Yaoming Li

A Pvdf Sensor for Monitoring Grain Loss in Combine Harvester

Grain loss sensor is an important monitoring module to realize the intelligent automatic control in combine harvester. Polyvinylidene fluoride piezoelectricity film has been one of the most widely used macromolecule materials at present. Based on its piezoelectric effect, in this paper, a grain loss monitoring system for combine harvester by using the PVDF sensor was designed. The system is composed of the PVDF sensor, intelligent display, SCM and so on. The working principle was analyzed and the measuring accuracy was dynamic calibrated in laboratory. The polyvinylidene fluoride piezoelectricity film was selected as the sensing material and the signal processing circuit was mainly composed by charge amplifier, band-pass filter. In order to verify the accuracy of the sensor, the dynamic calibration experiments were carried out. By adjustment of the feeder’s vibration parameters, the different feeding velocity can be gained. In the dynamic calibration experiments, the sensor was fixed at the height of 19.5mm and the fixing angle of 33°~35°, the feeding velocity of 60 80 grain per second, 1000samples of full grain were randomly selected in the experiments, and the results indicated the detecting error was less than 5.7%. Using the different mixture ratio of the full grain, the imperfect grain, short stalk to do multi-group experiments, the result shows that the sensor can distinguish the grain from the mixture stalks. In the rig-tests, the monitoring sensor was fixed on the caudal region of the cylinder concave grid of the combine harvester in order to monitor the threshing compounds. The precision error of the monitoring sensor in combine is less than 8% through field experiments. So this advanced sensor can realize the aim of monitoring the entrainment loss in combine harvester.

Research of predictive system for feed quantity of combine based on fuzzy neural network

A predictive system for feed quantity of combine based on fuzzy neural network was proposed to solve the delay problem of measuring feed quantity. This paper constructed the predictive system according to the actual destination and then decided the input and output of the network and its fuzzy dividing. At the same time, this paper investigated the construction of fuzzy rule base and designed the topological structure of networks. From experiments, this paper selected the example data for testing and abstracted initial fuzzy rules and trained the network using the example data and checked the generalizing ability of the network. The result proved the network of predictive system to be practicable. It was useful to improve real-time performance of control system of combine.

Study on Integration of Measurement and Control System for Combine Harvester

Using CAN (controller area network) field bus and iCAN protocol, an integrated measuring and controlling system for combine harvester is constructed in this paper. ARM is used as the processor of host computer. Lower computers are several measuring and controlling modules of combine harvester. Host computer communicates with lower computers through CAN bus. Information between measurement and control modules can be shared so that combine harvester’s whole performance can be improved. LCD on host computer shows working state and values of relative parameters of each module in real-time. Experiment shows that the system can work correctly and reliably.