Study On Designing Method Of Multi-duct Air-and-screen Cleaning Unit And Grain Sieve Loss Monitoring And Controlling Technology

Rice is one of the most important grain crops in China with a planting area about 30million hm² and an annual yield over 200 million tons.Currently,the larger part of these fields are harvested by combine harvesters.With increasing rice grain yields and feed rates,the cleaning system of a rice combine harvester has to deal with larger volumes of threshed outputs.Together with the threshing and separation section,the cleaning device can be considered as the'digestive system'of a combine harvester.This makes it one of the core parts of the machine.Field experiments have shown that the classical single-duct cleaning systems used in most rice combine harvesters have become a limiting factor,resulting in a high grain loss and high grain impurity ratio.On the other hand,the settings of the cleaning system cannot autonomously optimized during harvesting,leading to a poor cleaning adaptability under different working condition.Therefore,in this thesis,computational fluid dynamics?CFD?,hot wire anemometer?HWA?,discrete element method?DEM?and sensor technology were comprehensively applied to study on working mechanism and design method of multi-duct air-and-screen cleaning unit and develop a control system for cleaning process to decrease grain sieve loss.The major research work is as follows:1.To identify the causes for this poor performance,the distribution of the threshed outputs on the cleaning shoe,the terminal velocities of the different components and turbine flow meter measurements were used to obtain insight in the airflow distribution inside the cleaning shoe.Based on these insights,a multi-duct cleaning device with return pan was proposed for tangential-longitudinal-flow combine harvesters.According to distribution varation of the threshed output inside the cleaning shoe during continous cleaning process,the airflow resistance model for different working parts were developed,and the resistance value were calculated according to ideal airflow velocity distribution in different locations,which can lay a good foundation for designing multi-duct fans and optimizing structural of the cleaning system2.Designing perforated plates with different opening ratio to simulate cleaning load,and utilzing porous media to define the airflow resistance of the cleaning load in Fluent simualtions.CFD and HWA have been used to study the airflow distribution inside the fan and effects of working load on fan performance was analyzed in detail,the curve of?ptot versus the total volume of the airflow Q were fitted,and the fan with good performance was selected based on the criterion of'the slope of the?p_tot-Q curve at the working area should be as steep as possible,which means that as the load increases,the flow rate should not drop excessively.On the basis of CFD simulation results which shown the newly designed cleaning system has an ideal airflow distribution,a cleaning test-bench which equipped with extra sensors to extract some valuable information concerning the cleaning section,such as airflow velocity,fan speed,grain loss was developed,and some settings such as sieve opening and vibration frequency and fan speed can be adjusted separately.The simulation results for the whole field flow calculations were validated through comparison of simulated airflow velocities at certain points inside the cleaning shoe.According to cleaning preformance in different conditions and corresponding airflow distribuion,an ideal airflow velocity distribution was put forward,that is,For‘winnowing stage',the optimum airflow velocity was recommended to 6-9 m/s,thus threshed output can be blown off instantly,most of the chaff will be blown out of the cleaning shoe directly owing to their small terminal velocity.Cleaning loading can be reduced in large extent.In the following steps,sufficient airflow rate is required to assist in positioning grains over the upper sieve.According to the measured aerodynamic properties of the rice outputs,the airflow velocity above the upper sieve should be around 5-6 m/s to help grains easily pass through the sieve openings.The airflow velocity at the tails should be<4 m/s to avoid that the remaining rice grains are blown out.Effects of working parameters on airflow distribuion also has been investigaed and corresponding prediction models have been established which can give an insight into developing control strategy.3.Rice grain and short straw particle models were established according to their physical properties,and DEM simulations were carried out to understand their collision behaviour with the sensor.The influence of grain shape,straw length and impact angle on variations of the maximum normal contact force and force rise-time were analyzed in detail.According to differences in normal collision force,and force rise-time occurred will lead to corresponding differences in signal frequency and voltage amplitude,a signal processing circuit,which mainly consisted of a band-pass filter circuit and a voltage comparator circuit,was designed to discriminate for full grains.To accelerate grain sieve loss monitoring sensor's detecting frequency,theoretical analysis was performed from view of structural vibration and a partial constrained viscoelastic layer damping?PCLD?treatment was carried out on the surface of the developed sensor.On the basis of analyzing the mechanical relationships among each layer of the sandwich in depth,optimal positions to past the viscoelastic layers were found.Calibration experiment results indicated that sensor's detecting frequency was significantly improved after treated with PCLD.Understanding the relationship between counted grain numbers and the total grain sieve loss is essential for monitoring the grain losses in real-time.To apply a grain sieve loss monitoring mathematical model,laboratory experiments were carried out on the cleaning test-bench.Distribution of grain cleaning losses on rear of cleaning sieve was studied,and the mathematical model between total losses of grain cleaning and grain volume tail on different regions was developed.Utilizing the developed mathmatical,grain sieve loss can be calculated in real-time.4.Selecting S7-1200 PLC as the main control unit to build the lower computer hardware system and utilsing ladder language to complete the system compilation,and LabVIEW 14.0 software was used to design the host computer interface,multi-duct cleaning device performance monitoring and control system was developed.MODBUS-TCP Ethernet protocol was selected to realise the task of data exchange between host computer interface and lower computer hardware system,the developed performance monitoring and control system for multi-duct cleaning device with functions of parameter setting,display,fault alarm,data storage and playback,automatic control,communication and emergency stop.Effects of fan speed,guide plate angle,sieve opening on sieve loss ratio,grain impurity ratio has been investigated through a large number of bench tests.The relevance level of operating parameters on performance parameters also has been determined,a fuzzy control model of the cleaning process was developed for the multi-duct cleaning system,and experiment results indicated that the designed fuzzy control model can fulfill the purpose of automation control of cleaning section settings,such as fan speed and guide plate angle.