Effects of air parameters on sewage sludge drying characteristics and regression analyses of drying model coefficients. (5th November 2021)

• Record Type: Journal Article
• Authors: Zheng, Qiushuang; Hu, Ziyuan; Li, Penggang; Ni, Long; Huang, Guanying; Yao, Yang; Zhou, Liangyu
• Is Part Of: Applied thermal engineering. Volume 198(2021)
• Publication Date: 2021-11-05
• Subjects: Sludge -- Heat pump dryer -- Relative humidity -- Regression analyses; Heat engineering -- Periodicals; Heating -- Equipment and supplies -- Periodicals; Periodicals; 621.40205
• DOI: 10.1016/j.applthermaleng.2021.117501
• Languages: English
• ISSNs: 1359-4311

Abstract

Highlights: The range of air parameters discussed was suitable for sludge heat pump dryer. A quantitative study on the effect of relative humidity on sludge drying was given. Sensitivity of drying time to air was: temperature > air velocity > relative humidity. Relative humidity was introduced into drying model coefficients. Support vector machine improved the prediction ability of drying model. Abstract: Heat pump drying is an energy-saving method for drying sludge. However, the existing research on sludge drying characteristics is not consistent with the operating temperature of a heat pump. In particular, the effect of relative humidity is often neglected. To comprehensively investigate the effects of air parameters, sludge drying characteristics were investigated under the conditions of air temperature 40–70 °C, air velocity 1.0–2.0 m/s and relative humidity 20–40%. The results showed that the effect of relative humidity on the sludge drying characteristics cannot be neglected under low-temperature conditions. Specifically, the drying time was extended by 10–60 min for every 10% increase in relative humidity. Moreover, the sensitivity indicators of temperature, air velocity, and relative humidity on the drying time were 0.470, 0.232 and 0.025, respectively. Furthermore, the relative humidity was introduced into the drying model coefficients to improve prediction ability. Under the verification conditions, the prediction deviation of the drying time of the multiple linear regression was 14.1%, while the least square support vector machine reduced the deviation to 5.5%.