Hasil Pencarian  ::  Simpan CSV :: Kembali

"In this research, a layered-recurrent artificial neural network (ANN) using the back-propagation method was developed for simulation of a fixed-bed industrial catalytic reforming unit called Platformer. Ninety-seven data points were gathered from the industrial catalytic naphtha reforming plant during the complete life cycle of the catalytic bed (about 919 days). Ultimately, 80% of them were selected as past horizontal data sets, and the others were selected as future horizontal ones. After training, testing, and validating the model with past horizontal data, the developed network was applied to predict the volume flow rate and research octane number (RON) of the future horizontal data versus days on stream. Results show that the developed ANN was capable of predicting the volume flow rate and RON of the gasoline for the future horizontal data sets with AAD% (average absolute deviation) of 0.238% and 0.813%, respectively. Moreover, the AAD% of the predicted octane barrel levels against the actual values was 1.447%, which shows the excellent capability of the model to simulate the behavior of the target catalytic reforming plant."

"Biomassa merupakan salah satu potensi energi alternatif untuk mengurangi ketergantungan penggunaan energi fosil. Indonesia memiliki potensi energi biomassa sebesar 49.810 MW yang berasal dari limbah dan tanaman. Pemanfaatan energi tersebut dapat dilakukan melalui proses gasifikasi yang mengubah biomassa menjadi gas sintetik. Salah satu metode untuk memodelkan proses tersebut adalah dengan menggunakan kecerdasan buatan atau artificial intelligence (AI). Studi literatur yang dilakukan menunjukkan bahwa metode artificial neural network (ANN) adalah pendekatan AI yang sering dipakai untuk melakukan pemodelan proses gasifikasi. Namun, ANN memiliki beberapa kekurangan dalam pemodelan dinamis yang kemudian disempurnakan melalui salah satu pengembangannya yang dinamakan recurrent neural network (RNN) yang mampu memodelkan variabel dependen terhadap waktu. Kesimpulan dari penelitian ini menyarankan agar pengembangan RNN dapat dijadikan acuan untuk membuat sistem kontrol pintar pada prototipe gasifier yang akan datang.

---

Biomass is one of the alternative energy sources to reduce the usage of fossil energy. The potential of biomass energy in Indonesia reaches 49,810 MW, which comes from organic wastes and plants. Gasification is a process to convert biomass to synthetic gas, which is one of the utilizations of biomass energy. Artificial Intelligence (AI) implemented to model the complex process of gasification. Artificial Neural Network (ANN) is a common approach in AI to model the process in the gasifier. Yet, ANN is still inferior in modeling dynamic process that leads to an improvement of ANN called recurrent neural network (RNN). The result of this study suggests that RNN could be the foundation for the development of smart control for the next prototypes."

"Penelitian ini membahas mengenai terjadinya unplanned maintenance pada mesin hydraulic D/E di area finishing mill, Pabrik Hot Strip Mill pada pabrik penghasil baja PT. Krakatau Steel. Unplanned maintenance menyebabkan terganggunya proses produksi. Oleh karena itu, unplanned maintenance akan diubah menjadi planned maintenance. Perancangan planned maintenance akan dilakukan dengan 2 metode, yaitu Artificial Neural Network dan Distribusi Lognormal. Kedua metode ini kemudian akan dibandingkan berdasarkan nilai mean square error (MSE), mean absolute percentage error (MAPE) dan mean absolute deviation (MAD) untuk melihat metode mana yang lebih sesuai untuk kasus ini. Setelah melakukan perbandingan kedua metode, maka diketahui bahwa neural network lebih akurat dibandingkan metode distribusi lognormal karena memiliki nilai error yang lebih kecil.

---

This study discusses the occurrence of unplanned maintenance on hydraulic D / E machines in the area finishing mill, Hot Strip Mill Plant PT. Krakatau Steel steelmaker. Unplanned maintenance led to disruption of the production process. Therefore, unplanned maintenance will be changed to planned maintenance. The design of planned maintenance will be done by 2 methods, namely Artificial Neural Network and lognormal distribution. Both of these methods will then be compared based on the mean square error (MSE), mean absolute percentage error (MAPE) and mean absolute deviation (MAD) to see which method is more appropriate for this case. After doing a comparison of the two methods, it is known that a neural network is more accurate than the lognormal distribution method because it has a smaller error."

