Pub Date : 2023-04-20DOI: 10.30811/teknologi.v23i1.3715
S. Safrida, S. Suryani, Zuhra Amalia
The production of eco-enzyme usually takes a long time for fermentation, which is three months using natural microbes found in fruit peels. This research is experimental by focusing on accelerating the production of the eco-enzyme using two types of microbes, namely saccharomyces cerevisiae and aspergillus oryzae whose mass is varied by 5 variations, and the fermentation time is set for 20 days. The results showed that the eco-enzyme made by adding microbes had the same characteristics as the natural fermentation eco-enzyme. Furthermore, the eco-enzyme obtained from the research was added as an antiseptic in making solid soap. From the test results, the best inhibitory power was obtained in the eco-enzyme with the addition of saccharomyces cerevisiae as much as 10 g and aspergillus oryzae as much as 25 g. The inhibitory power test was carried out by comparing the number of bacteria present on unwashed hands with hands that had been washed using soap. The test results showed the number of bacteria on unwashed hands was 85 CFU, which were washed using samples with aspergillus oryzae as much as 4 CFU, with saccharomyces cerevisiae as much as 2 CFU, this inhibition is better than commercial antiseptic soap with a total of 8 CFU
{"title":"Pengaruh Penambahan Saccharomyces Cerevisiae dan Aspergillus Oryzae terhadap Karakteristik Eco-Enzyme serta Pengaplikasiannya dalam Pembuatan Sabun Padat Antiseptik","authors":"S. Safrida, S. Suryani, Zuhra Amalia","doi":"10.30811/teknologi.v23i1.3715","DOIUrl":"https://doi.org/10.30811/teknologi.v23i1.3715","url":null,"abstract":"The production of eco-enzyme usually takes a long time for fermentation, which is three months using natural microbes found in fruit peels. This research is experimental by focusing on accelerating the production of the eco-enzyme using two types of microbes, namely saccharomyces cerevisiae and aspergillus oryzae whose mass is varied by 5 variations, and the fermentation time is set for 20 days. The results showed that the eco-enzyme made by adding microbes had the same characteristics as the natural fermentation eco-enzyme. Furthermore, the eco-enzyme obtained from the research was added as an antiseptic in making solid soap. From the test results, the best inhibitory power was obtained in the eco-enzyme with the addition of saccharomyces cerevisiae as much as 10 g and aspergillus oryzae as much as 25 g. The inhibitory power test was carried out by comparing the number of bacteria present on unwashed hands with hands that had been washed using soap. The test results showed the number of bacteria on unwashed hands was 85 CFU, which were washed using samples with aspergillus oryzae as much as 4 CFU, with saccharomyces cerevisiae as much as 2 CFU, this inhibition is better than commercial antiseptic soap with a total of 8 CFU","PeriodicalId":47541,"journal":{"name":"Jurnal Teknologi-Sciences & Engineering","volume":" ","pages":""},"PeriodicalIF":0.7,"publicationDate":"2023-04-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"43888778","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-04-20DOI: 10.30811/teknologi.v23i1.3602
Teuku Tarich Akbar, Reza Fauzan, A. Putra
This gas source that comes from the bowels of the earth contains many impurities and heavy hydrocarbons, both of which must be separated or even eliminated. To purify the treated gas to prevent damage to process equipment or pipe clogging. The removal process includes H 2 S removal, CO 2 removal, oily water removal, mercury removal, and water content removal in natural gas. This study aims to determine the effect of flow rate and ambient temperature differences on the performance of the wellstream cooler and the level of efficiency of the heat transfer process in the wellstream cooler. Observations will be focused on the offshore platform wellstream cooler cooling system at PT. PHE NSO. Based on observational data, the results obtained using the Hysys method by entering the required data completely, Hysys will automatically perform calculations according to what is required and the calculation results will be plotted in graphical form. Heat transfer efficiency based on Hysys application calculations all show results of 99%. Thus it can be seen the effect of changes in flow rate and ambient temperature on the heat transfer process. The greater the flow rate, the greater the heat transfer that occurs and the higher the ambient temperature, the smaller the heat transfer occurs, so that the lower the ambient temperature, the more effective the heat transfer occurs.
