PROVISION OF DIFFERENTIATED VOLTAGE SAG PERFORMANCE USING FACTS DEVICES

Transcription

1 rd International Conference on Electricity Distribution Lyon, - June Paper PROVISIO OF DIFFERETIATED VOLTAGE SAG PERFORMACE USIG FACTS DEVICES Huilian LIAO Sami ABDELRAHMA Jovica V. MILAOVIĆ University of Manchester U.K. University of Manchester U.K. University of Manchester U.K. huilian.liao@manchester.ac.uk sami.abdelrahman@postgrad.man.ac.uk milanovic@manchester.ac.uk ABSTRACT This paper presents the concept of provision of differentiated levels of voltage sag performance based on zonal customer requirements and proposes a mitigation strategy to fulfil this concept using FACTS devices. In the study, a system index of voltage sag performance, sag gap index, is proposed to present the satisfactory degree of the actual sag performance in accordance with the zonal customer requirements. Using the sag gap index as objective function, greedy algorithm is adopted to search the optimal mitigation solutions among initial device placements which are selected based on voltage sag performance and sensitivity analysis. In the study, a - bus generic distribution network is used as the test system. Using the proposed mitigation strategy, a reasonable small number of FACTS devices are required to mitigate voltage sag phenomena in order to meet the pre-set zonal sag performance thresholds, in comparison to the case of using system index which is generated based on a methodology that is widely used in literature to present network performance of voltage sag. ITRODUCTIO Voltage sags, as one of the most critical power quality problems, have been attracting significant attention from many utilities and industries, due to frequent disruptions to industrial processes and malfunction of electronic equipment, ultimately resulting in substantial financial losses []. It is very important to provide acceptable voltage sag performance required by customers. In reality, requirements about voltage sag performance vary for different customers in different areas, depending on the sensitivity of customer processes and equipment to voltage sags. For this case, offering a differentiated level of voltage sag performance is promising, especially in the age of digital economy and smart grids. This would improve the efficiency of electricity/energy distribution by only ensuring the voltage sag performance as required. In other words, the mitigation planning will target those customers who require higher sag performance than what they have already received. In this way, less mitigation effort is required, which subsequently reduces the cost of investment and improves the efficiency, compared to the case when the sag performance is improved over the whole network and all customers benefit from better sag performance even though they do not need it. Furthermore, provision of differentiated voltage sag performance helps utilities to price the electricity and plan mitigation solution based on customers willingness to pay in different areas, which provides a fair way to subsidize the mitigation activity. However, provisioning differentiated levels of voltage sag performance to various customers has not been properly addressed in literature. In order to ensure provision of required voltage sag performance, it is important to have a proper index which can evaluate sag performance accurately and represent the disruptive impacts of voltage sags to equipment/system operation. A number of single-site indices have been proposed in literature to assess voltage sag performance at one location/site. Sag severity at certain location can be presented through probability density and distribution functions [], sag tables [], or SAIFI-based index [4]. An alternative approach is the use of single-index methods, which calculate the sum or average of the single-event characteristics of all events occurring at the location [,, ]. To reflect the potential impact of voltages sags on equipment/system operation, it is essential to take into account important factors, e.g., equipment sensitivity to voltage sags and the uncertainty in sag severity assessment. The severity of voltage sags is strongly related to the response of equipment to voltage sags, and furthermore the response of equipment/process to voltage sags is influenced by a number of uncertain factors [, ]. To address these factors, a single numerical index, Bus Performance Index (BPI), was proposed recently to evaluate the sag severity []. Based on single-site indices, a system index, which is usually calculated by taking average or weighted average of the obtained single-site indices from sites within the system, is used to represent the voltage sag performance of the whole network. The system indices generated in this way are typically used to select optimal mitigation solution []. This system level index is acceptable if the mitigation target is to improve sag performance of the whole network simultaneously. However, it is not appropriate in the case of provisioning differentiated levels of voltage sag performance. Therefore, a new system index which can represent local requirements in terms of sag performance is required. To reduce the disruptive impact of voltage sags to system/network operation, a number of mitigation techniques have been fully explored over the last decade. Flexible ac transmission system (FACTS) devices have CIRED /

