng swtch to be operated n order to solve EDS loss mnmzaton problem[3]. A Tabu search algorthm for the Reconfguraton of EDS for mnmzng the real power loss s proposed n [4]. An approach usng Dstrbuton Load Flow soluton and a Network Reconfguraton Algorthm (NRA) s consdered n [5, 6] whch mproves voltage profle, Relablty and Voltage Stablty Index (VSI) besdes mnmsng losses. A method was proposed to operate EDS at ts optmum performance wth Network Reconfguraton n [7]. A method to reduce the Actve Power Loss (APL) and Reactve Power Loss (RPL) and to mprove the Voltage Stablty n EDS wth nstallaton of capactor banks s proposed n [8, 9]. A smple modfcaton to the Bnary Partcle Swarm Optmzaton (BPSO) called selectve partcle swarm optmzaton (SPSO) [10]. The man objectve of ths paper s to conceptualze and realze EDS wth mproved Relablty and Voltage Stablty that wll contrbute to the substantal reducton n the EDS loss by usng Network Reconfguraton. II. PROBLEM FORMULATION It s consdered that the voltage should be wthn the specfed tolerance lmts. Radal Dstrbuton System s only consdered, Feeder reconfguraton s performed by selectng among all possble confguratons, the one that gves mnmum falure ndces and ncurs the smallest power losses, and satsfes a group of constrants. The objectve functon s to mnmze the Actve power losses of dstrbuton system PL and relablty ndces, consderng the followng constrants. 1. de voltage constrant: V V V mn max where V mn and V max are the mnmum and maxmum permssble RMS voltages of node respectvely. 2. Load connectvty: Each load bus should be connected va one path to the feeder. 3. Radal Network structure: loops are allowed n the network. www.jsret.org
Internatonal Journal of Scentfc Research Engneerng & Technology (IJSRET), ISSN 2278 0882 693 4. Power losses and Relablty ndces: 0<P L,Q L PLb, Q Lb ; 0<SAIFI (SAIFI)b ;0<SAIDI (SAIDI)b ; 0<CAIDI (CAIDI)b ;0<ASUI (ASUI)b ; ASUI 1 ASAI U N N 8760 N s number of customers at load pont. (8) III. LOAD FLOW AND NETWORK RECONFIGURATION ALGORITHMS 1. Load Flow Analyss The formaton of Bus Injecton Branch Current (BIBC) and Branch Current Bus Voltage (BCBV) matrces are explaned n [6]. These matrces explore the topologcal structure of dstrbuton system. 2. Relablty Evaluaton Basc Probablty Indces (BPI) to evaluate system relablty are gven by Average falure rate (λ) k ks Average annual outage tme (U) U f/yr (1) krk ks hrs / yr (2) Average outage tme (r) r U hrs (3) Where λ s the system falure rate at th load pont, U s system annual outage duraton at th load pont, λ k, r k are the falure rate and average repar tme of k th dstrbutor segment, S s the set of dstrbutor segments connected n seres up to th load pont. The Customer orentated Performance Indces that are most commonly used are defned n [2] as: SAIDI U N U CAIDI ASAI N f / customer (4) N N hr / yr hr N 8760 U N N 8760 (5) (6) (7) 3. Network Reconfguraton Algorthms The Network Reconfguraton Algorthms (NRA) wth the followng condtons have been consdered. 1. Wth VSI condton, consderng only adjacent sectonalzng swtch wth Maxmum VSI dfference. 2. Wthout VSI condton, consderng only adjacent sectonalzng swtch wth Maxmum VSI dfference. 3. Wth VSI condton, consderng only adjacent sectonalzng swtch wth Mnmum VSI dfference. 4. Wthout VSI condton, consderng only adjacent sectonalzng swtch wth Mnmum VSI dfference. 