Effect of Wind Energy Unit Availability on Power System Adequacy

Transcription

1 , Vol 9(28), DOI: /ijst/2016/v9i28/97962, July 2016 ISSN (Print) : ISSN (Online) : Effect of Wind Energy Unit Availability on Power System Adequacy Athraa Ali Kadhem 1*, N. I. A. Wahab 1, Ishak Bin Aris 1, Jasronita bt Jasni 1 and Ahmed N. Abdalla 2 1 Center for Advanced Power and Energy Research, Faculty of Engineering, University Putra Malaysia, Selangor 43400, Malaysia; athraaonoz2007@yahoo.com 2 Faculty of Engineering Technology, University Malaysia Pahang, Kuantan 26300, Malaysia Abstract Wind power has remarkable economic and environmental advantages when compared to other power generation sources. Presently, wind power is considered to be an essential alternative source for generating power. The growing pervasiveness of Wind Energy Conversion System (WECS) in power systems has a huge influence on the electrical system s reliability in relation to other conventional sources for power generation. If the variation in the speed of the wind of a specific site is significant, then, the power output from the wind turbine may get severely affected. The output power from the WECS may also get affected by the unavailability of wind generating units for a considerable period of time. The impact of wind turbine units on the reliability of power generating systems and operating reserves is explained in this paper, while considering wind power units that are frequently unavailable and for a considerable amount of time. The paper presents the impacts of the duration and frequency of failures of the Wind Turbine Generators (WTGs) on the WECS output power. A Sequential Monte Carlo Simulation (SMCS) technique along with Frequency and Duration method shows it is effective for estimating the WECS output power, and the simulation was conducted on IEEE RTS-79 bus system. Keywords: Generating Capacity Adequacy, Sequential Monte Carlo Simulation, Wind Power, Wind Energy Conversion System 1. Introduction In order to achieve power system requirements, renewable energy sources such as solar, wind 1, wave power, tidal stream, and biomass are being used frequently, as seen evident in nations such Denmark, Spain, Germany, China, and USA 2,3, where the WECS is widely utilized. The WECS system constitutes of two major models, namely; the Speed model and the WTG model 4. The speed of the wind is the most essential factor that requires consideration when designing a WECS. Moreover, WECS is characteristically less reliable than a conventional power generating unit and, hence its role is of great significance to ascertain how reliable the electric energy supply will be to the power systems 5. The capability of the power generating systems necessary to meet the load demands 6,7, is determined principally by analysing the sufficiency of the energy source and the reliability of the systems. The power generating facilities typically consist of conventional and unconventional units. The conventional units include thermal, hydro, and nuclear units, and also include the scope for collaborative benefits from the different units in an interconnected environment; unconventional units offer options such as WTGs and PV cells to the system designer 8,9. Inconsistent wind speed and unavailability of wind turbine units are the primary factors that impact the WECS output power. The WECS contain one or many of the WTGs, therefore, the unavailability of the WTG in the WECS is not be considering in some practical situation. The unavailability of the WTG can express in terms of unit failure rate and repair rate. This paper evaluates the reliability of power generation systems using wind power 10,11. The effects of the failure rate of WTG and Mean Time to repair (MTTR) was analyzed by applying the reliability technique on the system s adequacy. The objective of this study is to understand the * Author for correspondence

