Whole-range Self-adaptive Control System Of Combine Harvester Operating Speed Based On Association Rules

As the automatic operating control of combine harvester, the changes of cropping density, cropping moisture content, and even terrain all can influence on the change of feeding quantity, which causes these rotation speed changes of cutting table auger, conveyer trough and threshing rotor. Importantly, the change of threshing rotor rotation speed will directly influence on its grain threshing performance and separating performance. Therefore, some relevance is existed among these rotation speeds, forward speed, feeding quantity and grain harvesting loss rate. Analyzing the relevance among multi-source information, studying whole-range self-adaptive control system of combine harvester operating speed based on association rules and exploring self-adaptive control rules of combine harvester operating speed not only provide the basis for these designs and optimizations of existing intelligence control devices, but also enlighten finding new intelligence control strategies, which has important practical significance and scientific research value. In this paper, combined with the National High Technology Research and Development Program of China(863 Program) and Science and Technology Support Project of Jiangsu Province, the study of combine harvester operating information monitoring and intelligence control is developed by comprehensive applications of these techniques and theories which are model analysis, kinetic analysis and modeling, association rule data mining technique, computer simulation and embedded technique. The major research work is as follows:(1) Based on the analysis of combine harvester operating system models and the association rules mining of operating parameters, the association rules among operating parameters and feeding quantity or loss rate are selected and then the valuable knowledge is gained which can influence on combine harvester feeding quantity and loss rate. Through the analysis of operating parameters’ model of main work parts in combine harvester, it is certain that some relevance is existed among the operating parameters(cutting table auger rotation speed, threshing rotor rotation speed, conveying trough rotation speed and forward speed) and feeding quantity or loss rate. A group of sample data of combine harvester operating information is mined by using association rule technique. The strong association rules are selected from association rules base, an association knowledge is gained which influences on feeding quantity and loss rate, and according to the confidence and importance of these strong association rules the weight factors of threshing rotor, cutting table auger and conveyer trough impacting on harvester’s harvesting performance are obtained approximately using normalization method. However, considering if taking the incompleteness of the data into account, influencing the harvesting performance of the combine harvester should be defined as finding an the optimal trade-off among these weight factors of the threshing rotor, cutting table auger and conveyer trough, so these weight factors’ value ranges set are respectively [0.4 0.6], [0.3 0.5] and [0 0.3].(2) The kinetic model of combine harvester threshing system is built and operating speed control system simulation is constructed, and then fusing the operating parameters’ association knowledge an operating speed control model based on association rules is established. Aimed to the question that the existed model did not take into account movement states of other work parts, taking the XG610 combine harvester for example, the kinetic model for threshing system is established based on kinetic analysis of work parts. According to the kinetic model of threshing rotor, an operating speed ordinary fuzzy control model is constructed. Seen from the simulation change curves of threshing rotor rotation speed and forward speed, and the change range of the rotor rotation speed is not beyond the change range allowed, which meets the operating speed control requirements of combine harvester, so the kinetic model of threshing system constructed is reasonable and feasible. At the same time, based on the ordinary control model and fusing the operating parameters’ association knowledge, an operating speed control model based on association rules is established, and its simulation results is compared to the ordinary control model’s results. Through the comparison of their simulation results, under the almost same harvesting performance the forward speed adjustment performance of the operating speed control model based on association rules is superior to the ordinary control model, and its stable relative change ranges decreases 0.70% than that of the later and its maximum relative change range decreases 1.50% than that of the later. Furthermore, the system adjustment time of the control model based on association rules is about 11 s shorter than 16 s of the ordinary control model adjustment time,and the whole control stability of the former is better than that of the later.(3) The whole-range self-adaptive fuzzy control model based on association rules is constructed, and makes a simulation comparison to the ordinary control model and the control model based association rules. Through the study of self-adaptive fuzzy control theory and structure, based on the association rules operating speed control model, and the whole-range self-adaptive fuzzy control model based on association rules is constructed and self-adjustment factor fuzzy controllers are designed. At the same time, when the feeding quantity increases about 15%, the maximum relative change range of threshing rotor rotation speed is about 5.48% and its stable relative change range is about 2.62%, and the maximum relative change range of forward speed is about 9.00% and its stable relative change range is about 7.80%, and the system adjustment time is about 8s. In addition, the feeding quantity and the unit loss rate is respectively 3.88 kg/s and 0.55%/(kg/s). Compared to the simulation curves of the former two control models, the whole-range operating speed self-adaptive control model based on association rules is not only superior to the former two control models on these control performance aspects of the stable change range of threshing rotor rotation speed, the maximum change range of forward speed, the system whole stability and the system adjustment time, but also superior to the former two control models on the overall harvesting performance.(4) The hardware system construction and software system development of combine harvester operating speed control are studied and then tested in indoor simulation speed regulation device of combine harvester. According to the features of the operating information, the hardware of combine harvester operating speed control system is designed based on ARM9 embedded technology combining with kinds of inputting and outputting hardware, and two interfaces of video model and GPS model are left in the system in order to monitoring field operation and harvester position in the later. The driving programs of these external hardware devices are developed. The application software of combine harvester operating speed control system is developed by using Qt Designer. The software includes five parts which are respectively system main interface, parameters setting interface, operating speed monitoring and intelligence control interface, video monitoring interface and GPS information monitoring interface. Then it is tested in indoor simulation speed regulation device of combine harvester. The results show that operating information monitoring data is stable and timely, and according to field simulation experimental condition the stepper motor control can meet respectively their respective forward speed regulation by using the ordinary control algorithm, the control algorithm based on association rules and the whole-range self-adaptive control algorithm based on association rules.(5) The operating speed control system is debugged and made field test aimed to three control algorithms from three control models. Debugging works include main work parts’ rotation speed calibration, forward speed calibration and automatic control operation tests. In field tests, the ordinary control algorithm, the control algorithm based on association rules and the whole-range self-adaptive control algorithm based on association rules are used respectively to make harvesting rice experiments, and then their experimental data is analyzed and contrasted. Seen from the overall change trends of threshing rotor rotation speed and forward speed in three control algorithms are accorded with the simulation results, and the maximum change range of the rotor rotation speed are all not beyond the change range allowed, so it further verified that the established kinetic model of threshing system is reasonable and feasible. At the same time, the comparison results show that the stable average change range of threshing rotor rotation speed is about 2.97%, the maximum change range of forward speed is about 9.00%, and the system takes about 7 s from the step change of these operating speeds to basic stable state, the control performance data using the whole-range self-adaptive control algorithm based association rules is superior to that using the ordinary control algorithm and the control algorithm based on association rules. At the same time, the average feeding quantity using the whole-range operating speed self-adaptive control algorithm based on association rules is slightly smaller than the latter two control algorithms, but the average loss rate decreases respectively 0.29% and 0.22% than that using the latter two control algorithms and the unit loss rate reduces respectively 0.06%/(kg/s)and 0.05%/(kg/s)than that using the latter two control algorithms, the reduction range of the loss rate is larger compared to the feeding quantity. Therefore, the whole-range operating speed self-adaptive control system based on association rules is not only superior to the ordinary control system and the control algorithm based on association rules in control performance, but also better than the latter two control systems in overall harvesting performance.