5. Wth VSI condton, consderng the sectonalzng swtch gves mnmum losses wth Maxmum VSI dfference. 6. Wthout VSI condton, consderng the sectonalzng swtch gves mnmum losses wth Maxmum VSI dfference. 7. Wth VSI condton, consderng the sectonalzng swtch gves mnmum losses wth Mnmum VSI dfference. 8. Wthout VSI condton, consderng the sectonalzng swtch gves mnmum losses wth Mnmum VSI dfference. The Network Reconfguraton Algorthms wth the above condtons have been appled to 33-bus Radal Dstrbuton System and results are analysed and t s found that NRA wth 6 th Condton gves better mprovement n voltage profle, Relablty and reducton n power loss and hence t has been explaned n the next subsecton. 4. Optmal Network Reconfguraton Algorthm(NRA wth 6 th condton) The proposed Network Reconfguraton Algorthm s used to search the better swtchng combnaton that mproves the voltage stablty, voltage profle and Relablty and reduces losses by consderng a te swtch and ts adjacent sectonalsed swtch, one at a tme, and contnues the search process untl there s no further mprovement n voltage stablty and there s no further reducton n power loss and relablty Indces. Before the reconfguraton s accepted, a load flow analyss s needed to make sure the losses, performance ndces are decreased. If the result s negatve, the reconfguraton network must reset to prevous network condton. The algorthmc steps for Network Reconfguraton of Dstrbuton System are gven as follows: www.jsret.org
Internatonal Journal of Scentfc Research Engneerng & Technology (IJSRET), ISSN 2278 0882 694 Step 1: Read Lne data, bus data, Probablty of the dstrbuton system and set the flag equal to zero for all te swtches. Step 2: Run the Dstrbuton Load Flow usng BIBC, BCBV matrx approach and compute voltages, VSI, real and reactve power losses, performance ndces SAIFI, SAIDI,CAIDI and ASUI. Step 3: Check whether all the bus voltages are wthn the specfed tolerance lmts or not. V V V.e. wthn 6% of the rated mn max voltage; 0.94 V 1. 06, If so, go to step 10. Step4: Fnd the VSI dfference between the end nodes k and m of te swtches wth zero flag. Choose the teswtch wth largest VSI dfference. Step 5: Check whether VSI at k th node s larger than VSI at m th node, f so, go to step 7 Step 6: Open the sectonalzng swtch between k and k-1 or k-1 and k-2 or any swtch that gves mnmum losses and performance Indces and go to Step 8. Step 7: Open the sectonalzng swtch between m and m-1or m-1 and m-2 or any swtch that gve mnmum losses and performance Indces (PI). Step 8: Connect the te swtch and set flag equal to 1. Step 9: Calculate power losses f not 0 < P L, Q L P Lb, Q Lb, open te swtch and close sectonalzng swtch and go to step 2. Step 10: Calculate relablty ndces: SAIFI, SAIDI, CAIDI and ASUI If not 0<SAIFI (SAIFI) b ; 0<SAIDI (SAIDI) b ;0 < CAIDI (CAIDI) b ; 0 < ASUI (ASUI) b, open te swtch and close sectonalzng swtch and go to step 2. Step 11: Check the te swtches flag equal to 1 or not, If not go to step 2. Step 12: Prnt V, (VSI), P L, Q L, SAIFI, SAIDI,CAIDI and ASUI. The algorthmc steps are shown as Flowchart n Fgure 1. IV. RESULTS AND ANALYSIS Consder a 33-bus RDS shown n Fg.2. Lne, bus and relablty data of 33-bus RDS are gven n [5]. The NRA s mplemented n MATLAB, appled to 33-bus RDS and analyss