2 Effect of Wind Energy Unit Availability on Power System Adequacy effects of the WTG availability indices on the reliability of the system, and also to explore the advantages of increasing the reliability of the renewable energy units. The application of SMCS technique in conjunction with Frequency and Duration procedure was carried out, when modelling the wind power output and its reliability on the power generating systems. The Weibull distribution model is applied to replicate the wind speed per hour for the simulation process. The attainable power generated from the wind was computed, and the reliability index of the WTG suggested technique indicates the efficiency of estimating the WECS power output. 2. Generating System Adequacy Model In the quest to evaluate the adequacy of the generating systems, an elementary model consisting of a wind power generator model, a conventional generator model and a convolution model were employed as shown in Figure 1, these models incorporate a suitable load model to produce the risk model. To represents the system risk, one or more quantitative risk indices were used. The fundamental reliability indices that are evaluated in this paper to estimate the reliability level of the power generating systems with the inclusion of WECS are Loss of Load Expectation (LOLE), Loss of Load Frequency (LOLF), Loss of Energy Expectation (LOEE), and Loss of Load Duration (LOLD). These indices can be calculated by using the SMCS technique combined with the Frequency and Duration procedure. model or a multiple state model 12 as shown in Figure 2. In the Markovian 2-state model, the generating units are regarded as either completely out of service (down) or totally in service (up). The application of SMCS technique along with the Frequency and Duration procedure 13, the output power of the conventional power generating unit was simulated. The simulation process is defined by the operating cycle of each generating unit, and this is expressed in terms of a its failure rate, λ, implying how many times the unit is out of service per annum, and the MTTR, implying the period for which it is out of service per year. By combining two parameters (λ, MTTR) for a certain duration (usually one year), the operation cycle of every system unit can be estimated. From the simulation of all the components of the system, the available capabilities of the components can be obtained and these capacities can be grouped to estimate the total capacity of the system. Figure 2. Markovian model for a generating unit Load Model The two main modes by which the Load model is generally represented are chronological and non-chronological. Both of these modes can be used in conjunction with the SMCS technique, where either the Load Duration Curve (LDC) will generate values in each hour giving (8760) separate values for a particular year, or the Daily Peak Load Variation Curve (DPLVC) which will generate (365) values for that particular year 12. Figure 3, illustrates the annual hourly load model for IEEE RTS-79 bus system. Figure 1. The risk model in generating system adequacy. 2.1 Model Conventional Units The generating unit was represented separately 2-state Figure 3. Hourly peak load model for IEEE RTS-79 per. 2

3 Athraa Ali Kadhem, N. I. A. Wahab, Ishak Bin Aris, Jasronita bt Jasni and Ahmed N. Abdalla 2.3 Modeling WECS Wind Speed Model Various models have been utilised to simulate the WECS power system for forecasting the wind speeds and evaluating the reliability of the output power Since the Weibull distribution has a property that can modify parameters, such as the shape k and the scale c, it is generally used in simulating the variation in the speed of the wind. The objective of the wind speed sampling per hour is for the estimation of the hourly output power of the WECS that is available. Hence, the Weibull distribution model can be employed in simulating the speed of the wind at any point in time during the simulation period The wind speed (v) can be reproduced artificially 13, by applying the function in Equation (1). The simulated wind speed for (300) hours is depicted in Figure 4. By applying the speed of the wind into the output model of the wind turbine, the Equation 3 of the power output of the wind turbine can be derived as shown in Equation (1): (1) where c and k are Weibull parameters, U is uniform distribution of random variables between [0, 1], and v is speed of the wind. the relationship between the WTG output power and the wind speed 3. Figure 5 shows a graphical representation of a typical WTG power curve. (2) Figure 5. Typical WTG power curve 20. From Figure 5, the WTG does not produce reasonable amount power when the wind speed ws (m/s) is lesser than the cut-in speed V ci (m/s) and invariably shuts down power production when the wind speed is greater than the cut-out speed V co (m/s). When the output power (Pr) of the WTG increases as the speed of the wind increases within the range of V ci and the rated speed of the wind V r and the V r (m/s) remains fixed and WTG generates a capacity of power to which it is rated. A, B and Cx are presented in 21. Figure 6 simulation of the output power of wind turbine for (300) hours is depicted. Figure 4. Simulated wind speed sample for (300) hours The Power Model of a Wind Turbine The wind turbine power can be considered effective economically because the WTG units can provide the energy with zero fuel cost 20. The relationship between the output power of the WTG at any hour of time and that of the wind speed is not constant. Equation (2) represents Figure 6. Simulation depicted the wind turbine output power for (300) hours. 3

