E-Journal Politeknik Negeri Cilacap
Not a member yet
919 research outputs found
Sort by
Identifikasi Persediaan Makanan di dalam Lemari Pendingin Berbasis Raspberry Pi dan Deep Learning
No full textSistem ini dibuat atas dasar permasalahan yang terjadi dalam kehidupan sehari-hari, salah satunya yaitu tidak terpantaunya persediaan bahan makanan di lemari pendingin. Ketika dibutuhkan suatu bahan makanan dari lemari pendingin dan ternyata tidak ada, maka akan menjadi masalah. Oleh karena itu, dibuatlah sebuah sistem yang mampu mengidentifikasi objek makanan di dalam lemari pendingin. Masukan dari sistem ini berupa foto objek makanan yang diambil menggunakan Raspberry Pi Camera dan terhubung langsung dengan Raspberry Pi di dalam lemari pendingin. Setelah diproses dengan algoritma pembelajaran mesin, maka keluaran yang dihasilkan berupa identifikasi objek makanan yang terdapat di dalam lemari pendingin tersebut. Objek makanan yang diuji berupa pisang, mentimun, brokoli, dan jeruk. Dari hasil pengujian, terlihat bahwa program mengidentifikasi objek dengan benar pada objek pisang dan jeruk yang ditunjukkan dengan confidence level tertinggi sebesar 56,98% dan 45,88%. Identifikasi objek mentimun dikenali sebagai zukini dengan confidence level tertinggi sebesar 78,61%. Adapun identifikasi objek paling rendah terdapat pada objek brokoli dengan confidence level kurang dari 1%
Pemurnian Bioetanol Menggunakan Adsorben Silika Gel dari Limbah Botol Kaca di Industri Kecap
Get PDFWaste glass bottles is an inorganic waste that amounts to 0.7 million tons per year with the main content of silica, so it can be used as the main ingredient to produce silica gel. In the soy sauce industry, glass bottle waste is generally produced from broken glass bottles during depalletizer activity, namely the activity of moving glass bottles to the glass bottle cleaning area. The glass bottle waste is generally only accommodated and has not been used optimally. This research aims to utilize waste glass bottles as a raw material for producing silica gel adsorbents using the hydrothermal method and the sol-gel approach. Silica gel becomes an adsorbent in the purification of bioethanol from wastewater washing dissolving tanks using the adsorption method. Variations in the bioethanol production are yeast weight as 0, 2, 5, and 8 g as well as fermentation time for 4, 7, and 10 days. The bioethanol purification process used variations in adsorption time for 40, 60, and 80 minutes. Characterization of silica gel using BET and SEM-EDX test. Bioethanol levels after going through the adsorption process were analyzed using the GC-MS method. The BET test results show that activated silica gel has a surface area of 231,851 m2/g. Analysis with SEM-EDX showed that activated silica gel particles were in the form of porous lumps with chemical content of Si and O elements of 40.94% and 51.92%, respectively. Based on the results of the GC-MS test, 60 minutes is the best adsorption time to increase the bioethanol content from 39,8 % to 72,0 %
Studi Kualitas Air Kolam Ikan Air Tawar di Balai Benih Ikan Sentral Masni, Kabupaten Manokwari, Provinsi Papua Barat
Get PDFWater quality parameters in pond include ammonia, nitrite, nitrate, phosphate, BOD, DO, turbidity, temperature, and pH. It is important to know the concentration of each component to properly minimize the negative impacts of their disturbances for the overall water quality and the biota in it. This study aims to determine the concentration of ammonia, nitrite, nitrate, phosphate, BOD, DO, turbidity, temperature, and pH in freshwater fish ponds at the Central Fish Seed Office (BBIS) of Masni, Manokwari Regency, West Papua Province. The study was conducted in May 2017, representing the rainy season, and August 2017, representing the dry season. Water samples were taken from settling, rearing, hatchery, and brood ponds. Meanwhile, standard method was implemented in the component measurements. The study results show that the concentrations of ammonia, nitrite, nitrate, and phosphate were 0.02-0.18 mg/L, 0.00-2.00 mg/L, 0.40-1.00 mg/L, and 1.01-21.83 mg/L, respectively. Furthermore, the concentrations of BOD and DO were 6.12-8.19 mg/L and 7.30-12.00 mg/L, respectively. Finally, the values of turbidity, temperature, and pH were 0.40-9.99 NTU, 26-30°C and 7.7-8.5, respectively
Uji Karakteristik Briket Berbahan Baku Bonggol Jagung Berdasarkan Variasi Jumlah Perekat
Get PDFCharcoal briquettes are a renewable form of energy from biomass. This briquette is an alternative to fossil fuels. In this study, the process of analyzing the effect of the amount of adhesive on the water content, index of destruction, ash content, and calorific value of corn cob briquettes was carried out. The use of tapioca flour adhesive concentrations of 0%, 3%, 5%, and 7%. This study used a completely randomized series or RAL for a single factor with ANOVA analysis to determine the effect of the use of various adhesives on corn cob briquettes. The results obtained were, 0% produced briquettes with values of moisture content, ash content, heating value, and destruction index were 0.19%, 0.14%, 0.19% and 5.655 Cal/gr, respectively. For 3% adhesive, the yield of moisture content, ash content, calorific value, and destruction index were 0.21%, 0.33%, 0.28% and 5.398 Cal/gr, respectively. At 5% adhesive concentration the test results of moisture content, ash content, calorific value, and destruction index were 0.22%, 0.35%, 0.34% and 4.431 Cal/gr, respectively. Meanwhile, the adhesive concentration of 7% was 0.31%, 0.89%, 0.38% and 3.382 Cal/gr, respectively. If this result is based on SNI, it can be said that it has met these standards
Rancangan Alat Elektroplating dan Eksperimen Pelapisan Berbahan CuSO4 Terhadap Ketebalan Lapisan
