| The 21st century agriculture is moving in the direction of precision agriculture,precision agriculture will be information technology into agricultural production,to achieve the crop from the sowing,fertilization,top dressing,weeding,to harvest a comprehensive intelligence,improve the efficiency of agricultural operations.Variable fertilization is an important part of achieving precision agriculture.Variable fertilization is based on the fertility of farmland and the nutritional status of crops needed to adjust the amount of fertilizer to achieve on-demand fertilization,and minimize the abuse of fertilizer,reduce farmland pollution,reduce agricultural production costs,improve the quality of crops.Planting fertilizers is the main form of variable fertilization.The control of sowing is the key to realize the variable fertilization.At present,the mainstream sowing detection method is to quantify the quantitative relationship between the sowing and the sowing volume of the outer tanker by sowing and fertilizing one machine.According to the actual working speed of the sowing and fertilizing machine,the speed of the sowing and fertilizing machine is adjusted and the sowing speed is realized the amount of control.However,this method is essentially open-loop control,through the calibration model to calculate the real-time broadcast.The accuracy of the sowing is susceptible to the influence of the operating conditions,such as the brush wear,the change in the size of the fertilizer particles,the wetting of the fertilizer after the moisture in the outer tanker,and the factors that affect the quantitative relationship of the calibration model,and affect the fertilization effect.At present,there are very few studies on the method of directly detecting the amount of fertilizer applied and the corresponding detection device.In order to improve the accuracy of the sowing detection of the sowing detection method,the sowing quality of the sowing detection method is improved,and the sowing quality of the sowing detection method is improved.Detection system for mathematical modeling analysis and interference information analysis,proposed a detection system for information processing methods,can effectively reduce the external interference and the physical characteristics of the system caused by the measurement error.The specific research contents and research results are as follows:(1)A method of testing the sowing and sowing of the sowing and fertilizing machine is put forward.The embedded structure design of the weighing sowing detection device is carried out.(2)According to the principle of measurement and the physical characteristics of sowing and fertilizing the box,the core parts of the detection device are selected and the test device based on the weaving machine of the outer tanker sowing and fertilizing machine is made.The experimental device can carry out the outer slot control of the variable fertilization experiments,but also can be obtained based on the weighing sensor sowing detection test.(3)The spectral characteristics of the jamming signal of the load cell are analyzed,and the main interference components are identified as power frequency interference and white noise interference.The corresponding low-pass filter and mean filter algorithm are designed to effectively suppress the interference signal and improve the detection signal to noise ratio.(4)The symmetric mass sowing detection device is modeled and analyzed to determine the transfer function of the system.The adaptive compensation algorithm is designed by designing the system pole to adjust the damping ratio of the compensated system to 0.707,which effectively improves the output of the detection device Signal dynamic response characteristics.(5)Experiments on the characteristics of sowing detection devices in stable working environment.In the experiment,the weft sowing and fertilizing machine equipped with the weighing sowing detection device was suspended on the tractor,and the fertilization and sowing experiment was carried out on the flat road,and the sowing speed was maintained by controlling the outer ring speed.The sowing detection accuracy of the weighing-type sowing detecting device is tested based on the sowing value obtained by the trough rotational speed control as a reference value.The results show that the target sowing rate of wheat,urea and compound fertilizer is 10.4kg,5.2kg and 12.9kg,respectively,after the low-pass filter,mean filter and system pole configuration adaptive compensation of the load cell detection signal The 95%confidence intervals for the average absolute error of cumulative sowing were[0.14 kg,0.16 kg],[0.08 kg,0.10 kg]and[0.11 kg,0.13 kg].(6)The field performance test and error analysis of the weighing broadcast detection device were carried out.The influence of vibration and the inclination of the box on the measurement results is analyzed experimentally,and a solution based on low-pass filter is proposed.Based on the sowing speed obtained by the control of the wheel speed,the field sowing detection test was carried out,and the data was analyzed by using a 77 meter as a test unit.The results showed that after the single test was completed,Setting wheat,urea and compound fertilizer target sowing 8.0kg,4.0kg,8.0kg,the maximum relative error of total accumulated sowing of wheat,urea and compound fertilizer was 6.73%,7.38%and 4.94%respectively.Setting the target sowing of wheat,urea and compound fertilizer were 6.6kg,2.7kg,6.6kg,respectively,The 95%confidence interval of the cumulative absolute error of wheat,urea and compound fertilizer was[0.48 kg,0.64 kg],[0.38 kg,0.60 kg]and[0.61 kg,0.79 kg].The general research shows that:the proposed method of weighing broadcast detection is feasible.Due to the defects in the mechanical structure,the suspension structure of the sowing detection is not stable,and the suspension in the field is easy to cause random swing due to vibration.It Result in a large random error,dynamic response quality deterioration.At the end of this paper,an improved design scheme of mechanical structure is proposed. |
PDF Full Text Request