"ABSTRAKKondisi cuaca merupakan faktor yang signifikan untuk berbagai sektor seperti keselamatan transportasi, pembangunan, kesehatan dan lain-lain oleh karena itu dibutuhkan akurasi yang tinggi dalam melakukan peramalan keadaan cuaca kedepannya. Banyak cara yang digunakan untuk memprakirakan kondisi cuaca, seiring berkembangnya teknologi, prakiraan Hujan dapat dilakukan dengan menggunakan teknologi Artificial Intelligence (AI) atau kecerdasan buatan sehingga hasil yang diperoleh lebih optimal. Dalam penelitian ini, jaringan saraf tiruan yang digunakan memiliki algoritma feedforward neural network dengan data pelatihan berupa suhu, tekanan udara, kelembaban udara, titik embun, kecepatan angin tiap 3 (tiga) jam di Stasiun pengamatan BMKG di Jawa Timur dari tahun 2019 dengan target adalah intensitas curah hujan. Data pelatihan dilakukan pada periode 1 Januari 2019 sampai 28 Februari 2019 dan selanjutnya, data diuji pada periode 1 sampai 31 Maret 2019. Berdasarkan hasil analisis, model Jaringan Saraf Tiruan memiliki performa yang cukup baik dalam prakiraan intensitas curah hujan di Jawa Timur. Model terbaik ditunjukkan oleh model dengan arsitektur 7-60-1 dengan tingkat korelasi yang dihasilkan sebesar 0,87 dengan nilai error sebesar -0.03 serta akurasi 76 persen dengan lokasi penelitian di Stasiun Meteorologi Bawean. Dengan adanya model ini, diharapkan dapat menjadi salah satu pertimbangan forecaster dalam membuat prakiraan hujan khususnya prakiraan jangka pendek dengan interval tiap 3 (tiga) jam.

---

ABSTRACTWeather conditions are a significant factor for various sectors such as transportation safety, development, health, etc. Therefore, high accuracy is needed in forecasting future weather conditions. Many methods are used to predict weather conditions, as technology develops, Rain forecast can be made using Artificial Intelligence (AI) technology so that the results obtained are more optimal. In this study, the artificial neural network used has a feedforward neural network algorithm with training data in the form of temperature, air pressure, humidity, dew point, wind speed every 3 (three) hours at the BMKG observation station in East Java from 2019 with the target being rainfall intensity. The training data was conducted in the period January 1 2019 to February 28 2019 and subsequently, the data were tested in the period 1 to 31 March 2019. Based on the results of the analysis, the Artificial Neural Network model performed reasonably well in the forecast of rainfall intensity in East Java. The best model is shown by a model with 7-60-1 architecture with a resulting correlation level of 0,87 with an error value of -0.03 and an accuracy of 76 percent with the research location at the Bawean Meteorological Station. With this model, it is expected to become one of the forecaster considerations in making rain forecasts, especially short-term forecasts at intervals of every 3 (three) hours."

"Di Indonesia, dengue telah menjadi salah satu penyakit yang bersifat hiperendemis. Dengue diderita oleh masyarakat dari berbagai kalangan usia, baik pria maupun wanita. Dengue memiliki manifestasi klinis yang terdiri dari tiga fase: fase demam, fase kritis, dan fase penyembuhan. Banyak pasien dengue meninggal pada fase kritis karena pengobatan yang tidak dilaksanakan tepat waktu. Oleh karena itu, dibangunlah model-model yang dapat memprediksi tingkat keparahan dengue berdasarkan hasil uji laboratorium dari pasien yang bersangkutan menggunakan Artificial Neural Network (ANN) dan Analisis Diskriminan (AD). Dalam pembangunan model-model tersebut, digunakan data dengan jumlah yang sangat kecil, yakni sebesar 77 data. Dalam data tersebut, terdapat informasi mengenai hasil uji laboratorium dan diagnosis dari pasien yang bersangkutan. Diagnosis tersebut dikelompokkan ke dalam tiga kategori keparahan dengue, yakni DF sebagai tingkat ringan, DHF grade 1 sebagai tingkat sedang, dan DHF grade 2 sebagai tingkat parah. Dalam penelitian ini, dilakukan tiga pemisahan data, yakni dengan rasio data training : data testing sebesar 70% : 30%, 80% : 20%, and 90% : 10%. Berdasarkan hasil yang diperoleh, model-model prediksi ANN yang dibangun menggunakan fungsi aktivasi logistik dan tangen hiperbolik dengan persentase data training sebesar 70% menghasilkan akurasi (90.91%), sensitivitas (91.11%), dan spesifisitas (95.51%) tertinggi. Model-model tersebutlah yang diajukan dalam penelitian ini. Model-model tersebut akan mampu membantu para dokter dalam memprediksi tingkat keparahan dengue dari pasien yang bersangkutan sebelum memasuki fase kritis. Lebih jauh, model-model tersebut dapat memudahkan para dokter dalam mengobati pasien dengue secara dini, sehingga kasus-kasus fatal atau kematian dapat dihindari.

