Nano-organic Working Fluid Stability, Thermophysical Properties And Modeling Based On Artificial Neural Network

In the industrial production process,a large amount of waste heat is generated.Due to the low taste and distribution of low-temperature waste heat resources,a large number of low-grade waste heat resources are not utilized.Organic Rankine Cycle（ORC）is one of the effective ways to recover and efficiently utilize low temperature waste heat.At present,pure organic working fluid is generally used as the circulating working fluid of the ORC system.However,the thermal conductivity of pure organic working fluid is low,which affects the efficiency of system waste heat utilization.It has problems such as low thermal conductivity of organic working fluid and low recovery and utilization rate in heat transfer process.In this paper,the combination of experiment and theory is used to study the stability of single and mixed nanofluids in the process of easy precipitation of nanofluids,and to study the variation of parameters such as mass fraction,temperature and mixing ratio.According to the experimental data,the thermal conductivity and viscosity were modeled and predicted based on artificial neural network method,which provided a theoretical basis for the application of nanofluids in the recovery of medium and low temperature waste heat.It mainly includes the following aspects:1 The stability and thermal properties of Al₂O₃-EG single nanofluids were studied.Firstly,the stability of Al₂O₃-EG nanofluids under different ultrasonic time was quantitatively analyzed by particle size distribution and velocity ratio.It was found that the ultrasonic time was the best ultrasonic time at 60 min.Then,the effects of temperature and mass fraction on the thermal properties of Al₂O₃-EG nanofluids were investigated.The results show that the Al₂O₃-EG nanofluid with mass fraction of 2.0wt.%increases the corresponding thermal conductivity from 9%to 16%when the temperature is increased from 25°C to 60°C.In the case of ultrasonic 60 min and60°C,the maximum increase in viscosity of the nanofluid with a mass fraction of 2.0wt.%was 18.2%.According to different heat transfer performance evaluation criteria,it is found that the Al₂O₃-EG nanofluid is advantageous in heat transfer applications in laminar and turbulent conditions,and serves to enhance heat transfer.2 The stability and thermal properties of CuO-ZnO hybrid nanofluids were studied.Firstly,the effects of ultrasonic time and surfactant on the stability of CuO-ZnO hybrid nanofluids were studied.It was found that 120 min was the best ultrasonic time and PVP could improve the stability of CuO-ZnO hybrid nanofluids.Then,the effects of temperature,mass fraction,base ratio and surfactant on the thermal properties of mixed nanofluids were discussed.The results show that at 60°C,the mass fraction of 5 wt.%CuO-ZnO mixed nanofluidφ_v is 20:80%,40:60%,50:50%,60:40%and 80:20%,corresponding The thermal conductivity increased by 26.1%,22.9%,19.3%,16.3%,and 14%.At a normal temperature of 25°C,the viscosity of the CuO-ZnO mixed nanofluid with a base solution ratio of 60:40（EG:W）is 20%higher than that of the mixed base liquid.When the ratio of the base liquid is 40:60（EG:W）,the optimum addition amount of PVP is 1.0 wt.%.3 Based on experimental data,the thermal conductivity and viscosity were modeled and predicted by neural network method.The thermal property prediction model of single nanofluid and hybrid nanofluid based on artificial neural network method is studied,and the formula prediction model is compared with the artificial neural network model.It was found that the artificial neural network model predicts that the thermal properties of single and mixed nanofluids are more accurate than the formula predicts thermal properties.