Yolanda, Jessica and Dr.Agr.Sc. Dimas Firmanda Al-Riza,, ST., M.Sc. and Im’roah Ikarini,, STP., MP. (2022) Pengembangan Model Prediksi Karakteristik Fisikokimia Jeruk Keprok (Citrus reticulata Blanco) Varietas Batu 55 Berbasis Spektroskopi Reflektansi-Fluoresensi dan Teknik Kemometrika. Sarjana thesis, Universitas Brawijaya.
Abstract
Jeruk keprok (Citrus reticulata Blanco) varietas Batu 55 adalah salah satu varietas jeruk unggul yang dikembangkan di Balai Penelitian Tanaman Jeruk dan Buah Subtropika (Balitjestro). Buah ini tergolong dalam kelompok mandarin, memiliki bentuk buah yang bulat dan salah satu keunggulannya yaitu rasanya manis menyegarkan. Produksi komoditi buah jeruk keprok selalu mengalami peningkatan, pada tahun 2019 tercatat sebanyak 2.44 juta ton lalu mengalami peningkatan produksi pada tahun 2020 yaitu tercatat sebanyak 2.59 juta ton. Besarnya peluang pasar bagi komiditi jeruk keprok Batu 55 tidak sebanding dengan proses penyortirannya seperti menggunakan mesin sortasi berdasarkan ukuran diameter buah dan penyortiran secara manual yang membutuhkan waktu lama serta merusak buah jeruk dalam penilaiannya. Berdasarkan urgensi permasalahan tersebut, solusi yang ditawarkan yaitu prediksi tingkat kematangan secara cepat, mudah, dan tidak merusak buah menggunakan teknik kemometrika berupa regresi Partial Least Square (PLSR) berbasis spektroskopi reflektansi dan fluoresensi. Penelitian ini bertujuan untuk membangun model prediksi karakteristik fisikokimia jeruk berbasis metode spektroskopi reflektansi dan fluoresensi dengan pendekatan teknik kemometrika serta menganalisis karakteristik fisikokimia buah jeruk keprok (C. reticulata Blanco) varietas Batu 55 berdasarkan tekstur buah (firmness), total padatan terlarut (brix), total asam (acidity), dan rendemen (yield) minyak atsiri sesuai SNI 3165:2009. Metode penelitian ini diawali dengan tahap pengambilan 120 sampel jeruk keprok Batu 55 yang terdiri dari tiga warna berbeda berdasarkan tingkat kematangan (hijau, hijau kekuningan, dan oranye), pengambilan data spektroskopi reflektansi dan fluoresensi menggunakan spektrofotometer Vernier, dan pengukuran tekstur buah (firmness), total padatan terlarut (brix), total asam (acidity), serta pengujian minyak atsiri kulit jeruk keprok Batu 55 sebagai penelitian pendukung. Pengolahan data spektra diawali dengan pre-processing yang dilakukan dengan dua tahap untuk spektroskopi reflektansi yaitu smoothing moving average (MA) dan Savitzky-Golay (SG); pre-processing MSC (Multiplicative Scatter Correction) dan SNV (Standard Normal Variate), sedangkan spektroskopi fluoresensi dan kombinasi kedua spektroskopi hanya menggunakan proses smoothing. Data hasil pre-processing dimodelkan dengan regresi Partial Least Square (PLSR) untuk menentukan hubungan parameter brix-acid ratio dalam memprediksikan tingkat kematangan sampel. Pemodelan PLSR terbaik menghasilkan nilai koefisien korelasi (R2) dan nilai galat (RMSE) spektroskopi reflektansi (R2 = 0.8885; RMSE = 2.7025), fluoresensi (R2 = 0.8796; RMSE = 2.8087), dan kombinasi dari spektroskopi reflektansi dan fluoresensi (R2 = 0.9080; RMSE = 2.4555) berdasarkan metode regresi Partial Least Square (PLSR). Parameter kualitas internal dan eksternal sampel jeruk keprok Batu 55 saling memiliki korelasi satu sama lain. Hasil pengujian minyak atsiri kulit jeruk keprok Batu 55 telah dianalisis menggunakan ANOVA dan menunjukkan bahwa hubungan tingkat kematangan yang berbeda yaitu mentah (±189 hari setelah bunga mekar), setengah matang (±224 hari setelah bunga mekar), dan matang (±259 hari setelah bunga mekar) tidak berbeda nyata (p-value >0.05) terhadap nilai rendemen (yield) minyak atsiri, namun berbeda nyata (p-value <0.05) terhadap aktivitas antioksidannya dengan kemampuan aktivitas antibakteri berturut-turut yakni sangat aktif, aktif parsial, dan sangat aktif.
