energy required to agitate the pulp, measurements were corrected by subtracting the no-load power. A Guided Radar Laser (GRL) was utilized to measure the position of the surface (Lewis, 2007; Morrison, 2017). Using a Fast Fourier Transform (FFT), the variation of surface position over time was converted into an interface oscillation parameter. The oscillation parameter corresponded to the largest amplitude in the frequency spectrum generated by the FFT (Morrison, 2023). The amplitude produced by the FFT does not represent physical displacement directly. It is the result of a mathematical transformation of the time domain signal into the frequency domain. FFT magnitudes are useful for studying relative changes in signal strength across frequencies, but they do not correspond to actual physical dimensions. 3. RESULTS AND DISCUSSION The experimental data were analyzed using statistical regression modelling to evaluate the effects of the studied variables on the critical impeller speed (Njs), power draw (Pjs), and interface oscillations (Θjs). The Means of Means method was applied to identify key trends, which were interpreted in relation to the likely underlying mechanisms. 3.1 Effect of baffling and airflow rate on the critical impeller speed (Njs) A statistical regression model was developed to assess the influence of the variables on Njs. The derived model is shown in Equation 1. The model achieved an adjusted R2 of 97%, as summarized in Table 2. Njs = 465.9 + 689.9 Jg - 102.8 Baffling - 203.0 Jg2 - 180.3 Baffling*Jg (1) Table 2 – Metrics of the Njs regression model Indicator Value R2 98.3 R2 adjusted 97.0 Standard Error (RPM) 36.1 Observations 10 Table 3 presents the p-value of the Njs model terms. Terms with p<0.05 (95% confidence) were considered statistically significant. Table 3 – P-values for Njs model terms Term P-Value Constant < 0.001 Jg < 0.001 Baffling 0.054 Jg2 0.008 Jg*Baffling 0.010
RkJQdWJsaXNoZXIy MTM0Mzk2