HARMONIC ANALYSIS OF LIGHTING TECHNOLOGY APPLICATION CASE STUDY IN DISTRIBUTION NETWORK: SMART CITY BUZIOS

Transcription

1 HARMONIC ANALYSIS OF LIGHTING TECHNOLOGY APPLICATION CASE STUDY IN DISTRIBUTION NETWORK: SMART CITY BUZIOS Ana PAULA, Adriano FRAGOSO Márcio FORTES, Vitor FERREIRA Antonio PEREIRA LABLUX/UFF Brazil PPGEET/TEE/UFF Brazil LABLUX/UFF - Brazil acspaula@id.uff.br, lablux@vm.uff.br mzf@vm.uff.br, vitor@vm.uff.br am_estrela@id.uff.br ABSTRACT This work presents the impact assessment in quality of energy resulting from the implementation of residential artificial lighting and public lighting technologies in the Smart City Búzios, Rio de Janeiro State, Brazil. In the scope are the measurements made in LABLUX - Technical Lighting Laboratory of Federal Fluminense University, with compact fluorescent lamps and lamps (Light Emitting Diode), and two different types of luminaires: Sodium Vapour and. The simulation of Búzios network was carried out with help of HarmZs, software developed by CEPEL - Center for Electricity Energy Research. Eight application scenarios of lamps and luminaires were made, and checked the voltage distortion generated in the 13.8 kv network, by analysing the criteria established by PRODIST (8 th Module).Through this analysis it was shown that the technology has advantages with regard to reduction of distortions as compared to other technologies, for residential lighting or public roads. However, it is necessary a continuous improving of the technology, and new evaluations with the criteria of PRODIST or the long-existing regulation, to ensure the least possible interference harmonics in Power System. INTRODUCTION With the development and evolution of technology, and in parallel with the advancement of power electronics, some equipment became known and used more significantly by society both in residences and in industries. In the case of non-sinusoidal waves, the load is generated by a non-sinusoidal voltage, which causes distortions in the current waveform, producing nonsinusoidal waves. [1] Examples of non-linear household appliances include: microwave appliances, televisions, computers and fluorescent lamps. In the industry, the main ones are: frequency inverters and discharge lamps. In Brazil, more specifically in the area of lighting, technology has been widely used in various applications such as traffic lights, public lighting, residential lighting and emergency lighting. Public lighting and traffic light are stimulated by a subprogram of PROCEL (National Program for the Conservation of Electric Energy), PROCEL RELUZ. However, it is important to note that luminous efficiency does not necessarily means power quality. According to IEEE standard , power quality can be described as providing adequate power and grounding, aiming the correct operation of equipment. [2] In this context, the tests for conformity assessment with regard to energy quality are inserted. Conformity assessment seeks to provide the consumer with the appropriate degree of confidence that a product, process or service meets predefined requirements in standards or technical regulations. [3] According to ABILUMI - Brazilian Association of Importers of Lighting Products, the number of bulbs sold in Brazil increased from 4 million in 2011 to 25 million in The need for more economical alternatives due to the increase in electric energy tariffs, in addition to the gradual departure of incandescent bulbs from the Brazilian market and the fall in product prices favored this growth of use in the country. [4] Thus, with the growth of the use of this technology and due to this application, it is necessary to evaluate its use not only in terms of luminous efficiency and energy savings, but also, its impacts on the electricity network as regards energy quality. This work aims to analyze this impact on the quality of energy with a focus on harmonic distortions. Different scenarios of applications of lamps and Luminaires, compact fluorescent lamps (CFL s) and Sodium Vapor (SV) Luminaires were simulated and a comparison of the products was performed to verify if the technology corresponds to an alternative solution, with respect to the quality criteria of energy. METHODOLOGY Description of the tests The tests were performed in LABLUX Technical Lighting Laboratory of Federal Fluminense University. LABLUX has expertise and structure with equipment and controlled environmental conditions needed for the development of tests. Some examples of researches by LABLUX are discussed in references. [5-6] In this work, tests were carried out on light bulbs and a luminaire. Other two types of technologies were also tested: Compact Fluorescent Lamps and sodium vapor (SV) luminaire. All tests were performed at nominal voltage of 220 V, at controlled temperature of 25 C± 1 C. Equipment The harmonics distortions were measured in an Embrasul Power Quality Analyzer Model RE 600, connected to the input of LABLUX measuring room. The control and CIRED /5