English Abstract
Mandarin orange (Citrus reticulata Blanco cv. Batu 55) is one of the superior citrus varieties developed at the Citrus and Subtropical Fruit Research Institute (Balitjestro). This fruit belongs to the mandarin group, has a round fruit shape and one of its advantages is that it tastes sweet and refreshing. Production of the mandarin orange commodity has always increased, in 2019 it was recorded at 2.44 million tons, then production increased in 2020, to 2.59 million tons. The large market opportunity for the Batu 55 mandarin orange commodity is not comparable to the sorting process such as using a sorting machine based on the size of the fruit diameter and sorting manually which takes a long time and damages the citrus fruit in its assessment. Based on the urgency of the problem, the solution offered is to predict the level of ripeness quickly, easily, and not damage the fruit using a chemometric technique in the form of Partial Least Square regression (PLSR) based on reflectance and fluorescence spectroscopy. This study aims to build a prediction model for the physicochemical characteristics of oranges based on reflectance and fluorescence spectroscopy methods with a chemometric approach and to analyze the physicochemical characteristics of Mandarin orange (C. reticulata Blanco cv. Batu 55) based on fruit texture (firmness), total dissolved solids (brix), total acidity, and yield of essential oils according to SNI 3165:2009. This research method begins with the stage of taking 120 samples of Mandarin oranges Batu 55 consisting of three different colors based on the level of ripeness (green, yellowish green, and orange), collecting reflectance and fluorescence spectroscopy data using a Vernier spectrophotometer, and measuring fruit texture (firmness), total dissolved solids (brix), total acid (acidity), and testing of the essential oil of the Mandarin orange Batu 55 peel as supporting research. Spectral data processing begins with pre-processing which is carried out in two stages for reflectance spectroscopy, namely smoothing moving average (MA) and Savitzky-Golay (SG); MSC (Multiplicative Scatter Correction) and SNV (Standard Normal Variate) pre-processing, while fluorescence spectroscopy and the combination of the two spectroscopy only use the smoothing process. Data from pre-processing results were modeled using Partial Least Square (PLSR) regression to determine the relationship between the brix-acid ratio parameter in predicting the maturity level of the sample. The best PLSR modeling produces correlation coefficient (R2) and error values (RMSE) of reflectance spectroscopy (R2 = 0.8885; RMSE = 2.7025), fluorescence (R2 = 0.8796; RMSE = 2.8087), and a combination of reflectance and fluorescence spectroscopy (R2 = 0.9080 ; RMSE = 2.4555) based on the best model of the Partial Least Square (PLSR) regression method. The internal and external quality parameters of the Mandarin orange Batu 55 samples have a correlation with each other. As supporting data, the results of testing the essential oil of the Mandarin orange Batu 55 peel were analyzed using ANOVA and showed that correlation of the different maturity levels were unripe (±189 days after the flowers bloom), half-ripe (±224 days after the flowers bloom), and ripe (±259 days after the flowers bloom). bloom) was not significantly different (p-value >0.05) from the yield value of essential oils, but significantly different (p-value <0.05) from its antioxidant activity with antibacterial activity, namely very active, partially active, and very active.
| Item Type: | Thesis (Sarjana) |
|---|---|
| Identification Number: | 0522100378 |
| Uncontrolled Keywords: | Fluoresensi, PLSR, Reflektansi, Spektroskopi,Fluorescence, PLSR, Reflectance, Spectroscopy |
| Subjects: | 600 Technology (Applied sciences) > 630 Agriculture and related technologies |
| Divisions: | Fakultas Teknologi Pertanian > Keteknikan Pertanian |
| Depositing User: | Sugeng Moelyono |
| Date Deposited: | 30 Jan 2023 02:17 |
| Last Modified: | 30 Jan 2023 02:17 |
| URI: | http://repository.ub.ac.id/id/eprint/197061 |
![[thumbnail of DALAM MASA EMBARGO]](http://repository.ub.ac.id/style/images/fileicons/text.png) |
Text (DALAM MASA EMBARGO)
Jessica Yolanda.pdf Restricted to Registered users only until 31 December 2024. Download (10MB) |
Actions (login required)
 |
View Item |