4 Effect of Wind Energy Unit Availability on Power System Adequacy Simulation proposed of WTG The wind turbine units may be down at any moment due to forced outages. It is considered that the WECS constitutes of (n) number of identical units of WTG. The available power output at an hour, t from the WECS is equivalent to n p t. The output power is not considered if the WTG unit is out of service. Due to the random failures accompanying the WTG unit, the procedure for modelling the wind power will be as follows: the system will be simulated with the aid of the SMCS technique combined with the failure rate (λ = 1/MTTF) per year, and MTTR per year during the simulation period 13. For each unit of the WTG system, a random value (U) between the interval [0 and 1] in an hour t is created, by using the sample simulate values of MTTR and failure rate (λ) per year of the unit to attain the operation cycles (up-down-up). The unavailability of WTG is taken into account 22 for finding the hourly power output of the WECS, the total power output from the WECS is calculated by multiplying the total number of available WTG units with p t. 3. Case Study paper is implemented for an IEEE-Reliability Test System, which contains the WECS. The IEEE-RTS consists of 32 generation units, ranging from 12 MW to 400 MW unit capacities. The total power output is 3405 MW and the system s has a peak load is 2850 MW. Hourly wind data can be obtained from reference 8. The Weibull distribution model is used in simulating the hourly wind speed. To calculate the available wind power, the speed of the wind and the availability of wind turbine units are considered. The WTGs that are installed in the WECS have the following specifications: V ci = 6.3 m/s, V r = 12 m/s, and V co = 21 m/s, and the rated output power of every WTG unit is given as P r = 2.5MW 8. Figure 7 shows the simulated output power with associated failure of individual units in the sampling year and on installing 50 WTGs of 2.5 MW, while the total installed capacity is 250 MW. In addition to this, the number of outages is depicted in Figure 8, which represents the Frequency and Duration for (60) selected sampling years. Figure 9 represent the available capacity of the system obtained from generating unit and WECS. In figure 10 the available capacity for the power system from the simulate process which superimposition with the chronologically load model is shown. The reliability simulation technique suggested in this Figure 7. Simulated output power with associated failure for sampling year. 4

5 Athraa Ali Kadhem, N. I. A. Wahab, Ishak Bin Aris, Jasronita bt Jasni and Ahmed N. Abdalla Figure 8. Simulated of outages output power with associated failure. Figure 9. The available capacity of the system obtained from generating unit and WECS. Figure 10. The available capacity of the system which, superimposition with chronological available load model. 5