2 interface between the operator and the measuring system is made by using the computer software installed. Figure 1 shows the connection for the test performed in the Laboratory and the analyzer under measurement. PRODIST Figure 1- Test connection in Laboratory The PRODIST - Procedure for Distribution of Electric Energy in the National Electric System - are documents developed by ANEEL- Brazilian Energy Agency - together with associations of the Brazilian Electric Sector that establish criteria and standardization for technical activities of the systems in Distribution of electricity. In this work the focus is given in Module 8 Power Quality. This module addresses requirements both regarding the quality of the product and the service provided. The reference values established by PRODIST for total voltage harmonic distortions in distribution bars are given in Table 1. [7] Table 1 PRODIST Limits for Voltage THD Nominal Voltage in bar Limit THD Voltage (%) V N 1 kv 10 1 kv < V N 13.8 kv kv < V N 69 kv 6 69 kv < V N < 230 kv 3 MEASUREMENT DATA The tests were initially performed with no load attached and later each set of lamps and luminaires according to their technology were individually powered. Table 2 shows the characteristics of the tested that were chosen from different models in the Laboratory and Table 3 shows the harmonic data of each technology measured. Table 2 - Characteristics of tested Technology Model 14W 127/220V 6000K 12W 127/220V 6000K LAMP 9W 127/220V 6000K 5W 127/220V 6000K 9W 127/220V 2700K 20W 220V 6400K 15W 220V 6400K CFL 25W 220V 6400K 20W 220V 6400K 15W 220V 6400K LUMINAIRE 130W V 6173K SV LUMINAIRE 150W 220V 1960K Frequency (Hz) Table 3 - Harmonics Measurements CURRENT INDIVIDUAL HARMONICS (% FUND) FCL Luminaire SV Luminaire THD (%) CIRED /5

3 CASE STUDY: BUZIOS ELECTRICAL NETWORK The use of an electronic load individually, due to the small value of power, does not interfere in a damaging way in the electrical system. However, when applied in a large set, these loads can affect the quality of the energy supplied, causing losses and increasing losses for all agents of the national interconnected system. [8] In this step, a case study was carried out from the harmonics measurements generated by the contribution of each lighting technology. These loads were applied to a real network through a simulation in HarmZs software. For this modelling part of the electrical network of the city of Búzios was chosen, complementing one of the lines of research of this project that is the analysis of new sources of distributed generation to the network. In this particular case, the insertion of new generation sources is not being evaluated, but rather the massive insertion of new technology, especially lighting technologies. The load of the region of Búzios is predominantly residential and in 2012, customers were served including commercial and industrial. In this analysis, Búzios feeder 7 was chosen, responsible for feeding 1,583 customers. [9] According to data provided by the concessionaire, the region presents an increase of 5% per year, providing an estimate of the number of customers in each feeder for the coming years. The prediction for feeder 7 in 2016 was about 1,924 customers. For the simulation, an average of 25 lamps was used for each residence supplied by feeder 7. For public lighting simulation, 800 luminaires were used in the total length of the feeder (approximately 24 km), one every 30 meters (average distance for positioning of public lighting poles). In this way, approximately 50 luminaires were placed in each bar. The original system is composed of 116 bars and had been reduced to 16 bars due to software limitations. A power factor of 0.9 was applied for all the bars and the base voltage used was 13.8 kv. The harmonic current distortion values measured in the Laboratory in 220V were converted to the voltage level of 13.8 kv (network voltage). Eight different scenarios were constructed to analyse the behaviour of the network with the insertion of harmonic