Hasil Pencarian  ::  Simpan CSV :: Kembali

"Neural networks and fuzzy systems are model free control design approaches that represent an advantage over classical control when dealing with complicated nonlinear actuator dynamics. Neuro-Fuzzy Control of Industrial Systems with Actuator Nonlinearities brings neural networks and fuzzy logic together with dynamical control systems. Each chapter presents powerful control approaches for the design of intelligent controllers to compensate for actuator nonlinearities such as time delay, friction, deadzone, and backlash that can be found in all industrial motion systems, plus a thorough development, rigorous stability proofs, and simulation examples for each design. In the final chapter, the authors develop a framework to implement intelligent control schemes on actual systems. Rigorous stability proofs are further verified by computer simulations, and appendices contain the computer code needed to build intelligent controllers for real-time applications. Neural networks capture the parallel processing and learning capabilities of biological nervous systems, and fuzzy logic captures the decision-making capabilities of human linguistics and cognitive systems."

"Pengontrol aliran banyak digunakan di berbagai industri, seperti di industri perminyakan untuk mengalirkan minyak dari minyak lepas pantai ke darat atau digunakan untuk distribusi minyak. Pengontrol aliran yang paling banyak digunakan dalam industri adalah pengontrol berbasis PID konvensional yang diimplementasikan menggunakan PLC. PLC banyak digunakan dalam industri karena kekompakannya, memiliki konektivitas standar dan memiliki keandalan yang tinggi. Dalam penelitian ini, pengontrol non-konvensional, yaitu pengontrol Neuro-Fuzzy, diterapkan pada pabrik prototipe yang mengandung air sebagai agen alirannya. Pabrik prototipe terdiri dari tangki air, pompa air, katup gerbang, katup kontrol, flow meter, dan sistem perpipaan. Kontroler Neuro-Fuzzy dalam penelitian ini dirancang berdasarkan algoritma ANFIS, dengan input berupa kesalahan dan perubahan kesalahan dari variabel proses yang diamati, dalam hal ini aliran air pada pipa keluaran pabrik prototipe. Pengontrol dioperasikan di lingkungan MATLAB/SIMULINK pada PC, yang memperoleh informasi laju aliran berasal dari flow meter yang terhubung ke PLC. PLC berkomunikasi dengan pengendali melalui fasilitas OPC. Output dari pengontrol, yang berupa bukaan katup kontrol, akan dikirim ke PLC melalui OPC, oleh karena itu PLC dapat mengontrol bukaan katup sesuai dengan laju aliran air yang diinginkan. Setelah menjalani proses pelatihan, pengendali berbasis ANFIS yang dikembangkan diuji dengan berbagai titik setel debit air untuk mendapatkan informasi kinerjanya. Dari penelitian ini ditemukan bahwa pengontrol berbasis ANFIS adalah pengontrol dengan kinerja yang baik, yang memiliki waktu naik rata-rata 16,88 detik, waktu penyelesaian 30,68 detik, dan dengan overshoot 0% dan 35,65%, dan memiliki relatif kecil kesalahan 2,59%.

---

Flow control is widely used in various industries, such as in the oil industry to flow oil from offshore to onshore oil or used for oil distribution. The most widely used flow controller in the industry is conventional PID-based controller which is implemented using PLC. PLCs are widely used in industry because of their compactness, standard connectivity and high reliability. In this study, a non-conventional controller, the Neuro-Fuzzy controller, is applied to a prototype plant that contains water as its flow agent. The prototype plant consists of a water tank, a water pump, a gate valve, a control valve, a flow meter, and a piping system. The Neuro-Fuzzy controller in this study was designed based on the ANFIS algorithm, with input in the form of errors and error changes of the observed process variables, in this case the flow of water in the prototype factory output pipe. The controller is operated in a MATLAB / SIMULINK environment on a PC, which gets flow rate information from a flow meter connected to the PLC. PLC communicates with controllers through OPC facilities. The output from the controller, which is the control valve opening, will be sent to the PLC via OPC, therefore the PLC can control the valve opening according to the desired flow rate. After undergoing the training process, the ANFIS-based controller that was developed was tested with various water discharge set points to obtain performance information. From this study it was found that ANFIS-based controller is a controller with good performance, which has an average rise time of 16.88 seconds, a completion time of 30.68 seconds, and with 0% and 35.65% overshoot, and has relatively small errors 2.59%."

"Designed to meet the needs of a wide audience without sacrificing mathematical depth and rigor, Adaptive Control Tutorial presents the design, analysis, and application of a wide variety of algorithms that can be used to manage dynamical systems with unknown parameters. Its tutorial-style presentation of the fundamental techniques and algorithms in adaptive control make it suitable as a textbook. Adaptive Control Tutorial is designed to serve the needs of three distinct groups of readers: engineers and students interested in learning how to design, simulate, and implement parameter estimators and adaptive control schemes without having to fully understand the analytical and technical proofs; graduate students who, in addition to attaining the aforementioned objectives, also want to understand the analysis of simple schemes and get an idea of the steps involved in more complex proofs; and advanced students and researchers who want to study and understand the details of long and technical proofs with an eye toward pursuing research in adaptive control or related topics. The authors achieve these multiple objectives by enriching the book with examples demonstrating the design procedures and basic analysis steps and by detailing their proofs in both an appendix and electronically available supplementary material; online examples are also available. A solution manual for instructors can be obtained by contacting the authors."

