Kalibrasi Analitik Dan Estimasi Ketidakpastian Pengukuran Pada Monitoring Solar Energy System Dengan Fuzzy Regression Analysis

Saladin, Muhammad Nimran (2020) Kalibrasi Analitik Dan Estimasi Ketidakpastian Pengukuran Pada Monitoring Solar Energy System Dengan Fuzzy Regression Analysis. Undergraduate thesis, Institut Teknologi Sepuluh Nopember.

[img] Text
02311640000008-Undergraduate_Thesis.pdf
Restricted to Repository staff only

Download (2MB) | Request a copy

Abstract

Kegiatan kalibrasi menghasilkan nilai ketidakpastian dari pengukuran yang telah dilakukan. Salah satu sumbernya yaitu dari model regresi. Penelitian ini dilakukan untuk melakukan kegiatan kalibrasi dengan menggunakan konsep analisis regresi fuzzy pada monitoring solar energy system. Pemodelan regresi fuzzy dilakukan dengan metode linear programming (LP) dan quadratic programming (QP). Ketidakpastian model regresi fuzzy direpresentasikan dengan nilai h-level himpunan data Y. Ketidakpastian dari model regresi linier juga dihitung dan dibandingkan dengan model regresi fuzzy. Secara kasar, dengan tingkat kepercayaan 99% pada model regresi linier untuk mendekati model regresi fuzzy yang mencakup seluruh sampel data, diperoleh hasil perbandingan nilai ketidakpastian diperluas yang bervariasi dari masing-masing sumber pengukuran. Nilai ketidakpastian diperluas terendah diperoleh dengan komponen ketidakpastian model regresi fuzzy LP yaitu sebesar 0,09915 V, 0,23381 A, dan 0,05303 V masing-masing dari sumber tegangan dan arus photovoltaic serta tegangan baterai. Hingga saat ini masih belum ada standar yang dapat digunakan untuk menggabungkan ketidakpastian model regresi fuzzy dengan sumber-sumber ketidakpastian lainnya. ============================================================ Calibration activities produce uncertainty values from measurements that have been made. One of them is from the regression model. This research was to conduct calibration activities using the concept of fuzzy regression analysis on solar energy system monitoring. Fuzzy regression modeling is done by linear programming (LP) and quadratic programming (QP) methods. The uncertainty of the fuzzy regression model is represented by the h-level value of the Y data set. The uncertainty of the linear regression model is also calculated and compared with the fuzzy regression model. Roughly speaking, with a 99% confidence level in the linear regression model to approach the fuzzy regression model that includes the entire sample of data, obtained a varied comparison of the extended uncertainty values from each source of measurement. The lowest expanded uncertainty value is obtained by the uncertainty component of the LP fuzzy regression model that is equal to 0.09915 V, 0.23381 A, and 0.05303 V respectively from the voltage and photovoltaic current sources and battery voltage. Until now there is no standard that can be used to combine the uncertainty of the fuzzy regression model with other sources of uncertainty.

Item Type: Thesis (Undergraduate)
Uncontrolled Keywords: kalibrasi, ketidakpastian pengukuran, calibration, measurement uncertainty, fuzzy regression analysis, linear programming, quadratic programming
Subjects: Q Science > QA Mathematics > QA278.2 Regression Analysis
Q Science > QA Mathematics > QA39.3 Fuzzy mathematics
Q Science > QA Mathematics > QA248_Fuzzy Sets
T Technology > T Technology (General) > T57.6 Operations research--Mathematics. Goal programming
T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK1056 Solar powerplants
Divisions: Faculty of Industrial Technology and Systems Engineering (INDSYS) > Physics Engineering > 30201-(S1) Undergraduate Thesis
Depositing User: Muhammad Nimran Saladin
Date Deposited: 03 Aug 2020 07:48
Last Modified: 03 Aug 2020 11:01
URI: http://repository.its.ac.id/id/eprint/76866

Actions (login required)

View Item View Item