| The real-time and high-precision monitoring of the seeding performance of the air suction corn precision planter is one of the important ways to save fine seeds,increase production and achieve high-speed operation.Reasonable design of the capacitive sensor plate structure,study on the capacitive pulse signal processing algorithm,and correction compensation for the sensor nonlinearity,which is of great significance for improving the detection accuracy of seeding performance and ensuring the seeding quality.In this paper,theoretical analysis,experimental research,modeling and simulation and algorithm research are used to study and explore the detection mechanism of seeding performance.1.The effects of electric field frequency,moisture content,temperature and bulk density on dielectric properties of maize seeds were studied by single factor experiment.It is concluded that temperature has little effect on the dielectric properties of corn seeds within the range of safe moisture content under the condition of certain bulk density and electric field frequency,and the effect of temperature on dielectric properties increases with the increase of moisture content.With the increase of frequency,the dielectric constant decreases rapidly and tends to be stable at different temperature.The electric field frequency is determined as 12 Hz.2.In order to solve the problems of non-linearity and low electric field intensity of single-chip interdigital electrode structure in high-speed seeding detection,high-speed photography was used to analyze the position of corn seeds in different rotation speeds of seeding axis.Based on the Maxwell electrostatic field simulation analysis,the structure of E-type orthogonal interdigitated capacitor plate is determined.Based on Matlab,the constraint range of plate structure parameters is analyzed,and the optimum combination of plate structure parameters is determined by three factors and three levels orthogonal test.3.For the small capacitance increment(PF level)produced by the single corn seed passing through the capacitor plate,the micro capacitance detection circuit is designed by using the special integrated micro capacitance chip PCAP01-AD.The capacitance pulse signal is collected by bench test,the waveform characteristics are analyzed,and the threshold of wavelet based on EMD decomposition is used to filter the baseline drift disturbance,and the smooth curve preserving the original signal characteristics is obtained single peak,overlapping peak,weak peak and leakage can be located accurately.The test results show that the precision of seed metering performance is 100%.4.Through the air-suction two-row corn precision planter,the outdoor experiment and the static test of the capacitance sensor were carried out based on the capacitance sensor seeding performance monitoring system,and the non-linearity between the real value and the detected value of the seeding amount was determined.Based on GA-RBF neural network correction method,a non-linear correction model of capacitance sensor is constructed,and the genetic algorithm is used to optimize the network structure parameters quickly,to solve the problem of RBF neural network learning falling into local minimum,the relative error is less than 1% and the precision of is over 99.57%.5.The visual monitoring system of precision corn planter is designed.The control and detection data transmission with the slave computer is realized by wireless communication.The man-machine interface can display the performance parameters of seeding and fertilizer performance,such as locomotive data,seeding amount,reseeding rate,leakage rate and empty or blocked alarm of the seed pipe.The results of field test show that the precision of each row is not less than 99.37%,the maximum deviation between the reseeding rate and the leakage rate is 0.3%,the alarm accuracy of abnormal situation is 100%,and the delay distance of alarm is not more than 1.2 meters. |
PDF Full Text Request