{"title":"Optimasi Kinerja Wellstream Cooler Sebagai Sistem Pendinginan Gas Alam dari Sumuran terhadap Pengaruh Laju Alir dan Perbedaan Suhu Lingkungan di Platform PHE NSO Offshore","authors":"Teuku Tarich Akbar, Reza Fauzan, A. Putra","doi":"10.30811/teknologi.v23i1.3602","DOIUrl":"https://doi.org/10.30811/teknologi.v23i1.3602","url":null,"abstract":"This gas source that comes from the bowels of the earth contains many impurities and heavy hydrocarbons, both of which must be separated or even eliminated. To purify the treated gas to prevent damage to process equipment or pipe clogging. The removal process includes H 2 S removal, CO 2 removal, oily water removal, mercury removal, and water content removal in natural gas. This study aims to determine the effect of flow rate and ambient temperature differences on the performance of the wellstream cooler and the level of efficiency of the heat transfer process in the wellstream cooler. Observations will be focused on the offshore platform wellstream cooler cooling system at PT. PHE NSO. Based on observational data, the results obtained using the Hysys method by entering the required data completely, Hysys will automatically perform calculations according to what is required and the calculation results will be plotted in graphical form. Heat transfer efficiency based on Hysys application calculations all show results of 99%. Thus it can be seen the effect of changes in flow rate and ambient temperature on the heat transfer process. The greater the flow rate, the greater the heat transfer that occurs and the higher the ambient temperature, the smaller the heat transfer occurs, so that the lower the ambient temperature, the more effective the heat transfer occurs.","PeriodicalId":47541,"journal":{"name":"Jurnal Teknologi-Sciences & Engineering","volume":" ","pages":""},"PeriodicalIF":0.7,"publicationDate":"2023-04-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"43213179","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The LNG Receiving Hub facility is an existing facility that has not been used for a long time. In order for this facility can be used again, all equipment and piping systems must first be conditioned to a temperature close to the operational temperature of LNG, which is -160 o C. The purpose of this approach is to determine the theoretical and actual volume of LNG used for the cooling down process of the LNG Hub facility, as well as to compare it with the cooling down process for the regasification facility. An analytical approach is used in this study to determine the ideal amount of LNG needed for cooling down the facility and compare it with the actual amount of LNG that is being used for cooling down the facility. From the results of the analysis, the theoretical LNG volume needed for cooling down the LNG Hub facility is 8,176 m 3 and the actual volume is 7,013.6620 m 3 , while the actual value for cooling down the regasification facility is 11,573.19 m 3 . The actual volume for the LNG Hub facility is smaller compared to regasification which can be caused by the smaller volume of pipelines and the difference in the cooling down mechanism in both facilities.
{"title":"Analisis Volume LNG Ideal untuk Cooling Down Fasilitas LNG Hub","authors":"Dhian Nugraha, Harunsyah Harunsyah, Munawar Munawar","doi":"10.30811/teknologi.v23i1.3687","DOIUrl":"https://doi.org/10.30811/teknologi.v23i1.3687","url":null,"abstract":"The LNG Receiving Hub facility is an existing facility that has not been used for a long time. In order for this facility can be used again, all equipment and piping systems must first be conditioned to a temperature close to the operational temperature of LNG, which is -160 o C. The purpose of this approach is to determine the theoretical and actual volume of LNG used for the cooling down process of the LNG Hub facility, as well as to compare it with the cooling down process for the regasification facility. An analytical approach is used in this study to determine the ideal amount of LNG needed for cooling down the facility and compare it with the actual amount of LNG that is being used for cooling down the facility. From the results of the analysis, the theoretical LNG volume needed for cooling down the LNG Hub facility is 8,176 m 3 and the actual volume is 7,013.6620 m 3 , while the actual value for cooling down the regasification facility is 11,573.19 m 3 . The actual volume for the LNG Hub facility is smaller compared to regasification which can be caused by the smaller volume of pipelines and the difference in the cooling down mechanism in both facilities.","PeriodicalId":47541,"journal":{"name":"Jurnal Teknologi-Sciences & Engineering","volume":" ","pages":""},"PeriodicalIF":0.7,"publicationDate":"2023-04-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"45833107","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-04-20DOI: 10.30811/teknologi.v23i1.3781
