| With the development of the current field of wind-vector detection,there are some difficult problems such as Micro air vehicles(MAVs)wind-vector detection.It is extremely important to accurately perceive the airflow velocity and direction information of the environment.The traditional flow direction detection technology has limitations such as large mechanical loss,large size,low precision and high cost,so it can not meet the requirements of accurate flow direction detection.In nature,the trichobothria grown on the surface of scorpions has excellent wind-vector sensing performance,which can accurately locate the flying prey.This paper reveals the supersensitive cooperative sensing mechanism and functional integration mechanism of scorpion trichobothria on wind-vector,and prepares a bionic wind-vector sensor for airflow velocity direction detection.The main research work is as follows:(1)In this thesis,using the scorpion Heterometrus petersii as a biological template,the geometric parameters of scorpion trichobothria were statistically analyzed by optical and electron microscopy,and the flat and distorted morphology of scorpion trichobothria rod was verified.The relationship parameters such as the specific length ratio of the flat structure of the scorpion trichobothria rod were revealed.This paper tested the directional selectivity of scorpion trichobothria,and verified the perception characteristics of scorpion trichobothria for wind-vector.On this basis,the cooperative sensing mechanism of the flat structure and the cylindrical structure on the directional selectivity and velocity sensitivity of the scorpion trichobothria rod was revealed by using the finite element analysis method.(2)Based on the morphological characteristics of the scorpion trichobothria rod,the bionic hair rod structure model was established,the base sensing diaphragm structure based on the bionic hair rod structure was determined,and the feasibility of sensor design was demonstrated by finite element analysis method.Based on 3D printing,MEMS and other advanced micro and nano scale processing technologies,a bionic hair rod preparation method,a substrate diaphragm preparation method and a sensor micro-assembly method were proposed,and a bionic wind vector sensor was successfully prepared.(3)A wind tunnel experiment platform was built to test the sensitivity,repeatability and other performance indexes of the bionic wind-vector sensor.The maximum wind direction sensitivity was 0.53μV/°,the maximum wind speed sensitivity was17.18μV/m/s,and the maximum range was 0-5 m/s.It shows the good directional selectivity and wind-vector sensing sensitivity of the sensor in this paper.In the multiperiod wind tunnel switch cycle test,the sensor’s average voltage change is only Δu =± 0.00026 V under static state.In the response state,the average voltage change of the measured sensor is only Δu = ± 0.00032 V,which shows the excellent repeatability of the sensor in this paper.(4)The sensor array that can be used for wind-vector measurement was built,the response time series signals of the sensor array about the wind-vector were obtained,the data set based on the response signals of the sensor array was constructed,the windvector recognition method based on BP neural network was proposed,the neural network model structure was designed,and the model training and prediction analysis were carried out.On the randomly constructed test set in this paper,the average error of wind direction prediction is only 3.149°,and the average error of wind speed prediction is only 0.00186 m/s.The results show that the sensor array in this paper has a very high ability to recognize and detect wind-vectors. |
