Hasil Pencarian  ::  Simpan CSV :: Kembali

"Optimasi multi-objektif adalah salah satu alat untuk mengambil keputusan yang optimal dimana terjadi trade-off antara dua atau lebih objektif yang saling bertentangan. Pada studi ini optimasi multi-objektif dijalankan untuk sistem multi generasi panas bumi. Metode yang dipakai adalah artificial neural network-genetic algorithm (ANN-GA) dan genetic algorithm (GA), dimana keduanya nanti akan diperbandingkan. Digunakan software Engineering Equation Solver (EES) dan MATLAB. Batasan (constraint) yang dipakai adalah konsentrasi ammonia (YB), temperature geotermal (TGEO) dan mass extraction ratio (MER). Nilai optimal terbaik dari optimasi multi-objektif metode ANN-GA adalah exergy destruction 3955.51 kW, sum unit cost of the product (SUCP) 97.84 $/GJ dan exergoenvironmental 724.92 mPt/s, nilai optimal ANN-GA tersebut dicapai pada YB 0.415, TGEO 130.02oC dan MER 0.399. Sedangkan nilai optimal terbaik dari optimasi multi-objektif metode GA adalah exergy destruction 3522.59 kW, SUCP 93.86 $/GJ dan exergoenvironmental 813.29 mPt/s, nilai optimal ini didapat pada YB 0. 0.477, TGEO 159.79oC dan MER 0.203. Analisa life cycle analysis (LCA) yang ada dalam studi ini dari software SIMAPRO menunjukkan dampak lingkungan dari steel adalah 156 mPt/kg, steel low alloy 247 mPt/kg dan cast iron 227 mPt/kg.

---

Multi-objective optimization is a tool for making optimal decisions where there is a trade-off between two or more conflicting objectives. In this study, multi-objective optimization is carried out for the geothermal multi-generation system. The methods used are artificial neural network-genetic algorithm (ANN-GA) and genetic algorithm (GA), both of which will be compared later. Engineering Equation Solver (EES) and MATLAB software are used. The constraints used are ammonia concentration (YB), geothermal temperature (TGEO) and mass extraction ratio (MER). The best optimal values from the multi-objective optimization of the ANN-GA method are exergy destruction 3955.51 kW, sum unit cost of the product (SUCP) 97.84 $/GJ and exergoenvironmental 724.92 mPt/s, the optimal values for ANN-GA were achieved at YB 0.415, TGEO 130.02oC and MER 0.399. While the best optimal value of the multi-objective optimization GA method is exergy destruction 3522.59 kW, SUCP 93.86 $/GJ and exergoenvironmental 813.29 mPt/s, this optimal value is obtained at YB 0. 0.477, TGEO 159.79oC and MER 0.203. Life cycle analysis (LCA) that is done in this study from SIMAPRO software showed the environmental impact from steel is 156 mPt/kg, steel low alloy is 247 mPt/kg and cast iron is 227 mPt/kg."

"This book describes hybrid intelligent systems using type-2 fuzzy logic and modular neural networks for pattern recognition applications. Hybrid intelligent systems combine several intelligent computing paradigms, including fuzzy logic, neural networks, and bio-inspired optimization algorithms, which can be used to produce powerful pattern recognition systems. Type-2 fuzzy logic is an extension of traditional type-1 fuzzy logic that enables managing higher levels of uncertainty in complex real world problems, which are of particular importance in the area of pattern recognition. The book is organized in three main parts, each containing a group of chapters built around a similar subject. The first part consists of chapters with the main theme of theory and design algorithms, which are basically chapters that propose new models and concepts, which are the basis for achieving intelligent pattern recognition. The second part contains chapters with the main theme of using type-2 fuzzy models and modular neural networks with the aim of designing intelligent systems for complex pattern recognition problems, including iris, ear, face and voice recognition. The third part contains chapters with the theme of evolutionary optimization of type-2 fuzzy systems and modular neural networks in the area of intelligent pattern recognition, which includes the application of genetic algorithms for obtaining optimal type-2 fuzzy integration systems and ideal neural network architectures for solving problems in this area."

"Pengendalian pesawat terbang merupakan suatu tahap terpenting dalam pengembangan teknologi aviasi yang hanya dapat dilakukan jika memiliki data penerbangan dan model pesawat. Pengambilan data penerbangan dilakukan menggunakan simulator penerbangan ultra-realistis, X-Plane. Algoritma Neural Networks dipilih sebagai metode untuk memodelkan dan mengidentifikasi sistem pesawat terbang juga sebagai pengendali sistem tersebut yang akan terbentuk dalam sebuah kesatuan Direct Inverse Control. Pengujian dan pembelajaran open loop pada sistem Direct Inverse Control dilakukan untuk mengetahui keandalan sistem kendali yang dirancang. Batasan pada penelitian ini adalah kondisi cruising ideal dimana merupakan kondisi terbang pesawat yang memakan hampir 90% dari total penerbangan. Dari hasil pengujian dapat diketahui bahwa data yang dihasilkan simulator sesuai dengan dinamika pergerakan pesawat terbang pada kondisi cruising dan sistem kendali yang dibuat memiliki keandalan yang baik.

---

Flight control is the most important stage in the development of aviation technology which can only be done if flight data and aircraft models have been acquired. Flight data acquisition is carried out using an ultra-realistic flight simulator, X-Plane. Neural Networks algorithm is chosen as a method for modeling and identifying aircraft systems as well as controlling the system which will be formed in a Direct Inverse Control unit. Open loop testing and learning in the Direct Inverse Control system is carried out to determine the reliability of the designed control system. The limit of this study is in the ideal cruising conditions which consume almost 90% of total flights time. From the test results, it can be seen that the data generated by the simulator is in accordance with the dynamics of aircraft movements in cruising conditions and the designed control system has good reliability."

"Jaringan Syaraf Tiruan memiliki kemampuan yang sangat baik dalam mengenali suatu pola (pattern recognition). Cara kerja JST dalam mengenali pola memiliki kesamaan dengan cara kerja otak manusia. Salah satu metode yang termasuk ke dalam JST adalah metode perambatan balik. Dengan kemampuannya dalam mengenali pola ini diharapkan metode JST - Perambatan Balik dapat memodelkan sistem Pressure Process Rig serta Unmanned Aerial Vehicle yang datanya digunakan pada Skripsi ini dan dapat merancang pengendali untuk sistem tersebut. Kode untuk algoritma perambatan balik pada Skripsi ini dikembangkan dengan menggunakan perangkat lunak MATLAB.

---

Artificial Neural Networks had a very good ability for pattern recognition. Artificial Neural Networks techniques in recognising pattern had something in common with the works of human brain. One of its method which included in is backpropagation method. With its ability to recognise these patterns, it was expected that artificial neural networks method can model Pressure Process Rig and Unmanned Aerial Vehicle systems which data had been used in this paper and can design controller for that systems. Backpropagation algorithm code in this paper developed using MATLAB software."