s done. The converged values of Voltage magntudes, phase angle and VSI for the Base confguraton shown n Fg. 2 are gven n Table 1. Actve power loss (APL) of the system s 202.6650 kw. The VSI dfference between two sdes of the te swtch n between 22-12 s larger so ths te-swtch s to be closed frst. As VSI of 22 s greater than VSI of 12, the swtch n the branch 12-11 s opened. w the APL s 156.7283 kw whch s smaller than the ntal value so keep on searchng n the drecton. Fnally when the swtch n between 9-10 closed, APL s 154.07 kw whch s mnmum for ths loop wth te swtch n between 22-12 s closed. For ths reconfguraton, next te swtch s to be closed s n between 25-29, and then repeat the procedure andsoluton s to open the swtch n between 28-29 wth APL 145.88 kw. Next te swtch to be closed n between 9-15, and the swtch to open n between 14-15, wth APL 142.56 kw. The procedure s repeated untl the fnal optmal confguraton s acheved. Thus the APL s 140.00 kw after reconfguraton and the fnal optmal confguraton s shown n Fg. 3, the te swtches are n between 9-10, 28-29, 33-32, 14-15 and 7-8. 19 20 21 22 Bus Sectonalzng Swtch Te Swtch Slack Bus Fg. 2 Lne Dagram of 33-bus RDS for Base Confguraton 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 23 24 25 26 27 28 29 30 31 32 33 www.jsret.org
Internatonal Journal of Scentfc Research Engneerng & Technology (IJSRET), ISSN 2278 0882 695 Start Read Lne, Bus and Probablty data Set Vmn = 0.94, Vmax = 1.06 Run Dstrbuton Load Flow and calculate power losses, relablty ndces Open te swtch l and connect sectonalsng swtch Is V Vj mn V max Is 0 < P L, Q L P Lb, Q Lb Fnd the VSI dfference between the end nodes k and m of te swtch l wth zero flag, choose te swtch l wth the largest VSI dfference Calculate Relablty ndces Is Open the swtch whch gves mnmum losses, PI between m and m-1or m-1 and m- 2 etc. If VSI(k) > VSI(m) Open the swtch whch gves mnmum losses, PI between k and k-1 or k-1 and k-2 etc. 0 <SAIFI (SAIFI) b; 0 <SAIDI (SAIDI) b; 0 < CAIDI (CAIDI) b; 0 < ASUI (ASUI) b set flag=1 and connect te swtch l Calculate Power Losses Is flag=1 for all teswtches Prnt values Stop Fg. 1 Flow chart for Network Reconfguraton of 33-bus RDS www.jsret.org
Internatonal Journal of Scentfc Research Engneerng & Technology (IJSRET), ISSN 2278 0882 696 Bus. Table 1 Converged values of bus voltages magntude, phase angle and VSI Before Reconfguraton Voltage Voltage Phase Angle VSI Phase Bus. Magntude Magntude Angle VSI 1 1 0 1 18 0.9131-0.0086 0.6951 2 0.997 0.0003 0.9882 19 0.9965 0.0001 0.6934 3 0.9829 0.0017 0.9331 20 0.9929-0.0011 0.9720 4 0.9755 0.0028 0.9053 21 0.9922-0.0014 0.9692 5 0.9681 0.004 0.8781 22 0.9916-0.0018 0.9668 6 0.9497 0.0023 0.8127 23 0.9794 0.0011 0.9529 7 0.9462-0.0017 0.8014 24 0.9727-0.0004 0.8950 8 0.9413-0.0011 0.7851 25 0.9694-0.0012 0.8829 9 0.9351-0.0023 0.7644 26 0.9477 0.003 0.8761 10 0.9292-0.0034 0.7456 27 0.9452 0.004 0.7980 11 0.9284-0.0033 0.7429 28 0.9337 0.0055 0.7599 12 0.9269-0.0031 0.7381 29 0.9255 0.0068 0.7336 13 0.9208-0.0047 0.7187 30 0.922 0.0086 0.7225 14 0.9185-0.0061 0.7117 31 0.9178 0.0072 0.7095 15 0.9171-0.0067 0.7074 32 0.9169 0.0068 0.7067 16 0.9157-0.0071 0.7032 33 0.9166 0.0066 0.7058 17 0.9137-0.0085 0.6970 Bus. Table 2 Converged values of bus voltages magntude, phase angle and VSI After Reconfguraton