CD Skripsi
Sistem Hidung Elektronik Untuk Identifikasi Infeksi Ganoderma Pada Tanaman Kelapa Sawit
Oil palm plant is one of the main commodities for Indonesia. Determination of
diseases that cause productivity decline is important to identify. Rot disease that
causes complete damage to oil palm plants due to fungal infection G. boninense sp
has organic volatile compounds that can be detected using an electronic nose.In this
Study an electronic nose system has been were designed with 6 sensor arrays namely
TGS 2612, TGS 822, TGS 2611, TGS 2610, TGS 813, and TGS 2620 which are
sensitive to certain VOC compounds. The Samples oil palm plants aged 4 months.
The Samples were divided into 4 groups consisting of 15 seedlings. Three groups of
samples were inoculated by Ganoderma on roots at different times in one week
intervals and labeled as samples A, B, C. The other sample of is were classified as D,
without inoculation. some plant samples in the field. The detection process was
carried out on the roots, stems and leaves of oil palm plants. Python program was
used as a data acquisition system in voltage retrieval. The obtained voltage is
processed and further analyzed as a trapezoid area to determine the sensor response
and the level classification using the Principal Component Analysis (PCA) method.
The results of the trapezoidal area showed that TGS 2611 had high response
compared to other sensors in identifying Ganoderma both in the laboratory and in
the field, while analysis using PCA was able to detect attacks on the roots with a
cumulative variance percentage level of 91.37% on Lab experiment and 98, 48% on
the Field. The results of this PCA analysis show that the electronic nose is potential
at distinguishing oil palm plants attacked by Ganoderma with 4 classifications A, B,
C and D.
Keywords: Electronic Nose, TGS, Python, PCA, Palm Oil, Rot
Tidak tersedia versi lain