Fanglan Ma

Research on the working mechanism and virtual design for a brush shape cleaning element of a sugarcane harvester

Abstract The paper elaborates the working mechanism of a brush shape cleaning element of a sugarcane harvester, and analyzes the state of the cleaning procedure and the reacting force to which the cleaning element is subjected. Combined with earlier stage experiments and research, it was found that the factors which affect the cleaning efficiency and fatigue life of the cleaning element are numerous, and also complex and random. Thus it was difficult to analyze and design a cleaning element in theory. Moreover, if the design parameters were to be obtained from the experiment it would be time consuming, laborious, and costly. The method of a numerical value simulation orthogonal experiment is used to analyze the reacting force by means of establishing a finite element analysis model for the cleaning element on ANSYS. According to the orthogonal experiment principle, different parameter groups were made for finite element calculation, and then orthogonal experiment analysis and regression analysis were conducted to determine the effect on the calculated result that was caused by the changes of different parameter groups, and thus obtain the optimal design parameters of the cleaning element. In summary, this virtual design method is a quick and efficient way in innovative mechanical product design.

Knowledge acquisition based on neural networks for performance evaluation of sugarcane harvester

Expertise acquisition is always the obstacle and bottleneck in the development of intelligent design system. In order to generalize and accumulate the expertise and experience of simulation analysis and experiments, the intelligent design system of sugarcane harvester is introduced. In the intelligent system of sugarcane harvester, the neural network is applied to overcome the difficulty of knowledge acquisition (KA). In this study, the application of neural network in the system is illustrated, including data predisposal, generation and management of the knowledge. An example is given to explain the application as well. The research shows using neural network can simplify the procedure of knowledge acquisition. It can also evaluate and forecast the performance of sugarcane harvester in design phrase. And it is beneficial to enhance the development success rate of the digital product and to lessen the development cost of physical prototype.

Cutting Force Prediction of High-Speed Milling Hardened Steel Based on BP Neural Networks

It is absolutely necessary to machine the complex mould cavity with micro-ball end mill in high speed machining. Because of the priceless and frangibility of the tool, it is significant to predict the cutting force in order to lessen the tool’s disrepair, to ensure the quality and to improve efficiency. This paper established the cutting force prediction model based on BP neural network when machining arc surface of hardened steel in high speed. According to the sample set of experimental results used to train and test the neural network, we realize the cutting force prediction and simulation through introducing the elastic grads’ decrease method to improve the speed of convergence and precision in the process of cutting. The practice showed that most of the error values are about 5% except some special individual. Obviously, to predict the cutting force is feasible in the process of non-stability and non-linear extremely of cutting through the non-linear neural network.

Simulation Analysis and Experimental Study on Rigidity Enhancement of Sugarcane Harvester Cutter

The rigidity of sugarcane cutter has notable influence on the quality of sugarcane cutting,the most concerned problem of sugarcane harvester developing currently,as it effects its vibration and the vibration in turn brings great damage to the cutting quality on which rare research has been done. This paper took altering a sensitive parameter of the cutter rigidity, the distance between the bears of the shaft as a case to investigate how the enhancement of the cutter rigidity brought the improvement of the cutting quality by applying ADAMS program to perform dynamic simulation of the cutting process after designing their 3D models in Pro/Engineering and making ANSYS analysis of the according effects. Further, a physical experiment following showed that proper design of the distance between the cutter shaft bearings that enhancing 23.67% of its rigidity could lead to 29.20% improvement of the sugarcane cutting quality.

Fuzzy Comprehensive Evaluation on Cleaning Performance of Sugarcane Harvester

The cleaning performance of sugarcane harvester is influenced by many factors. In order to find out the effects on the cleaning performance the orthogonal experiment is integrated with fuzzy comprehensive evaluation to study the remainder ratio of leaf. First, the orthogonal experiment is carried out on different parameter combinations to research the influences of different targets on the cleaning element. Then, the fuzzy comprehensive evaluation method is used to dispose the experiment results. Finally, the evaluation results are combined with the range analysis of orthogonal experiment to determine the optimal parameter combination. Therefore, the multi-target experiment is changed into single-target experiment. The research shows that using fuzzy comprehensive evaluation can get better application effect.

Research on Case Retrieval of Case-Based Reasoning of Motorcycle Intelligent Design

The case retrieval model based on neural network is presented to enhance the efficiency and quality of retrieving case in the case-based reasoning system of the motorcycle intelligent design. In the retrieval model, the adaptive resonance theory neural network was used to dynamically cluster the cases in the case base to narrow the searching range. The back propagation neural network was applied to memory the index of cases to retrieve quickly the similar case from the narrowed case base. Thus the efficiency and quality of retrieving case are improved. Finally, an example of the plan selection of motorcycle general design was given. Its result was contrasted with that of case retrieval based on the nearest neighbor method to demonstrate the effectives of the case retrieval model. The research shows that it is practicable and effective using the adaptive resonance theory and BP neural network to modeling the reasoning mechanism.

Dynamical Simulation Analysis and Experimental Study on a New Sugarcane Cutting Device

The cutting stiffness has notable influence on the sugarcane cutting quality, so great importance should be laid on enhancing the cutting stiffness when designing the cutting device. Based on cutting mechanism analysis, this paper proposes a new design of cutting device to increase the cutting stiffness by properly adding an auxiliary support bar and builds up 3D model of it. Then combining of virtual simulation analysis and physical experiment, it compares and analyzes the different stiffness and cutting quality with support bar in various support position or without it. The result shows that the new design with properly mounted support bar can raise 15.30% of the cutting stiffness, and consequently improve the sugarcane cutting quality distinctly.