Genetic Algorithm In Forest Harvest Adjustment Application As Well As Software Research

In the knowledge-based system, especially in the expert system, the acquisition of knowledge is a very difficult issue, and it has been recognized as a "bottleneck" . In this paper, I systematically introduced the theory and application technology of knowledge acquisition based on artificial neural network technology and proposed an application method in forest management decision-making. Then, in the MATLAB software environment, the application and theory proposed in the article was verified and the self-learning knowledge acquisition process based on the neural network was achieved.By using knowledge acquisition techniques, according to the forest resources for the inventory data, I established the neural network forecast model of the category of forest land types and the forest future harvest (including the age group structure prediction and the prediction of forest harvest). Also ,I tested the model. Test results showed that using the artificial neural network method to forecast the dynamic change of forest land types and the forest harvest can meet the accuracy requirements of forestry production and operation. That the knowledge acquisition based on artificial neural network is feasible. According to the knowledge acquired,I predicted and analyzed the dynamic change trends of forest land types in Jianyang of Fujian and the development trends of Chinese fir and the forest harvest in Fujian. To find the key to the development and make a scientific decision, which is of great practical significance to the maintenance of ecological balance and rational development of forestry production.In this paper, using artificial neural network to achieve the automatic acquisition of knowledge, no longer needed by the knowledge engineers to organize, summarize and digest the knowledge of experts in the field, only needed the problem-solving examples of experts in the field to train the neural network, so that in the same input, artificial neural network could get the same output given by experts as much as possible.