sources of current. Scenario 1 and 2: CFL lamp applied on the bars First, in scenario 1, CFL s were applied to 3 bars of the system in order to analyse the THD of each. The bars that received compact fluorescent lamps were 9 and 16. The remaining bars of the system were not changed. In scenario 2, CFL s were applied to all 16 system bars. The results presented by the software for voltage THD in each bar are shown in Table 4. Table 4 Results for scenarios 1 and 2 9,16 9,16 Bar Bar Bar Bar Bar Bar Bar Bar Bar Bar Bar Bar Bar Bar Bar Bar It can be noted that, the increase of the quantity of CFL s cause a significant increase in voltage THD in. The values found are below the required for PRODIST in scenario 1 and above PRODIST limits in scenario 2. Scenario 3 and 4: lamp applied on the bars In scenario 3 lamps were applied to 3 bars of the system in order to analyse the THD of each. The bars that lamps were 2, 9 and 16. The remaining bars of the system were not changed. In scenario 4, s were applied to all 16 system bars. The results presented by the software for voltage THD in each bar are shown in Table 5. Table 5 - Results for scenarios 3 and 4 9, 16 9, 16 Bar Bar Bar Bar Bar Bar Bar Bar Bar Bar Bar Bar Bar Bar Bar Bar It can be seen that, in the same way as with fluorescent lamps, with the increase of the number of lamps applied, there is a significant increase in the voltage THD in each bar. However, with the values are still below the required by PRODIST for a voltage level of 13.8 kv. Scenario 5 and 6: Luminaires applied on all the bars In this case, first 50 SV luminaires were distributed per bar in order to analyse the THD of voltage. After this, the same was verified for 100 SV luminaires per bar. Then, CIRED /5

4 the case was done for Luminaires. The results are shown in Table 6. Table 6 for scenario 5 and 6 Technology SV Bar Number The values found for both technologies with 50 luminaires are within the limits of PRODIST. However, in a projection of expansion of this network and inclusion of more luminaires, the tendency is the increase of this value. Scenario 7: SV Luminaires mixed with CFL on all bars In this scenario, the compact fluorescent lamps used were placed in the bars along with public lighting, using sodium vapour lamps, in order to analyze the voltage THD in each of the bars generated by the residential illumination with public lighting. The results presented by the software for voltage THD in each bar are shown in Table 7. Table 7 - Results for scenario 5 Bar Bar THDv Number Number (%) The voltage THD values found for this scenario are above the limits specified by PRODIST, which is 8% for the voltage level of 13.8 kv. Scenario 8: Luminaires mixed with lamps on In this scenario 50 luminaires were distributed per bar with lamps in order to analyse the total distortion of voltage. The results presented by the software for Voltage THD in each bar are shown in Table 8. Table 8 - Results for scenario The Total Voltage Distortion values found for this scenario are above the limits specified by PRODIST, which is 8% for the voltage level of 13.8 kv. Summary of Scenarios Results Table 9 summarizes the results for the different scenarios simulated in this work and also the compliance with the requirements of Voltage Total Harmonic Distortion, established by PRODIST. It is important to note that, in this analysis, a balanced network was considered and only the harmonics generated by the illumination were analyzed, disregarding the other loads in residences. Table 9 Summary of results THDv PRODIST TECNOLOGY average LIMIT RESULT FCL 9,16 2% PASS FCL 12% FAIL 9,16 1.5% PASS 5% PASS Luminaires (50 ) 2% PASS Luminaires 4.5% PASS (100 ) 8% SV Luminaires (50 2% PASS ) SV Luminaires (100 ) 6% PASS SV Luminaire+CFL 15% FAIL Luminaire+ 9% FAIL CONCLUSION In this work, concepts were presented on the generation of harmonics through the application of artificial lighting. It was presented a case study of the impacts caused in the distribution network of the city of Búzios, considered balanced and modeled with 16 bars, using as criteria the CIRED /5

