Design Of New Cloud Water Content Sensor And Array Spherical Temperature Sensor

Meteorological detection is of great significance to agricultural production,disaster prevention and mitigation,transportation and energy security,national defense and other fields.Meteorological sensors provide raw measurement data for applications such as climate change observation and numerical weather prediction,and play an indispensable role in the meteorological industry..This paper designs a cloud water content sensor with a "Y" structure suitable for radiosondes,which is used to measure the liquid water content in the cloud,and specifically solves the problems of traditional measuring instruments such as difficulty in manual operation and high power consumption;An array temperature sensor is also proposed,which can reduce solar radiation errors and improve the accuracy of temperature observation.It can be used in ground weather stations or carried by radiosondes for high-altitude observations.This paper uses computational fluid dynamics(CFD)method to simulate the relationship between the heating power of the cloud water content sensor and the contacted liquid water content.In order to ensure the temperature control effect of the cloud water content sensor,a single neuron PID algorithm is used to heat The temperature of the component is adjusted in real time.In order to carry out the experimental test,a set of environmental simulation experiment platform was built.The results show that when the cloud water content is 1.0g/m3,the average power consumption of the cloud water content sensor is 1.74 W.In order to reduce the solar radiation error,an array type spherical temperature sensor is designed,which does not require the use of traditional shutters or radiation shields.Using the radiation error ratio relationship between the probes of the array type spherical temperature sensor,the true value of the ambient temperature can be derived.It is proposed to use the extreme learning machine(ELM)algorithm to obtain the relationship between the ratio and environmental factors.The algorithm fitting root mean square error is 0.0245 K,the real measurement error is 0.145 K,and the root mean square error is 0.16 K,which is good Measurement accuracy.