2 rd International Conference on Electricity Distribution Lyon, - June Paper attracted great attention, as they can control network parameters including current, voltage and impedance flexibly. FACTS devices have been reasonably and widely investigated in power systems for various purposes, e.g. restoring bus voltages locally or globally [], enhancing transfer capability [] and maximizing power system loadability [], etc. Apart from that, their application has been widely investigated in power systems for mitigating voltage sag issues [-]. Even though these devices are expensive, placing FACTS devices for power quality mitigation is beneficial in the long run, as the financial benefits resulting from their installation will cover the initial capital investment within a few years after installation [, ]. In this paper, a mitigation strategy is proposed to provision differentiated levels of voltage sag performance required in different zones, based on FACTS devices. A new system index, i.e., sag gap index (SGI), is proposed to present how well the received sag performance meets the customer requirements. Before applying optimization algorithm, location of mitigation devices, including SVC, STATCOM and DVR, are pre-selected based on sag performance and the sensitivity of voltage variation at a bus due to the injection of active or reactive power. A greedy algorithm is used to find the optimal mitigation solution and minimize the sag gap index. The proposed mitigation strategy is tested on a -bus generic distribution network, and the results are compared with those obtained based on a system level sag index..heatmaps are used to present the effect of applying the optimal mitigation solution obtained by the proposed mitigation strategy. METHODOLOGY The System sag index and zonal sag index In this study, Bus Performance Index (BPI) which evaluates the level of voltage sag performance from the perspective of utilities and customers in distribution network is used. The index takes into account various sag characteristics simultaneously as well as sensitivity of equipment to voltage sags. It accounts for sag magnitude, sag duration, sag occurrence frequency, the sensitivity of equipment to voltage sags, the uncertainty of voltage tolerance curve, the stochastic nature of load variation and the uncertainty of sag characteristics. It reflects to a good approximation the practical consequence of voltage sags from the point of view of system/equipment operation. With this index, voltage sag performance across the network can be assessed, and customer requirements can be evaluated. Considering that the customers in the same area typically have similar economic activities and consequently similar requirements regarding voltage sag performance, the network can be divided into zones based on customer requirements []. The sag performance threshold of each zone can be determined accordingly and set and adjusted individually by users. The system sag index, which represents the sag performance of the whole network, is calculated by taking the sum or averages of all site indices within the network. This approach is widely used for the purpose of mitigation planning []. Using this methodology, the system sag index which is based on BPI can be calculated as: B i SI BPI = i= ( j= BPI i,j ) () which takes the sum of BPIs of all buses. To implement the concept of provision of differentiated levels of voltage sag performance, a new index, sag gap index (SGI), is proposed to indicate the level of the received voltage sag performance compared to imposed customer requirements. Different from general system indices, SGI accounts for zonal requirements of customers, as defined below: B i SGI = i= ( j= BPI i,j BPI TH,i BPIi,j ) () >BPI TH,i where B j denotes the total number of buses within power quality zone i; and BPI i,j denotes BPI of the j th bus in zone i. With this index, the areas where customers are most affected by voltage sags can be identified for the purpose of efficient mitigation. Problem definition and optimization procedure Given sag thresholds, the issue of offering differentiated levels of sag performance can be considered