This study aimed to determine the optimum multivariate model for monitoring the soluble solids content (SSC) of sugarcane billets being transferred on a conveyor. The study covered two main issues: the exploration of an appropriate spectral range (450–900 nm versus 700–900 nm) and the assessment of the influence of different levels of cane billets on an elevator via modelling to predict the SSC values. Partial least squares regression (PLSR) was used for model development. Modelling using the range of 450–900 nm employed 4 latent variables (LVs) and showed the coefficient of determination (R²) and root mean squares error of prediction (RMSEP) of 0.83 and 0.29 °Brix, respectively. This caused the model established using the range of 700–900 nm, employed 3 LVs and provided the R² and RMSEP values of 0.81 and 0.31 °Brix, respectively, seems more appropriate. In case of assessing the different cane levels on the conveyor, the outcomes presented model performance of the full and half cane levels in predicting half and full cane datasets with R² and RMSEP of 0.52 and 0.55 °Brix and 0.53 and 0.48 °Brix, respectively. This showed that the different levels affected the SSC predictive accuracy of the model. The combined model was developed to cover variations of this difference and was used to predict two external sets. The predictions of ninety and thirty samples that were collected from the same and different growing seasons as the samples for the modelling presented the R², RMSEP and RPD of 0.70, 0.42 °Brix and 1.83 and 0.56, 0.42 °Brix and 2.00, respectively.

这项研究旨在确定用于监测在传送带上转移的甘蔗坯料的可溶性固形物含量（SSC）的最佳多元模型。该研究涵盖两个主要问题：探索合适的光谱范围（450-900 nm与700-900 nm），以及通过建模以预测SSC值来评估不同水平的甘蔗坯料对电梯的影响。偏最小二乘回归（PLSR）用于模型开发。使用450–900 nm范围进行建模时，使用了4个潜在变量（LVs），并显示出确定系数（R ²）和预测均方根误差（RMSEP）分别为0.83和0.29°Brix。这导致使用700-900 nm范围建立的模型，使用了3个LV，并提供了R ²RMSEP值分别为0.81和0.31°Brix似乎更合适。在输送机上评估不同甘蔗水平的情况下，结果呈现充分和半甘蔗水平的模型性能预测被R半桥和全甘蔗数据集²和0.52 RMSEP和0.55°白利糖度和0.53和0.48°白利糖度，分别。这表明不同的水平会影响模型的SSC预测准确性。开发了组合模型以涵盖此差异的变化，并用于预测两个外部集。分别从同一生长季节和不同生长季节收集的九十个和三十个样品的预测结果表明，R ²，RMSEP和RPD分别为0.70、0.42°Brix和1.83和0.56、0.42°Brix和2.00。