Skripsi
PENERAPAN ALGORITMA AUTO-CALIBRATION PADA SENSOR SUHU MENGGUNAKAN JARINGAN SARAF TIRUAN (JST)
In this study, an automatic calibration system was designed for the BME280 sensor as a temperature gauge. Artificial Neural Network (ANN) is selected as the automatic calibration method that runs on Matlab software. With the BME 280 sensor producing untrained data (inputs) and as reference data used by elitech digital temperature sensors, measurements are performed simultaneously between the BME 280 sensor and the elitech sensor for 5 days to understand the character of the BME 280 sensor. ANN model selected during network training is trainCGB using 3 hidden layers, minimum gradient 10-7 and maximum validation of 7, then obtained the best validation performance with an average error value (Mean-Square Error) only around 0.001264 and regression coefficient (R) of 0.98924 shows excellent network performance. Furthermore, to demonstrate network consistency, testing was conducted with a new amount of data. The test results showed that the ANN model used was able to calibrate new enter data over time. That means this calibration system can be used at any time in a certain period, whether daily, monthly or yearly. The next step is to implement the best weight changes obtained in network training to be applied to the microcontroller of the temperature monitoring system with BME 280 sensor.
Inventory Code | Barcode | Call Number | Location | Status |
---|---|---|---|---|
2107003866 | T55063 | T550632021 | Central Library (Referens) | Available but not for loan - Not for Loan |
No other version available