as an optimisation problem which is to minimise the gap between the zonal sag thresholds and actual sag performance, i.e., SGI. In order to optimally place FACTS devices, potential and effective locations are made initially available for placement based on sag performance, sensitivity analysis and geographical constraints. Firstly, buses are sorted according to BPI, V j j= and V j Q j= in descending P order, individually. The ranking index of bus i with respect to BPI is denoted as R BPI (B i ), and the same applies to V j j= and V j Q j=. The R P BPI (B i )= suggests that bus i is experiencing the worst sag performance, and R V/ Q (B i ) = that voltage of bus i is the most sensitive to the injection of reactive power. With these rankings, the potential locations are chosen globally (i.e., based on the whole network) and zonally (i.e., based on zonal information) respectively. Global selection: the buses having R BPI = and the smallest R BPI + V j j= are selected as potential Q locations for installing SVC; the buses having R BPI = and the smallest R BPI + R V/ P +R V/ Q j=, are selected as potential locations for installing STATCOM and DVR. Following this appropriate FACTS devices are preliminarily paced at selected potential locations. The same selection procedure is then performed again to CIRED /

3 rd International Conference on Electricity Distribution Lyon, - June Paper select the second set of potential placements. Zonal selection: In the zonal selection, the procedure is the same as the global selection, except that the ranking procedure is performed within the zones rather than within the whole network. With these initially pre-selected locations, Greedy algorithm is used to search the potential solutions (i.e., optimal placement of FACTS devices and their optimal rating settings) to minimise the gap between the threshold and BPI achieved after the application of mitigation solutions. Before applying greedy algorithm, a pool of potential solutions, denoted as set U, should be determined based on the initial placement/locations and rating constraints. As mentioned above, the initial placements of FACTS devices including their location and type have been decided. Assume there are M D potential devices. For each potential device, an extra variable needs to be determined, i.e. the rating of the device. The rating range of each device can be divided into M I intervals, and for each interval, a rating is chosen by randomly selecting a value within the interval. Thus, a pool of M D M I potential solutions, denoted as set U, which consists of locations, types of devices and ratings, are made available initially for optimization process. With the initial set U, the greedy algorithm is applied to select the optimal mitigation solution. The flowchart of the application of greedy algorithm for this allocation problem is given in Fig., where s is the chosen solution which is corresponding to the minimum SGI at each stage; Γ denotes the devices selected so far; and X is the updated pool of potential solutions at each stage. At each stage, X is updated by removing its elements which have the same location and type of devise as the selected s. The optimization procedure can be terminated if the size of Γ reaches the preset maximum number of devices allowed to be installed, or if the improvement of SGI between two sequential stages is smaller than a preset threshold. Set Γ is selected as the final optimal mitigation solutions. X=U; Begin Input set U of potential devices End Γ=Φ; Install covered devices Γ; Update X by regenerate rating randomly within corresponding interval Select sϵx that minimizes objective function; X=X-{all elements in X which have the same location and type of device as s}; Γ=Γ {s}; Reach stop criteria? Yes Fig.. Flowchart of greedy algorithm. o SIMULATIO RESULTS AD AALYSIS Test system modeling In this study, a -bus generic distribution network (GD), as shown in Fig., is used. It comprises kv transmission in-feeds, kv and kv predominantly meshed sub-transmission networks, and kv predominantly radial distribution network. The network consists of lines including overhead lines and underground cables, transformers with various winding connections, loads (including unbalance loads) representing industrial, commercial and domestic loads, and distributed generators (including wind turbines, fuel cells and photovoltaic) connected to kv distribution network. The locations of unbalanced loads and distributed generators are marked by different labels in Fig.. The network is divided into three zones circled by solid red lines. All types of faults are considered. All simulations related to voltage sags are implemented in commercially available DIgSILET/PowerFactory. 