Zhijian Liu

Exploring the potential relationship between indoor air quality and the concentration of airborne culturable fungi: a combined experimental and neural network modeling study

Indoor airborne culturable fungi exposure has been closely linked to occupants’ health. However, conventional measurement of indoor airborne fungal concentration is complicated and usually requires around one week for fungi incubation in laboratory. To provide an ultra-fast solution, here, for the first time, a knowledge-based machine learning model is developed with the inputs of indoor air quality data for estimating the concentration of indoor airborne culturable fungi. To construct a database for statistical analysis and model training, 249 data groups of air quality indicators (concentration of indoor airborne culturable fungi, indoor/outdoor PM2.5 and PM10 concentrations, indoor temperature, indoor relative humidity, and indoor CO2 concentration) were measured from 85 residential buildings of Baoding (China) during the period of 2016.11.15–2017.03.15. Our results show that artificial neural network (ANN) with one hidden layer has good prediction performances, compared to a support vector machine (SVM). With the tolerance of ± 30%, the prediction accuracy of the ANN model with ten hidden nodes can at highest reach 83.33% in the testing set. Most importantly, we here provide a quick method for estimating the concentration of indoor airborne fungi that can be applied to real-time evaluation.

Extreme learning machine: a new alternative for measuring heat collection rate and heat loss coefficient of water-in-glass evacuated tube solar water heaters

BackgroundHeat collection rate and heat loss coefficient are crucial indicators for the evaluation of in service water-in-glass evacuated tube solar water heaters. However, the direct determination requires complex detection devices and a series of standard experiments, wasting too much time and manpower.FindingsTo address this problem, we previously used artificial neural networks and support vector machine to develop precise knowledge-based models for predicting the heat collection rates and heat loss coefficients of water-in-glass evacuated tube solar water heaters, setting the properties measured by “portable test instruments” as the independent variables. A robust software for determination was also developed. However, in previous results, the prediction accuracy of heat loss coefficients can still be improved compared to those of heat collection rates. Also, in practical applications, even a small reduction in root mean square errors (RMSEs) can sometimes significantly improve the evaluation and business processes.ConclusionsAs a further study, in this short report, we show that using a novel and fast machine learning algorithm—extreme learning machine can generate better predicted results for heat loss coefficient, which reduces the average RMSEs to 0.67 in testing.