Robert Napitupulu, S.ST., M.T., Yuli Dharta, Diviya Arsieka Putri, Devaned Parlindungan, A. Pratama, Clara Lavita Angelina
{"title":"Perancangan Mesin Pencacah Limbah Kelapa Muda Secara Vertikal Menggunakan Metode Sintesis Pandangan","authors":"Robert Napitupulu, S.ST., M.T., Yuli Dharta, Diviya Arsieka Putri, Devaned Parlindungan, A. Pratama, Clara Lavita Angelina","doi":"10.30811/teknologi.v23i1.3781","DOIUrl":"https://doi.org/10.30811/teknologi.v23i1.3781","url":null,"abstract":"","PeriodicalId":47541,"journal":{"name":"Jurnal Teknologi-Sciences & Engineering","volume":" ","pages":""},"PeriodicalIF":0.7,"publicationDate":"2023-04-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"49599220","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-04-20DOI: 10.30811/teknologi.v23i1.3819
Munawar Munawar
Broadband Wireless Access (BWA) is useful for spreading and improving internet services evenly to all regions in Indonesia. While the frequency spectrum used in this Broadband Wireless Access (BWA) service is 2.3 GHz. The antenna gain test method is carried out with 3 antenna models working at the same frequency, and before the gain test is carried out, it is confirmed that the Voltage Standing Wave Ratio (VSWR) value ≤ 2 on the 2.3 GHz frequency spectrum. The test results of the parabolic Bowtie reflector antenna show that the gain value for each diameter is 6.35 dB, 7.3 dB, 7.45 dB
宽带无线接入(BWA)有助于在印度尼西亚的所有地区均匀地传播和改善互联网服务。而该宽带无线接入(BWA)业务使用的频谱是2.3 GHz。天线增益测试方法采用3个天线型号在同一频率下工作，在进行增益测试前，确认在2.3 GHz频谱上电压驻波比(VSWR)值≤2。抛物线形鲍蒂反射面天线的测试结果表明，各直径下的增益值分别为6.35 dB、7.3 dB、7.45 dB
{"title":"Antena Bowtie dengan Reflektor Parabola untuk Layanan Broadband Wireless Access","authors":"Munawar Munawar","doi":"10.30811/teknologi.v23i1.3819","DOIUrl":"https://doi.org/10.30811/teknologi.v23i1.3819","url":null,"abstract":"Broadband Wireless Access (BWA) is useful for spreading and improving internet services evenly to all regions in Indonesia. While the frequency spectrum used in this Broadband Wireless Access (BWA) service is 2.3 GHz. The antenna gain test method is carried out with 3 antenna models working at the same frequency, and before the gain test is carried out, it is confirmed that the Voltage Standing Wave Ratio (VSWR) value ≤ 2 on the 2.3 GHz frequency spectrum. The test results of the parabolic Bowtie reflector antenna show that the gain value for each diameter is 6.35 dB, 7.3 dB, 7.45 dB","PeriodicalId":47541,"journal":{"name":"Jurnal Teknologi-Sciences & Engineering","volume":" ","pages":""},"PeriodicalIF":0.7,"publicationDate":"2023-04-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"45564799","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-04-19DOI: 10.11113/jurnalteknologi.v85.19221
N. I. Alimon, N. Sarmin, A. Erfanian
In chemistry, the molecular structure can be represented as a graph. Based on the information from the graph, its characterization can be determined by computing the topological index. Topological index is a numerical value that can be computed by using some algorithms and properties of the graph. Meanwhile, the non-commuting graph is a graph, in which two distinct vertices are adjacent if and only if they do not commute, where it is made up of the non-central elements in a group as a vertex set. In this paper, the Szeged index of the non-commuting graph of some finite groups are computed. This paper focuses on three finite groups which are the quasidihedral groups, the dihedral groups, and the generalized quaternion groups. The construction of the graph is done by using Maple software. In finding the Szeged index, some of the previous results and properties of the graph for the quasidihedral groups, the dihedral groups, and the generalized quaternion groups are used. The generalisation of the Szeged index of the non-commuting graph is then established for the aforementioned groups. The results are then applied to find the Szeged index of the non-commuting graph of ammonia molecule.