Voltage Voltage Phase Angle VSI Phase Angle Bus. Magntude Magntude VSI 1 1.00000 0.00000 1.00000 18 0.98874-0.00453 0.94004 2 0.99732 0.00027 0.98933 19 0.99583-0.00018 0.94322 3 0.98724 0.00172 0.94971 20 0.98357-0.00351 0.97016 4 0.98544 0.00169 0.94301 21 0.98027-0.00488 0.93333 5 0.98395 0.00160 0.93734 22 0.97724-0.00616 0.92235 6 0.98081 0.00060 0.92538 23 0.97987 0.00229 0.91105 7 0.98017-0.00008 0.92302 24 0.96515 0.00252 0.91995 8 0.97438-0.00723 0.91743 25 0.95385 0.00361 0.94172 9 0.97339-0.00755 0.85607 26 0.98053 0.00057 0.82179 10 0.97290-0.00770 0.88562 27 0.98027 0.00054 0.91378 11 0.97112-0.00750 0.88476 28 0.97974 0.00031 0.91967 12 0.97125-0.00754 0.86744 29 0.94883 0.00416 0.92025 13 0.96869-0.00780 0.88051 30 0.94564 0.00587 0.79032 14 0.96791-0.00808 0.87767 31 0.94223 0.00461 0.79632 15 0.99420-0.00251 0.85023 32 0.94156 0.00429 0.78651 16 0.99251-0.00287 0.91922 33 0.98842-0.00459 0.78547 17 0.98972-0.00435 0.92630 www.jsret.org
Internatonal Journal of Scentfc Research Engneerng & Technology (IJSRET), ISSN 2278 0882 697 19 20 21 22 Bus Sectonalzng Swtch Te Swtch Slack Bus 2 3 Fg. 3 Reconfgured 33-bus RDS 1. Comparson of Voltage Magntudes Comparson of voltage magntudes for base case and reconfgured network of 33-bus RDS s shown n Fg.4. 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 1 23 24 25 26 27 28 29 30 31 32 33 2. Power Loss Analyss The APL, RPL and Total Power Losses (TPL) of the 33-bus RDS, before and after reconfguraton are gven n Table 3 and ts comparson s shown n Fg.5. Table 3 Varaton n the power loss of 33-bus RDS Power loss Before NR After NR %Decrease Actve Power Loss(kW) 202.67 140.00 30.92 Reactve Power Loss (kvar) 135.13 104.09 22.97 Total Power Loss (KVA) 243.59 174.4 28.40 Fg. 5 Comparson of Power loss for Base Case and Reconfgured Network 3. Performance Indces The Performance Indces SAIFI, SAIDI, CAIDI, and ASUI of the 33-bus RDS are evaluated by consderng cutsets of varous load ponts usng cutset approach [5]. The values of performance ndces before and after Network Reconfguraton are gven n Table 4 and ts comparson s shown n Fg. 6. Table 4 Varaton n the Performance Indces of Dstrbuton System Index Before NR After NR %Decrease SAIFI (f/customer) 2.41 2.31 4.14 SAIDI (hrs/ yr) 2.04 1.48 27.4 CAIDI (hr) 0.85 0.64 24.7 ASUI 2.33e -4 1.66e -4 28.75 Fg.4 Comparson of voltage magntudes for Base Case and Reconfgured Network www.jsret.org
Internatonal Journal of Scentfc Research Engneerng & Technology (IJSRET), ISSN 2278 0882 698 Fg. 6 Comparson of Performance Indces for Base Case and Reconfgured Network V. OPTIMAL CAPACITOR PLACEMENT Capactor placement n dstrbuton network s one of the most common methods that used n the dstrbuton network to reduce the power loss and mprovng the voltage profle. ETAP software s used for Optmal Capactor placement. A capactor of 0.3MVAR s placed optmally at bus 3 of 33-bus RDS. The comparson of power loss values for base case and Reconfgured network of 33-bus RDS wth Optmal Capactor placement s gven Table 5. Power loss APL (kw) RPL (kvar) TPL (kva) VI. Table 5 Comparson of Power loss for Base Case and Reconfgured Network wth Optmal Capactor Placement Wthout Capactor Base Case Wth Capactor Reconfgured Network wth NRA 6 Wthout Wth capactor Capactor 202.66 197.9 140.00 136.3 135.13 132.7 