Research On Ground Temperature Measurement System Based On Computational Fluid Dynamic

Accurate temperature observations are of great importance to our national defense security,disaster prevention,and atmospheric science research.However,existing temperature observation systems are susceptible to heating by solar radiation and poor air circulation due to their structure and materials.The measured values of these systems are higher than the actual free air temperature,with radiation errors of up to 1 ℃.Therefore,compared with the error caused by the temperature measurement circuit system,the radiation error included by solar radiation has become the main error source.How to reduce the radiation error is a key point to improve the accuracy of surface air temperature observation.On this basis,this paper proposes a series of new surface air temperature observation systems,including a temperature measurement circuit system and a new radiation shield.The circuit system is a four-wire platinum resistance temperature measurement circuit based on a potentiometer bridge,which can effectively improve measurement accuracy.The new radiation shield is composed of silver-plated aluminum plates and airflow deflectors.The silver-plated aluminum plates with 95% reflectivity can effectively reduce the radiation impact.Airflow deflectors can guide the airflow to the temperature-sensing probe and promote the diffusion of radiation heat to achieve the design goal of reducing radiation error.The computational fluid dynamics(CFD)method was used to quantify the radiation error of the new surface air temperature observation system under different meteorological environments including solar radiation,reflected radiation from the underlying surface,longwave radiation,diffuse radiation,air density,solar elevation angle,and wind velocity.The calculation results show that the new surface air temperature observation system has better radiation protection performance and ventilation effect compared with the traditional natural ventilation temperature observation system.To further improve the accuracy of the new temperature measurement system,this paper uses the MLP neural network to train and learn the radiation error calculation data of the new system.The environmental variables and the radiation error were input parameters and output,respectively.The training results show that the differences between the sample values and the predicted values provided by the correction algorithm were mainly concentrated within± 0.005 ℃.To verify the measurement accuracy of the new surface air temperature observation system and the accuracy of the radiation error correction algorithm,the measured values of the 076 B artificially ventilated air temperature observation system were used as the reference temperatures during experiments.The experimental results show that the average radiation errors of the three new surface air temperature observation systems before correction are 0.07 ℃,0.18 ℃ and 0.15 ℃,respectively.After correction,the radiation errors of the three systems can be reduced to the order of 0.05 ℃.The three systems have high temperature measurement accuracy.After correction,the measurement accuracy of the systems can be further improved.Finally,to facilitate the application of correction algorithms by meteorologists,this paper proposes a highly applicable software for querying and correcting radiation errors,using Java language programming.The radiation errors and the corrected temperature data can be easily and intuitively queried.