{"title":"ON THE SZEGED INDEX AND ITS NON-COMMUTING GRAPH","authors":"N. I. Alimon, N. Sarmin, A. Erfanian","doi":"10.11113/jurnalteknologi.v85.19221","DOIUrl":"https://doi.org/10.11113/jurnalteknologi.v85.19221","url":null,"abstract":"In chemistry, the molecular structure can be represented as a graph. Based on the information from the graph, its characterization can be determined by computing the topological index. Topological index is a numerical value that can be computed by using some algorithms and properties of the graph. Meanwhile, the non-commuting graph is a graph, in which two distinct vertices are adjacent if and only if they do not commute, where it is made up of the non-central elements in a group as a vertex set. In this paper, the Szeged index of the non-commuting graph of some finite groups are computed. This paper focuses on three finite groups which are the quasidihedral groups, the dihedral groups, and the generalized quaternion groups. The construction of the graph is done by using Maple software. In finding the Szeged index, some of the previous results and properties of the graph for the quasidihedral groups, the dihedral groups, and the generalized quaternion groups are used. The generalisation of the Szeged index of the non-commuting graph is then established for the aforementioned groups. The results are then applied to find the Szeged index of the non-commuting graph of ammonia molecule.","PeriodicalId":47541,"journal":{"name":"Jurnal Teknologi-Sciences & Engineering","volume":" ","pages":""},"PeriodicalIF":0.7,"publicationDate":"2023-04-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"45369420","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-04-19DOI: 10.11113/jurnalteknologi.v85.19440
Maryam Zahaba, Ungku Aida Farzana Ungku Azly, Nur Sarah Fatihah Tamsi, Saiful Arifin Shafiee, Hazrin Abdul Hadi, M. Yusof, Mohd Shukri MOHD ARIS, Mohd Norhafsam Maghpor, Nor Mohd Razif Norain
Exposure to types of asbestos such as chrysotile and crocidolite increases respiratory disease risks such as lung cancer, mesothelioma, and asbestosis. Nevertheless, asbestos products banning in Malaysia is only limited to crocidolite as per stated in OSHA (Prohibition of Use of Substance) Order 1999, though other types are highly suspected to be found in asbestos-containing materials (ACM) like brake pads and linings. This study ascertains the presence of asbestos fibres, particularly chrysotile, in brake pads and linings used in Malaysia's vehicle and heavy vehicle sector. Seven different brake pads; three from vehicle brands, and four from heavy vehicle brands were collected by bulk sampling approaches from the market and field. Dust fibres were extracted using slow grinding method and analysed under Polarized Light Microscope (PLM). The fibre characteristics such as colour, morphology, pleochroism, extinction, and dispersion staining technique were examined, referring the National Institute for Occupational Safety and Health (NIOSH) Manual of Analytical Methods (NMAM) 9002. Additionally, the observed samples also were verified by an accredited lab to support the authenticity of the outcome. From the analysis and lab results, chrysotile fibres were consistently detected in all brake pad samples, fulfilling the fibre characteristics and positive elongation signs.