---

In Indonesia, dengue has become one of the hyperendemic diseases. Dengue is being suffered by many people of all ages, both men and women. Dengue has clinical manifestations that are divided into three phases: febrile phase, critical phase, and convalescence phase. Many patients have died in the critical phase due to the lack of timely treatment. Therefore, I developed models that can predict the severity of dengue based on the corresponding patients’ laboratory test results using Artificial Neural Network (ANN) and Discriminant Analysis (DA). In developing the models, I used a very small dataset, which only consisted of 77 data. The data contains information regarding the laboratory test results and the diagnosis of each of the corresponding patients. The diagnoses were classified into three categories of dengue severity, which are DF as the mild level, DHF grade 1 as the intermediate level, and DHF grade 2 as the severe level. I conducted three different data split, that is, with the ratio of training : testing = 70% : 30%, 80% : 20%, and 90% : 10%. It is shown that ANN models developed using logistic and hyperbolic tangent activation function with 70% training data yielded the highest accuracy (90.91%), sensitivity (91.11%), and specificity (95.51%). These ANN models are the proposed models in this research. The proposed models will be able to help physicians predict the dengue severity of a corresponding patient before entering the critical phase. Furthermore, it will ease physicians in treating dengue patients early, so deaths or fatal cases can be avoided."

"Penelitian ini membahas tentang perancangan sistem identifikasi pada sistem dinamik kapal Makara 03 dengan konfigurasi multi masukan dan multi keluaran. Penelitian ini merancang berbagai metode perombakan struktur Jaringan Saraf Tiruan (JST) baik metode sekuensial maupun fungsional untuk dapat menangkap dinamik yang ada pada dinamik kapal Makara 03. Metode-metode pada JST yang dibuat akan dibandingkan dengan hasil dari model matematika yaitu Transfer Function dan State Space untuk membuktikan keberhasilan dan keunggulan JST dalam membuat sistem identifikasi. Hasil dari perbandingan tersebut membuktikan semua metode yang dihasilkan pada penelitian ini mendapatkan hasil yang lebih baik dibandingkan dengan model matematika konvensional.

---

This research discusses the design of the identification system on the dynamic system of the Makara 03 ship with a multi-input and multi-output configuration. This study designed various structural reshuffle methods for sequensial and functional model of Artificial Neural Network (ANN) to be able to capture the dynamics of Makara 03. The methods in the ANN that were made will be compared with the results of mathematical models namely Transfer Function and State Space for prove the success and superiority of ANN in making identification systems. The results of this comparison prove that all the ANN methods produced in this study get better results compared to conventional mathematical models."