6 Effect of Wind Energy Unit Availability on Power System Adequacy Table 1 presents the reliability indices of 50 WTG units (250 MW) before and after they are added to the IEEE- RTS-79. The simulation will be terminated after a set number of samples (60 times). The failure rate for each WTG (4 f/yr) and the MTTR (90 hours) in this analysis. The results are compared with 6, as can be seen from the columns (1), (2) that have the same degree of accuracy LOLE, LOLF, and LOLD when to apply SMCS technique along with Frequency and Duration method in simulated. Also, it can be seen from Table 1, the estimated reliability indices values of the IEEE-RTS-79 are affected by number failures of every WTG within the simulation period of the WECS as shown in columns (3, 4). 4. Conclusion Table 1. Shows the estimation of adequacy indices of the IEEE-RTS-79 Reliability index Before adding WECS [6] Before adding WECS 50 WTG added without failure 50 WTG added with associated failure LOLE (h/yr) LOEE (MWh/) LOLF (occ/yr) LOLD (hrs/occ) ENSINT (MW/int) Simulated years Each wind unit has λ=4 f/year and µ=(1/90) 24. This study presented the SMCS technique along with the Frequency and Duration manner and wind power modelling for the adequacy assessment of power generation systems. The SMCS technique combined with Frequency and Duration procedure generates artificial operation of WTG. The Weibull distribution model is employed to imitate the hourly wind speed. The attainable wind power is computed by considering availability of wind turbine units and wind speed. The WECS has an important contribution to the reliability performance adequacy of the generating systems. There are several factors that influence the WECS, such as wind direction and speed. This paper researches the effects of the unavailability of WTG on the WECS output power. During the analysis of failure per year of every WTG within the simulated WECS, it was found that the unavailability of turbine units have a significant impact. This paper will help system designers to quantitatively evaluate the worth of the WECS, which will assist in decision making process while designing WTG. 5. References 1. Khare V, Nema S, Baredar P. Solar-wind hybrid renewable energy system: A review. Renew Sustain Energy Rev. 2016; 58: Chauhan A, Saini RP. Statistical analysis of wind speed data using Weibull distribution parameters st Int Conf Non Conv Energy (ICONCE 2014); p Shi S, Lo KL. An overview of wind energy development and associated power system reliability evaluation methods th Int Univ Power Eng Conf p Billinton R, Bai G. Generating capacity adequacy associated with wind energy. Energy. 2004; 19(3): Mosadeghy M, Yan R, Saha TK. Impact of PV penetration level on the capacity value of South Australian wind farms. Renew Energy. 2016; 85: Chaiamarit K, Nuchprayoon S. Modeling of renewable energy resources for generation reliability evaluation. Renew Sustain Energy Rev. 2013; 26: Almutairi A, Ahmed MH, Salama MMA. Probabilistic generating capacity adequacy evaluation: Research roadmap. Electr Power Syst Res. 2015; 129: Billinton R, Gan L. Wind power modeling and application in generating adequacy assessment. IEEE WESCANEX 93 Commun Comput Power Mod Environ - Conf Proc. p Billinton R, Chen H, Ghajar R. A sequential simulation technique for adequacy evaluation of generating systems including wind energy. IEEE Trans Energy Convers. 1996; 11(4): Karki R, Hu P, R. Billinton, Fellow L. Relibility evaluation considering wind and hydro power coordination. Power. 2010; 25(2): Zheng R, Zhong J. Generation adequacy assessment for power systems with wind turbine and energy storage Innov Smart Grid Technol; p Billington R, Allan RN. Reliability evaluation of power systems. New York and London: Plenum Press; Qual Reliab Eng Int. 1996; 1: Shi S, Lo KL. Reliability assessment of power system considering the impact of wind energy th International Universities Power Engineering Conference (UPEC); p

7 Athraa Ali Kadhem, N. I. A. Wahab, Ishak Bin Aris, Jasronita bt Jasni and Ahmed N. Abdalla 14. Billinton R. Incorporating wind power in generating capacity reliability evaluation using different models. IEEE Trans Power Syst. 2012; 26(4): Karki R, Hu P, Billinton R. A simplified wind power generation model for reliability evaluation. IEEE Energy Conversion. 2006; 21(2): Billinton R, Karki R, Gao Y, Huang D, Hu P, Wangdee W. Adequacy assessment considerations in wind integrated power systems. IEEE Trans Power Syst. 2012; 27(4): Odo FC, Offiah SU, Ugwuoke PE. Weibull distribution-based model for prediction of wind potential in Enugu, Nigeria. Adv Appl Sci Res. 2012; 3(2): Azad AK, Rasul MG, Alam MM, Uddin SMA, Mondal SK. Analysis of wind energy conversion system using weibull distribution. Procedia Eng. 2014; 90: Adaramola MS, Oyewola OM, Ohunakin OS, Akinnawonu OO. Performance evaluation of wind turbines for energy generation in Niger Delta, Nigeria. Sustain. Energy Technol Assessments. 2014; 6: Phoon HY. Generation system reliability evaluations with intermittent renewables Sep. p Giorsetto P, Utsurogi K. Development of a new procedure for reliability modeling of wind turbine generators. IEEE Trans Power Appar Syst. 1983; PAS-102(1): Mabel MC, Raj RE, Fernandez E. Adequacy evaluation of wind power generation systems. Energy. 2010; 35(12): Subcommittee PM. IEEE Reliability Test System. IEEE Trans Power Appar Syst. 1979; PAS-98(6): Karki R, Billinton R, Verma AK. Reliable and sustainable electric power and energy systems management;