TY - GEN
T1 - Soft Sensor Design of Solar Irradiance Using Multiple Linear Regression
AU - Asy'Ari, Muhammad Khamim
AU - Musyafa, Ali
AU - Noriyati, Ronny Dwi
AU - Indriawati, Katherin
N1 - Publisher Copyright:
© 2019 IEEE.
PY - 2019/8
Y1 - 2019/8
N2 - Solar irradiance is power per area received from the sun. Solar irradiance is an important variable that affects the solar panel output. Monitoring of solar irradiance can use a pyranometer sensor. However, the drawbacks of the sensor are expensive and less efficient if that is applied to solar power generation systems requiring a lot of sensors. The problem can be solved by soft sensors which can be designed using multiple linear regression models. The model is developed using training data from solar cell output. The variations given is the type of input to build the model. The model input comprises solar cell output voltage, solar cell output current, and a combination of both. Data testing is done by using multiple correlation tests, multiple linear regression significance tests, and multiple linear regression coefficient significance tests. The best design of the soft sensor is a combination of current and voltage as input models. It has the smallest mean square error value of 7.887 × 10-24 for training data and 3.876 for test data.
AB - Solar irradiance is power per area received from the sun. Solar irradiance is an important variable that affects the solar panel output. Monitoring of solar irradiance can use a pyranometer sensor. However, the drawbacks of the sensor are expensive and less efficient if that is applied to solar power generation systems requiring a lot of sensors. The problem can be solved by soft sensors which can be designed using multiple linear regression models. The model is developed using training data from solar cell output. The variations given is the type of input to build the model. The model input comprises solar cell output voltage, solar cell output current, and a combination of both. Data testing is done by using multiple correlation tests, multiple linear regression significance tests, and multiple linear regression coefficient significance tests. The best design of the soft sensor is a combination of current and voltage as input models. It has the smallest mean square error value of 7.887 × 10-24 for training data and 3.876 for test data.
KW - current
KW - multiple linear regression
KW - soft sensor
KW - solar cell
KW - solar irradiance
KW - voltage
UR - http://www.scopus.com/inward/record.url?scp=85078451604&partnerID=8YFLogxK
U2 - 10.1109/ISITIA.2019.8937150
DO - 10.1109/ISITIA.2019.8937150
M3 - Conference contribution
AN - SCOPUS:85078451604
T3 - Proceedings - 2019 International Seminar on Intelligent Technology and Its Application, ISITIA 2019
SP - 56
EP - 60
BT - Proceedings - 2019 International Seminar on Intelligent Technology and Its Application, ISITIA 2019
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 2019 International Seminar on Intelligent Technology and Its Application, ISITIA 2019
Y2 - 28 August 2019 through 29 August 2019
ER -