4kV kv kv kv kv.kv C D E H I J K L Unbalance Load Zone- Zone- Zone- F G G Fig.. Illustration of zone division in -bus generic distribution network. To present the sag performance more accurately, the variation of load profiles and network parameters are taken into account. Annual hourly loading curves were extracted from survey of different types of loads (including commercial, industrial and residential loads), and operating points are obtained. Since there are different variation patterns for industrial load, commercial load, domestic load and output in terms of day and season, some similar operating conditions reoccur during the year. In the study, Cluster Evaluation of Statistics Toolbox in Matlab is used to find the representative operating conditions. The industrial load, commercial load, domestic load and output are taken as the input to classification approach K-means. With this approach, representative points are obtained. Additionally, further operating points are considered corresponding to the maximum load, the maximum DG output, the maximum wind output, the maximum output, the maximum industrial load, the maximum commercial load, and the maximum domestic load. So, in O CIRED /

4 BPI SGI C D E H I J K C D E H I J K 4 4 SGI L 4 L rd International Conference on Electricity Distribution Lyon, - June Paper total there are characteristic operating points taken into account.. The average of the SGIs obtained using these operating points is taken as the objective function during optimization process. Simulation results The convergence curve which illustrates the relationship between the number of installed devices and SGI obtained by () is given in Fig.. It can be seen that with one device, SGI is significantly improved by.% (.. ). With four devices installed, the obtained SGI. is zero, which means that the sag performance at all buses meets the preset requirements/thresholds. Figure 4 provides the BPIs of all buses obtained with and without optimal mitigation solution, as well as the sag thresholds of all buses. It can be seen without mitigation, a number of BPIs are above the thresholds. After the implementation of mitigation solution obtained by the proposed methodology, BPIs of all buses are well below the corresponding thresholds. Fig.. Convergence characteristics between SGI and the number of devices selected based on (). Without mitigation. With mitigation Threshold umber of devices installed Fig. 4. BPIs obtained without and with mitigation (4 devices) derived based on (). To observe the sag performance of various buses visually, a heat map is used in the study. For each bus, the mean of the BPIs which are corresponding to operating points respectively is calculated and used to generate the heat map of the network. The heat maps obtained without and with mitigation are given in Fig. (a) and (b) respectively. The areas which are exposed to potential disruptive voltage sags are marked in red. It can be seen from Fig. (a) that zone is most vulnerable to voltage sags. However, with the placement of the obtained optimal mitigation solution (SVC at B and DVR at B, B and B), the area performance is greatly improved, as shown in Fig. (b). kv kv kv kv Unbalance load (fixed) on-linear load (fixed) (a) Derived without mitigation DVR SVC Unbalance load (fixed) on-linear load (fixed) (b) Derived with mitigation Fig.. Heatmaps of BPIs obtained without and with mitigation. Efficiency analysis To analyze the efficiency of the proposed methodology, the same optimisation procedure is performed by taking the system index generated in a general way, i.e., SI BPI in (), as the objective function to the greedy algorithm. The relationship between the calculated SGI and the number of devices obtained using SI BPI is given in Fig.. It can be seen that SGI reaches zero after devices are installed, i.e., three more than in case of Fig., % ( 4 ) more 4 resources are required to ensure that sag performance of all buses is within preset thresholds. 4 4 umber of devices installed Fig.. Convergence characteristics between SGI and the number of devices selected based on (). Since only 4 devices are needed to meet differentiated requirements of voltage sags if SGI is used as the objective function, the BPIs of all buses obtained with the installation of 4 devices found based on SI BPI are presented in Fig.. It can be seen that there are three segments of BPIs which are slightly above the thresholds. Between the cases of using indices SI BPI and SGI as objective function, the optimization based on SI BPI attempts to target the buses which have large BPI; while the latter attempts to mitigate the sags at buses which violate the thresholds. The selection of objective function between SI BPI and SGI depends on the concerns and purpose of mitigation. If the main target is to improve the sag performance of the whole network and meanwhile meet the thresholds, general system index, i.e., SI BPI, can be selected. However, if the main concern is to implement the mitigation activity efficiently and fulfill the concept of provisioning differentiated levels of sag STATCOM 4 CIRED 4/

