With the rapid development of the global economy,the improvement of people’s living standards make people’s demands for happiness higher and higher.In the past,people often use some mechanical tools to adjust the indoor air humidity.However,this adjustment method leading to indoor air pollution because of unventilation,and leading to waste of energy,it does not meet the strategic requirements of sustainable development.In this paper,based on this hot issue,orthogonal test method was used to study the ratio of zeolite concrete,and a large amount of data was accumulated for the subsequent development of the material.For the practical application of the material,especially in the regulating room wet environment has important scientific significance and practical value.The research indicates:(1)The X-ray diffraction analysis of the zeolite powder after the modified treatment showed that the structural components of the zeolite powder did not change,indicating that the natural clinoptilolite had good thermal stability.(2)The results of moisture absorption test of zeolite powder after thermal modification showed that zeolite powder had the best hygroscopicity when the modification condition was 500 ℃ for 2 hours.The optimum moisture absorption rate was 13.60%.The moisture content was 11.73% and the moisture absorption hysteresis was 13.75%.(3)In this experiment,the water-cement ratio,the sand rate and the zeolite powder were used as the influencing factors.The compressive strength,the flexural strength and the moisture absorption were used as the measurement indexes.The three-factor four levels were used to design the orthogonal test and the concrete specimens was poured.The test results showed that the optimal combination: group A is M1N3L0(water-cement ratio 0.34,40% sand,zeolite powder 0%);group B is M3N1L4 or M3N4L4(water-cement ratio 0.46,36% sand,or 42%,zeolite powder 20%);group C is M3N1L3(water-cement ratio 0.46,sand rate 36%,zeolite powder 15%).(4)In this study,the absorption and desorption properties of zeolite concrete were predicted by using the BP neural network model with 3-11-1(three layers of neurons in the input layer and 11 neurons in the hidden layer and the output layer 1 neurons)The MES,R~2 and MAPD of the training data were 583.09,0.984 and 2.87% respectively,and the MES,R~2 and MAPD of the predicted data were 399.48,0.929 and 0.34% respectively.The results show that the model is reasonable and reliable for predicting the hygroscopicity of zeolite concrete. |