"Proses sistem kendali adalah proses penting yang terjadi di dunia perindustrian, salah satunya di ranah industri hulu migas. Salah satu instrumen utama pada proses upstream migas adalah separator yang memiliki fungsi untuk memisahkan kandungan fluida minyak mentah yang mengalir melalui pipa menjadi beberapa wujud fase. Pada kenyataanya hampir semua proses pengendalian separator pada fasilitas produksi PT. Pertamina EP masih menggunakan model pengendalian PID konvensional yang harus terus dimonitoring oleh sumber daya manusia selama 24 jam per hari. Oleh karenanya, pada penelitian ini dirancang sebuah metode pengendalian berbasis intelligent system, yaitu simulasi pengendalian Neuro Fuzzy. Metode pengendalian Neuro-Fuzzy ini didesain menggunakan algoritma ANFIS dengan input berupa setpoint, error, dan selisih error dari proses variabel fluida separator, yaitu level (h) fluida. Penelitian dilakukan menggunakan aplikasi Simulink/MATLAB dengan memasukkan fungsi transfer dari model matematis separator lalu melakukan perbandingan dengan melihat grafik respon dan parameter antara model pengendali PID dan ANFIS. Hasil dari penelitian menunjukan bahwa performa pengendali model ANFIS secara rata-rata memiliki overshoot yang jauh lebih baik dari model PID karena selalu mendekati nol dalam tiap kondisi set point serta model ANFIS memiliki nilai error yang lebih baik pada saat set point bernilai 5 dengan perbedaan error 0,712 dari error model pengendali PID.

---

The control system process is an important process that occurs in the industrial world, one of which is in the upstream oil and gas industry. One of the main instruments in the upstream oil and gas process is a separator which has afunction to separate the crude oil fluid content flowing through the pipe into several phases. In fact, almost all separator control processes at PT. Pertamina EP still uses the conventional PID control model which must be continuously monitored by human resources 24 hours per day. Therefore, in this study, a control method based on intelligent systems is based on Neuro Fuzzy control of the level (h) of the fluid. The research was conducted using the Simulink/MATLAB application by entering the transfer function of the separator mathematical model and then making comparisons by looking at the response and parameter charts between the PID and ANFIS controller. The results of the study show that the ANFIS model controller performance on average has a much better overshoot than the PID model because it is always close to zero in each set point condition and the ANFIS model has a better error value when the set point is 5 with an error difference of 0.712. of the PID controller model error."

"Pengendali adaptif pada Pressure Process Rig Feedback 38-714 dengan mekanisme adaptasi, yaitu faktor pelupa lebih kecil dari 1, menunjukkan performa yang baik jika set-point yang diberikan cukup tereksitasi. Pada sistem dengan setpoint kurang tereksitasi, pengendalian adaptif dengan mekanisme adaptasi menghasilkan fenomena Bursting, yaitu fenomena di mana sistem tidak terkontrol akibat gagalnya kerja estimator. Untuk mengatasi hal tersebut dirancanglah suatu algoritma supervisi. Algoritma supervisi ini berfungsi untuk menata kinerja estimator dan sintesa pengendali dan untuk memastikan lup pegnendalian selalu stabil. Pada skripsi ini, algoritma supervisi memantau besaran sinyal rata-rata dan variansi kesalahan prediksi, autokorelasi sinyal kendali, variansi parameter model, dan letak kutub parameter model terestimasi. Besaran - besaran ini dihitung secara rekursif (setiap pencuplikan) dari besaran ? besaran yang dihasilkan oleh pengendali adaptif. Algoritma supervisi ini diaplikasikan pada pengendalian adaptif Pressure Process Rig (Feedback 38-714). Dari uji eksperimen terbukti bahwa pengendalian adaptif dengan supervise memberikan hasil pengendalian yang lebih baik dibandingkan dengan pengendalian adaptif tanpa supervisi.Hal tersebut dapat terlihat dari tidak adanya fenomena bursting yang terjadi pada pengendali adaptif dengan supervisi yang mempunyai mekanisme adaptasi dan set-point kurang tereksitasi.

---

Adaptive controlling on Pressure Process Rig Feedback 38-714 with adaptation mechanism, that has forgetting factor less than 1, shows good performance if the set-point given excite enough. In the system with less excitation, adaptive control with adaptation mechanism results Bursting phenomenon, that is phenomenon where system can?t be controlled anymore because of the estimator failure. Supervision algorithm is designed to cope with that problem. This supervision algorithm rule is organizing estimator?s work and controller design to make sure that control closed-loop always stable.

In this bachelor thesis, supervision algorithm monitors some parameters, there are mean and variance of prediction error signal, autocorrelation of control signal, variance of model parameter, and place of estimated model poles. These parameters are recursive calculated (every sample time) from adaptive control parameter yielded. This supervision algorithm is implemented on Pressure Process Rig (Feedback 38-714) with adaptive control.

From experiment test, it is proved that adaptive control with supervision gives better control result than adaptive control without supervision. It can be seen from no Bursting phenomenon that happened in adaptive control with supervision level that has adaptation mechanism and less excitation set-point."