Research On Detectiom Method For Orchard Tree Canopy Density Based On Ultrasonic Sensor

With the increasing of emphasis on food safety and environmental protection,people pay more and more attention to excessive pesticide residues in fruit and soil.The implementation of target-oriented variable-rate spraying is an important technical measure to reduce pesticide consumption and chemical pollution.What's more,the target detecting is the key of target-oriented variable-rate spraying.In this thesis the orchard tree canopy density was detected based on ultrasonic sensor and the canopy density models were established through orthogonal experiments.The canopy density model based on time-domain energy was of good universality,which could provide variable control basis for an upper computer to achieve precision spraying.(1)Tree canopy density detection test system and mathematical model experiments were designed.According to the general requirements of canopy density detecting,in this thesis a test bench and detecting system was developed under laboratory conditions.This system could change the canopy density by changing the weight of leaves which were arranged manually,and also could collect and storage the data from the ultrasonic sensor using the oscilloscope and its PC software.Besides,the method of orthogonal regression equation establishment and statistical verification method of significance of the regression coefficient,regression significance and lack of fit for model equation was introduced,and quadratic regressive orthogonal experiments were designed to establish the mathematical model of canopy density.(2)Characteristics of sensors and information extraction methods of ultrasonic echo wave were studied.Experiments of characteristic parameters for the MB7092 type ultrasonic sensor were carried out.Those results provided a basis for the ultrasonic signal processing and analysis.Meanwhile,this thesis conducted the correlation experiments to explore the relationship between ultrasonic echo and canopy density,and proposed two analytical methods including echo wave envelope analysis and time-domain energy analysis which can characterize the change of canopy density.According to the two analytical methods,the ultrasonic signal it was tested and analyzed respectively.(3)Canopy density model was set up based on echo wave envelope analysis.The experiments were conducted based on quadratic regression orthogonal design.The ultrasonic echo wave envelope which was calculated by MATLAB software was used as a result of the experiment.It established canopy density models with different canopy thickness with 2 layers,3 layers and 4 layers,and verified regression significance and lack of fit for the model equation.The results showed that the model equation had a good regression significance,but some model equations was lack of fit with larger errors.(4)Canopy density model was established based on a time-domain energy of echo.MATLAB software filtered the signal and calculated echo energy based on time-domain energy method so as to establish canopy density models with different canopy thickness with 2 layers,3 layers and 4 layers.The canopy density model of 4 layers was selected as the optimal model equation through statistical verification and comparison of those model equations.The verification test was conducted at the same density values and detecting distances with different layers,and canopy density model equation verification showed that the relative errors of model value and actual value were small.The maximum relative error of 4 layers model equations verification at 2 layers,3 layers,4 layers and 5 layers were 29.92%,21.33%,25.64% and 17.68%.