"ABSTRAK

Prostetik merupakan salah satu inovasi di bidang kesehatan yang berfungsi untuk membantu maupun menggantikan salah satu fungsi organ. Prostetik yang ada di Indonesia saat ini masih didominasi oleh produk-produk luar negeri. Hal ini akan membuat harga prostetik menjadi tidak terjangkau oleh pasien. Seiring perkembangan zaman, metode produksi masal yang menggunakan CNC machining atau casting sudah mulai berubah ke batch yang lebih sedikit dengan menggunakan additive manufacturing seperti 3D printing. Penggunaan teknologi 3D Printing memiliki beberapa keunggulan dibandingkan dengan teknologi untuk produksi masal, di antaranya ramah proses kustomisasi, menghemat biaya bahan baku, waktu produksi lebih singkat untuk batch dengan jumlah kecil, dan lebih ramah lingkungan karena menghasilkan sedikit bahan sisa produksi. Penelitian ini dilakukan untuk dapat meningkatkan performa dari prostetik melalui sebuah desain prostetik lokal. Penentuan performa prostetik dilakukan dengan simulasi Finite Element Analysis dengan membandingkan tegangan von mises. Hasil simulasi menggambarkan bahwa desain modifikasi prostetik mampu meningkatkan performanya. Selain itu, dalam penelitian ini juga akan membahas permodelan biaya produksi prostetik antara tiga metode produksi, yaitu CNC Milling, 3D Printer FDM kelas penghobi, dan 3D Printer FDM kelas industri. Dari permodelan tersebut, terdapat dua parameter yang dibandingkan yaitu perbandingan waktu periode profit dalam nilai investasi yang sama dan perbandingan nilai investasi dengan harga jual prostetik yang sama. Hasil permodelan biaya menggambarkan bahwa teknologi 3D Printing mampu menginterupsi teknologi produksi masal CNC machining.

---

ABSTRACT

Prosthetics is one of the innovations in health that serves to help or replace one of the organs' functions. Prosthetics in Indonesia is currently still dominated by foreign products. That will make prosthetic prices unaffordable for patients. Over the times, mass production methods that use CNC machining or casting have begun to change to fewer batches using additive manufacturing, such as 3D printing. 3D printing technology has several advantages compared to mass production technology, including friendly customization processes, saving raw material costs, shorter production time for batches in small quantities, and more environmentally friendly because it produces less material remaining production. This research conducted to improve the performance of prosthetics through a local prosthetic design. The determination of prosthetic performance is done by Finite Element Analysis simulation by comparing von mises stress. Simulation results illustrate that prosthetic modification design can improve performance. Besides, this research will also discuss the modeling of prosthetic production costs between three production methods, namely CNC Milling, hobbyist class FDM 3D Printer, and industrial class 3D Printer. From the modeling, two parameters are being compared namely the comparison of the period of profit in the same investment value and the comparison of investment values with the same prosthetic selling price. The cost modeling results illustrate that 3D Printing technology can interrupt CNC machining mass production technology."

"The two LNAI volumes 7208 and 7209 constitute the proceedings of the 7th International Conference on Hybrid Artificial Intelligent Systems, HAIS 2012, held in Salamanca, Spain, in March 2012. The 118 papers published in these proceedings were carefully reviewed and selected from 293 submissions. They are organized in topical sessions on agents and multi agents systems, HAIS applications, cluster analysis, data mining and knowledge discovery, evolutionary computation, learning algorithms, systems, man, and cybernetics by HAIS workshop, methods of classifier fusion, HAIS for computer security (HAISFCS), data mining: data preparation and analysis, hybrid artificial intelligence systems in management of production systems, hybrid artificial intelligent systems for ordinal regression, hybrid metaheuristics for combinatorial optimization and modelling complex systems, hybrid computational intelligence and lattice computing for image and signal processing and nonstationary models of pattern recognition and classifier combinations."

"The two LNAI volumes 7208 and 7209 constitute the proceedings of the 7th International Conference on Hybrid Artificial Intelligent Systems, HAIS 2012, held in Salamanca, Spain, in March 2012. The 118 papers published in these proceedings were carefully reviewed and selected from 293 submissions. They are organized in topical sessions on agents and multi agents systems, HAIS applications, cluster analysis, data mining and knowledge discovery, evolutionary computation, learning algorithms, systems, man, and cybernetics by HAIS workshop, methods of classifier fusion, HAIS for computer security (HAISFCS), data mining, data preparation and analysis, hybrid artificial intelligence systems in management of production systems, hybrid artificial intelligent systems for ordinal regression, hybrid metaheuristics for combinatorial optimization and modelling complex systems, hybrid computational intelligence and lattice computing for image and signal processing and nonstationary models of pattern recognition and classifier combinations."