104.9 103.0 243.58 238.27 174.9 170.8 COMPARISON OF POWER LOSS WITH EXISTING METHODS The comparson of Power loss wth exstng methods s gven n Table 6. Table 6 Comparson of Power loss wth exstng methods Actve Power Loss Wth Actve Power Loss wth Method Network Network Reconfguraton Reconfguraton and wthout capactor (kw) wth Capactor (kw) ABC method [2] 139.5 - GA method[11] 140.2 - PSO method [10] 139.7 - GSO [9] - 143.76 Proposed method 140 136.3 From Table 6, t can be concluded that wth the proposed algorthm and OCP the losses are further reduced and also t s proved that Relablty s also mproved. VII. CONCLUSIONS A Network Reconfguraton scheme for Voltage Stablty enhancement of Radal Dstrbuton Systems has been developed. Ths algorthm s based on VSI and mproves voltage profle, Relablty Indces, enhance the voltage stablty and reduce the system power losses. The proposed algorthm s appled to 33- bus RDS and the obtaned results from MATLAB are analyzed and compared wth exstng methods. Wth the Optmal Capactor Placement the losses are further reduced. REFERENCES [1] Dr. M.K. Khedkar, Dr. G.M. Dhole, A Text book of Electrc power Dstrbuton Automaton, USP, pp. 78-85 [2] Roy Bllnton, Allan RN, Relablty Evaluaton of Power Systems, Second Edton, Sprnger Internatonal Edton, Reprnted n Inda, BS Publcatons,2007, pp.229-231. [3] R.Srnvasa Rao,S.V.L Narsmham, M.Ramalngaraju Optmzaton of Dstrbuton Network for Loss Reducton usng Artfcal Bee Colony Algorthm, Internatonal Journal of Electrcal, Computer, Energetc, Electronc and Communcaton Engneerng Vol. 2,. 9, 2008 pp. 1964-1970. [4] Marcos A. N. Gumaraes,Carlos A. Castro Reconfguraton of dstrbuton system for Loss www.jsret.org
Internatonal Journal of Scentfc Research Engneerng & Technology (IJSRET), ISSN 2278 0882 699 Reducton usng Tabu search,15th PSCC, Lege, 22-26 August 2005 sesson 10, paper 2, pp. 1-6. [5] D. Rav Kumar and V.sankar, Loss reducton and relablty optmzaton n electrcal dstrbuton systems usng network reconfguraton, The Journal of CPRI, Vol. 11,.2, June 2015, pp. 259-268. [6] Jen-Hao Teng, A Drect Approach for Dstrbuton System Load Flow Solutons, IEEE Trans. PD, Vol. 18,.3, 2003 pp. 882-887. [7] S.A. Heydar, T. Heydarzadeh, Naser M. Tabatabae A Combned approach for loss reducton and voltage profle mprovement n dstrbuton systems, Internatonal journal on Techncal and physcal problems of engneerng, ISS, 26, Vol. 8,. 1, Mar. 2016 pp. 30-35. [8] Sarfaraz Nawaz, Nkta Jhajhara, Tanuj Manglan Reducton of Dstrbuton Losses by combned effect of Feeder Reconfguraton and Optmal Capactor Placement,Internatonal Journal of Recent Research and Revew, Vol. II, June 2012 pp. 30-38. [9] Mr. Y. Mohamed Shuab, Dr. M. Surya Kalavath Optmal Capactor Placement n Radal Dstrbuton System usng Group Search Optmzaton Algorthm, Internatonal Journal of Scentfc& Engneerng Research, Vol. 5, Issue 4, Aprl 2014, pp.1-6. [10] Tamer M. Khall, Alexander V. Gorpnch Reconfguraton for loss reducton of dstrbuton systems usng selectve partcle swarm optmzaton, Internatonal Journal of Multdscplnary Scences and Engneerng, Vol.3,. 6, June 2012, pp. 16-21. [11] Y. Y. Hong and S. Y. Ho, Determnaton of network confguraton consderng mult objectve n dstrbuton systems usng genetc algorthms, IEEE Trans. Power Syst., vol. 20, no. 2, May 2005,pp. 1062 1069. www.jsret.org