5 harmonic distortion levels established by PRODIST, for the voltage level of 13.8 kv. The work was restricted to the evaluation of harmonics generated by the lighting, disregarding the other loads existing in the residences. In the first two scenarios, only the effects of residential lighting in the electrical network were analysed, applying loads in 3 of the 16 modelled bars. It was noticed that with the use of compact fluorescent lamps, the voltage distortion was of the order of 4% while with s is in the order of 1.5% in each bar, both below the PRODIST limits. - In scenarios 3 and 4, the purpose was to increase the amount of residential lighting, in order to evaluate a horizon of expansion of application of the technology. The simulation took into account the existence of 1924 residences, using or CFL lamps connected to the 16 system bars. It was noticed that with the increase of the number of lamps used, the distortions in the bars also increased, presenting about 12% for CFL above PRODIST limits, whereas with, the distortions are of the order of 5%, below PRODIST limits. - Scenarios 5 and 6 evaluated the impacts of the type of luminaire chosen for public lighting. The simulation took into account the application of 50 sodium vapour as well as 50 luminaires in each of the system bars, totalling 800 luminaires. With the use of SV luminaires, the voltage distortion was about 3%, while that with luminaires was 2%. Both values are below the PRODIST limits. The application of double luminaires was also simulated and a significant increase of these values was observed, increasing to 6% for SV luminaire and 4% for. A more significant contribution of residential lighting was observed when evaluated separately from public lighting. - In scenarios 7 and 8, the purpose was to verify the harmonic impact of residential lighting with public lighting. One of them applying CFL's and SV luminaire, obtained a distortion of 15%, a value above the limit allowed by PRODIST. In the other case, lamps were applied with luminaires and the voltage THD per bar was of the order of 9%, also above the PRODIST limit. In these studies, simulation scenarios were generated to evaluate the impacts of lighting technologies on the real network studied. It is known that the scenarios can be conservative or extreme in the characterization of the actual load (it was considered a number of lamps per residence that can surpass what is actually applied). In this study, has been observed that more critical and close evaluations of actual data of the consumers connected should be performed. Knowledge of each customer's load curves is essential for this assessment. With the data currently available, this evaluation is not possible, but with the implementation of smart meters (this proposal of the Búzios Smart City project), this characterization will be possible and studies closer to reality can be made. Regardless of the characterization of the scenarios used, it is noticed that the massive deployment of new lighting technologies should be discussed in detail considering the impact they can cause in the network in the aspect of voltage distortion, in particular. There is a need to continue improving the regulations for reduce harmonic generation in new lighting equipment. Thus, it was concluded that this work provided important information for the area of power quality, as it presented cases of artificial lighting technology implantation. These cases may be occurring in several cities in Brazil, where lighting systems are being replaced. In addition, this work seeks to communicate to professionals in the field, the importance of considering in the implementation of efficiency projects with the use of new lighting technologies, the possible effects of these equipment on harmonic distortions in the electrical network and effects on equipment and installations already in operation. REFERENCES [1] LEÃO et al., Harmônicos em sistemas elétricos. Elsevier. Brazil. [2] PIRES, Igor Amariz Conceitos Iniciais - Qualidade de Energia e Harmônicos. Journal: O Setor Elétrico, [3] INMETRO, Portaria 453 de 17 de setembro de 2013 Vocabulário Inmetro de Avaliação da Conformidade. [4] ABILUMI, : a tecnologia do presente. Journal: Lumiere Electric, vol 207. [5] FASSARELA et al., 2014 Analysis and Suggested Solution of Power Quality in Lighting Laboratory, IEEE Latin America Trans., vol.12, pp , doi: /tla [6] Fortes et al., 2015 Analysis of CLF s Power Quality in the Brazilian Market, in Proc IEEE Workshop on Power Electronics and Power Quality Applications PEPQA, Bogota, 2015, pp.1-5. doi: /PEPQA [7] PRODIST, 2016 Procedimentos de Distribuição de Energia Elétrica, módulo 8 Qualidade da Energia Elétrica ANEEL. [8] Souza et al., Campanhas de medição dos indicadores de flutuação, harmônicos e desequilíbrio de tensão na Rede Básica constatações e resultados. IEEE. [9]Ampla, Levantamento do Perfil Energético de Búzios CIRED /5