5 BPI BPI rd International Conference on Electricity Distribution Lyon, - June Paper performance, the proposed SGI is better compared to the case of using SI BPI as objective function.. Fig.. BPIs obtained without and with mitigation (4 devices) derived based on (). COCLUSIOS This paper introduces the concept of provision of differentiated levels of voltage sag performance across the network. It proposes a sag gap index (SGI) for this purpose to reflect the level of meeting different performance requirements in different zones of the network. Finally it presents a mitigation strategy to fulfil the provision of differentiated voltage sag performance The initial placement of mitigation devices is decided based on bus sag performance and sensitivity analysis and its followed by application of a greedy algorithm to find the optimal mitigation solution based on SGI. It is demonstrated that choosing optimal placement of devices for the purpose of provision of differentiated sag performance in the network is more effective (% reduction in mitigation effort, and therefore cost) if SGI is used compared to the use of the general system sag index. Acknowledgments This work was supported by SuSTAIABLE Project under Grant. REFERECES Without mitigation With mitigation Threshold [] J. Y. Chan, J. V. Milanović, and A. Delahunty, "Riskbased assessment of financial losses due to voltage sag," IEEE Trans. Power Del., vol., pp. -,. [] M. H. J. Bollen and D. D. Sabin, "International coordination for voltage sag indices," in Proc. IEEE PES Transmission and Distribution Conference and Exhibition,, pp. -4. [] S. R. aidu, G. V. de Andrade, and E. G. da Costa, "Voltage sag performance of a distribution system and its improvement," IEEE Trans. Ind. Appl., vol. 4, pp. - 4,. [4] R. A. Barr, V. J. Gosbell, and I. McMichael, "A new SAIFI based voltage sag index," in Proc. th Int. Conf. on Harmonics Quality of Power,, pp. -. [] M. H. J. Bollen and I. Y. H. Gu, Signal Processing of Power Quality Disturbances. ew York, Y, USA: IEEE Press,. [] "IEEE Guide for Voltage Sag Indices," IEEE Std - 4, June 4, pp. -, 4. [] IEEE Recommended practice for evaluating electric power system compatibility with electronic process equipment: IEEE Std -,. [] H. L. Liao, S. Abdelrahman, and J. V. Milanović, "Identification of weak areas of power network based on exposure to voltage sags part I: development of sag severity index for single-event characterization," IEEE Trans Power Del. (accepted), 4. [] H. L. Liao, S. Abdelrahman, Y. Guo, and J. V. Milanović, "Identification of weak areas of power network based on exposure to voltage sags Part II: assessment of network performance using sag severity index," IEEE Trans Power Del. (accepted), 4. [] H. Masdi,. Mariun, S. Mahmud, A. Mohamed, and S. Yusuf, "Design of a prototype D-STATCOM for voltage sag mitigation," in Proc. ational Power and Energy Conf., Kuala Lumpur, Malaysia, 4, pp. -. [] Y. Xiao, Y. H. Song, C. C. Liu, and Y. Z. Sun, "Available transfer capability enhancement using FACTS devices," IEEE Trans. Power Syst., vol., pp. -,. [] E. Ghahremani and I. Kamwa, "Optimal placement of multiple-type FACTS devices to maximize power system loadability using a generic graphical user interface," IEEE Trans. Power Syst., vol., pp. -,. [] J. V. Milanović and Y. Zhang, "Modeling of FACTS devices for voltage sag mitigation studies in large power systems," IEEE Trans. Power Del., vol., pp. -,. [4] Y. Zhang and J. V. Milanović, "Global voltage sag mitigation with FACTS-based devices," IEEE Trans. Power Del., vol., pp. -,. [] M. M. El Metwally, A. A. El Emary, F. M. El Bendary, and M. I. Mosaad, "Using FACTS controllers to balance distribution systems based A," in Proc. th Int. Middle East Power Systems Conf.,, pp. -. [] R. Grunbaum, "FACTS for voltage control and power quality improvement in distribution grids," in Proc. CIRED Seminar Smart Grids for Distribution,, pp. -4. [] J. V. Milanović and Z. Yan, "Global minimization of financial losses due to voltage sags with FACTS based devices," IEEE Trans. on Power Del., vol., pp. -,. [] F. B. Alhasawi and J. V. Milanović, "Techno-economic contribution of FACTS devices to the operation of power systems with high level of wind power integration," IEEE Trans. Power Syst., vol., pp. -4,. [] J. Thomas, "Development of methodology for online provision of differentiated quality of electricity supply," MSc. Electrical Power Systems, Fac. of Eng. and Phys. Sciences, The University of Manchester,. CIRED /