{"title":"IDENTIFICATION OF CHRYSOTILE IN BRAKE PADS AND LININGS FROM MALAYSIAN VEHICLES AND HEAVY VEHICLES BY USING POLARIZED LIGHT MICROSCOPE (PLM)","authors":"Maryam Zahaba, Ungku Aida Farzana Ungku Azly, Nur Sarah Fatihah Tamsi, Saiful Arifin Shafiee, Hazrin Abdul Hadi, M. Yusof, Mohd Shukri MOHD ARIS, Mohd Norhafsam Maghpor, Nor Mohd Razif Norain","doi":"10.11113/jurnalteknologi.v85.19440","DOIUrl":"https://doi.org/10.11113/jurnalteknologi.v85.19440","url":null,"abstract":"Exposure to types of asbestos such as chrysotile and crocidolite increases respiratory disease risks such as lung cancer, mesothelioma, and asbestosis. Nevertheless, asbestos products banning in Malaysia is only limited to crocidolite as per stated in OSHA (Prohibition of Use of Substance) Order 1999, though other types are highly suspected to be found in asbestos-containing materials (ACM) like brake pads and linings. This study ascertains the presence of asbestos fibres, particularly chrysotile, in brake pads and linings used in Malaysia's vehicle and heavy vehicle sector. Seven different brake pads; three from vehicle brands, and four from heavy vehicle brands were collected by bulk sampling approaches from the market and field. Dust fibres were extracted using slow grinding method and analysed under Polarized Light Microscope (PLM). The fibre characteristics such as colour, morphology, pleochroism, extinction, and dispersion staining technique were examined, referring the National Institute for Occupational Safety and Health (NIOSH) Manual of Analytical Methods (NMAM) 9002. Additionally, the observed samples also were verified by an accredited lab to support the authenticity of the outcome. From the analysis and lab results, chrysotile fibres were consistently detected in all brake pad samples, fulfilling the fibre characteristics and positive elongation signs.","PeriodicalId":47541,"journal":{"name":"Jurnal Teknologi-Sciences & Engineering","volume":" ","pages":""},"PeriodicalIF":0.7,"publicationDate":"2023-04-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"49422590","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-04-19DOI: 10.11113/jurnalteknologi.v85.18684
Hanis Farhah Jamahori, M. Abdullah, Abid Ali
Most distributed renewable energy generation (DREG) planning studies are performed using a constant load model and a dispatchable generation unit. However, the renewable generation unit and load demand vary in real life, and the generation size at the peak demand varies accordingly with loading levels. Such considerations may lead to the erroneous conclusion: the power loss reduction and bus voltage improvement may not be optimal. Consequently, the generation unit must be adequately integrated to offer optimal capacity in the distribution system while considering non-constant load demand as a part of DREG planning. Therefore, the impact of integrating photovoltaic (PV) considering historical solar weather data and varying load demand for five different voltage-dependent load models is proposed in this study. Particle swarm optimization (PSO) is employed to find the optimal location and size of PV with the objective to minimize power losses in the distribution system using IEEE 33-bus and IEEE 69-bus test systems. The findings are evaluated based on the comparative analysis of power losses reduction, PV penetration level, power loss index, and voltage deviation index. Findings revealed that the proposed model is effective in determining the optimal location and size of PV with a significant reduction of power losses that varies between 13.84% to 32.71% in 33-bus, and between 18.56% to 43.80% in 69-bus. In addition, the improvement in minimum bus voltage and other performance indices are also significant.
{"title":"IMPACT AND EVALUATION OF OPTIMIZED PV GENERATION IN THE DISTRIBUTION SYSTEM WITH VARYING LOAD DEMANDS","authors":"Hanis Farhah Jamahori, M. Abdullah, Abid Ali","doi":"10.11113/jurnalteknologi.v85.18684","DOIUrl":"https://doi.org/10.11113/jurnalteknologi.v85.18684","url":null,"abstract":"Most distributed renewable energy generation (DREG) planning studies are performed using a constant load model and a dispatchable generation unit. However, the renewable generation unit and load demand vary in real life, and the generation size at the peak demand varies accordingly with loading levels. Such considerations may lead to the erroneous conclusion: the power loss reduction and bus voltage improvement may not be optimal. Consequently, the generation unit must be adequately integrated to offer optimal capacity in the distribution system while considering non-constant load demand as a part of DREG planning. Therefore, the impact of integrating photovoltaic (PV) considering historical solar weather data and varying load demand for five different voltage-dependent load models is proposed in this study. Particle swarm optimization (PSO) is employed to find the optimal location and size of PV with