Get PDFThe electroplating method has the aim of producing a surface that has characteristics by the coating metal. The coating process requires a device that can place the ions from the coating material (anode) into the coated metal (cathode) through an electro-deposition process. Parameters such as temperature affect the electroplating process and greatly determine the coating result. The purpose of this study was to design the electroplating device, as well as to test the coating process using CuSO4 anode. The research method was carried out with an experimental approach through the manufacture of electroplating tools and coating testing of time variations of 20, 26, 32, 38, and 44 minutes and temperatures of 60, 65, 70, 75, and 80 oC. The results of the electroplating device design consist of frame components, plating and rinsing tubs, electrical systems, and cathode hanger. The coating results obtained the highest layer thickness 2.890 μm at a temperature of 80 oC for 40 minutes
Deteksi Kadar Alkohol Menggunakan Sensor MQ3 Berbasis Website
Get PDFAccording to the Head of the Indonesian National Police (Kapolri) General Idham Azis, the number of deaths from accidents that occurred in 2019 reached 23,530. A total of 40% of the deaths from traffic accidents are caused by human error and the influence of alcohol. The high number of deaths from traffic accidents due to the influence of alcohol is a concern for all of us. Excessive alcohol consumption is dangerous when driving because consuming alcohol will affect a person\u27s temperament and worsen driving behavior by reducing awareness, leading to accidents. A system that is able to detect the alcohol level of a vehicle driver that can be monitored anywhere is needed, as information to alert motorists to the influence of alcohol in order to prevent traffic accidents online and on a web-based basis. Using the MQ-3 sensor for alcohol detection and the ESP8266 processor in WeMos WiFi. Â The average percentage error in the measurement of the measured sample was found to have an error value of: 3.6%. Research has also succeeded in reading alcohol levels through websites
Komparasi Model Prediksi Kurs Pada Masa Pandemi Covid-19 Menggunakan Neural Network Berbasis Genetic Algorithm dan Particle Swarm Optimization
Get PDFData from Bank Indonesia shows that the rupiah exchange rate against dollar weakened at the beginning of the Covid-19 pandemic. This exchange rate volatility is an important problem in the Indonesian economy. Therefore, the prediction model for the exchange rate against the dollar is needed during the Covid-19 pandemic to predict the exchange rate during the Covid-19 Pandemic. This study is proposed to compare the prediction of the rupiah exchange rate against the dollar using the GA-based Neural Network algorithm and the PSO-based Neural Network algorithm. Initially the data was collected in the period 2019 to 2021, then the data is preprocessed. Validation used the k-fold validation technique with a ratio of 70:30, while the evaluation is carried out with the output of RMSE. The results showed that the performance of PSO and GA was the same, namely 0.020 +/- 0.006
Peramalan Permintaan Pasokan Energi Berdasarkan Intensitas Konsumsi Listrik dan Kapasitas Pembangkit Listrik Terpasang
Get PDFThis research is an important step in energy planning which is the main in electricity planning in the long-term forecast for the demand and supply of energy supply. A simple model is presented using LEAP (Long Term Alternative Energy Planning System) as a broadcasting tool and Central Java as a case study case for this research. This research discusses future energy needs. Electricity needs in the household, industrial, business, and community sectors are calculated based on data on participation, electricity consumption, installed electricity generation capacity, electrification ratio, and electricity strengthening. The base year of this research is 2020 and 2045 is the final period of this forecast. The results of this study indicate that energy needs in Central Java will increase by an average of 23.44% in 2045 or five times compared to 2020 by taking into account various factors of regional economic growth and electricity demand each yea
Optimasi Klasifikasi Parasit Malaria Dengan Metode LVQ, SVM dan Backpropagation
Get PDFThe use of the classification method affects the accuracy of the test results. The accuracy of the classification method is affected by the number of classes in the image. The number of classes and the amount of data should be considered when making decisions in choosing a classification method. This study used 600 data, which were divided into 510 training data and 90 test data. The number of classes tested is 12 classes with the number of initial features used by 22 features. The characteristics used in the test consist of shape characteristics and texture characteristics. The classification methods used in this study are LVQ, Backpropagation, and SVM. The data has 22 features or attributes that are the result of texture and shape feature extraction. Texture features are energy 0o, energy 45o, energy 90o, energy 135o, entropy 0o, entropy 45o, entropy 90o, entropy 135o, contrast 0o, contrast 45o, contrast 90o, contrast 135o, homogeneity 00, homogeneity 45o, homogeneity 90o, homogeneity 135o, correlation 0o, Correlation 45o, correlation 90o, correlation 135o, features of área and perimeter shape. The test results using the Backpropagation method obtained 89.7% results, using the LVQ method obtained 77.78% results, and the SVM method obtained 99.1% results
An Enhanced Dynamic Signature Verification using the X and Y Histogram Features
Get PDFDynamic signature verification by using histogram features is a well-known signature forgery detection technique due to its high performance. However, this technique is often limited to angular histograms derived from vectors containing two adjacent points. We propose additional new features from the X and Y histograms to overcome the limitation. Our experiments indicate that our technique produced Under Curve Area AUC values 0.80 to detect skilled forgery and 0.91 for random forgery. Our method performed best when the verification system uses 12 of the most dominant features. This setup produced AUC values of 0.80 to detect skilled forgery and 0.93 for random forgery. These results outperformed the original technique when the X and Y histogram features are not used that produced AUC values of 0.78 to detect skilled forgery and 0.90 for random forgery