the objective to minimize power losses in the distribution system using IEEE 33-bus and IEEE 69-bus test systems. The findings are evaluated based on the comparative analysis of power losses reduction, PV penetration level, power loss index, and voltage deviation index. Findings revealed that the proposed model is effective in determining the optimal location and size of PV with a significant reduction of power losses that varies between 13.84% to 32.71% in 33-bus, and between 18.56% to 43.80% in 69-bus. In addition, the improvement in minimum bus voltage and other performance indices are also significant.","PeriodicalId":47541,"journal":{"name":"Jurnal Teknologi-Sciences & Engineering","volume":" ","pages":""},"PeriodicalIF":0.7,"publicationDate":"2023-04-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"43405126","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-04-19DOI: 10.11113/jurnalteknologi.v85.18744
Nurul Atikah Abbas, Mohd Ridzuan Ahmad
Due to the obvious advancement of artificial intelligence, keyword spotting has become a fast-growing technology that was first launched a few years ago by hidden Markov models. Keyword spotting is the technique of finding terms that have been pre-programmed into a machine learning model. However, because the keyword spotting system model will be installed on a small and resource-constrained device, it must be minimal in size. It is difficult to maintain accuracy and performance when minimizing the model size. We suggested in this paper to develop a TinyML model that responds to voice commands by detecting words that are utilized in a cascade architecture to start or control a program. The keyword detection machine learning model was built, trained, and tested using the edge impulse development platform. The technique follows the model-building workflow, which includes data collection, preprocessing, training, testing, and deployment. 'On,' 'Off,' noise, and unknown databases were obtained from the Google speech command database V1 and applied for training and testing. The MFCC was used to extract features and CNN was used to generate the model, which was then optimized and deployed on the microcontroller. The model's evaluation represents an accuracy of 84.51% based on the datasets. Finally, the KWS was successfully implemented and assessed on Arduino Nano 33 BLE Sense for two studies in terms of accuracy at three different times and by six different persons.
由于人工智能的明显进步，关键词识别已经成为一项快速发展的技术，几年前由隐马尔可夫模型首次推出。关键字识别是一种查找已经预先编程到机器学习模型中的术语的技术。但是，由于关键字定位系统模型将安装在小型且资源受限的设备上，因此它的尺寸必须最小。当最小化模型尺寸时，很难保持精度和性能。我们在论文中建议开发一个TinyML模型，该模型通过检测在级联架构中用于启动或控制程序的单词来响应语音命令。利用边缘脉冲开发平台建立关键字检测机器学习模型，进行训练和测试。该技术遵循模型构建工作流，其中包括数据收集、预处理、训练、测试和部署。从谷歌语音命令库V1中获取“On”、“Off”、“noise”和未知数据库，应用于训练和测试。使用MFCC提取特征，使用CNN生成模型，然后对模型进行优化并部署在微控制器上。基于数据集，该模型的评估准确率为84.51%。最后，KWS在Arduino Nano 33 BLE Sense上成功实现，并在三个不同的时间，由六个不同的人进行了两次准确性研究。
{"title":"KEYWORD SPOTTING SYSTEM WITH NANO 33 BLE SENSE USING EMBEDDED MACHINE LEARNING APPROACH","authors":"Nurul Atikah Abbas, Mohd Ridzuan Ahmad","doi":"10.11113/jurnalteknologi.v85.18744","DOIUrl":"https://doi.org/10.11113/jurnalteknologi.v85.18744","url":null,"abstract":"\u0000\u0000\u0000\u0000Due to the obvious advancement of artificial intelligence, keyword spotting has become a fast-growing technology that was first launched a few years ago by hidden Markov models. Keyword spotting is the technique of finding terms that have been pre-programmed into a machine learning model. However, because the keyword spotting system model will be installed on a small and resource-constrained device, it must be minimal in size. It is difficult to maintain accuracy and performance when minimizing the model size. We suggested in this paper to develop a TinyML model that responds to voice commands by detecting words that are utilized in a cascade architecture to start or control a program. The keyword detection machine learning model was built, trained, and tested using the edge impulse development platform. The technique follows the model-building workflow, which includes data collection, preprocessing, training, testing, and deployment. 'On,' 'Off,' noise, and unknown databases were obtained from the Google speech command database V1 and applied for training and testing. The MFCC was used to extract features and CNN was used to generate the model, which was then optimized and deployed on the microcontroller. The model's evaluation represents an accuracy of 84.51% based on the datasets. Finally, the KWS was successfully implemented and assessed on Arduino Nano 33 BLE Sense for two studies in terms of accuracy at three different times and by six different persons.\u0000\u0000\u0000\u0000","PeriodicalId":47541,"journal":{"name":"Jurnal Teknologi-Sciences & Engineering","volume":" ","pages":""},"PeriodicalIF":0.7,"publicationDate":"2023-04-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"45836353","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：