Research On Tomato Fruit Development And Yield Formation Model Under Facility Condition

Research of crop imitation model is a new and dynamic interdiscipline. Every subject's generation has its own dramatic significance as well as crop imitation model which is the base of precisely agricultural management and research of agricultural expert system. There is no doubt that it will promote the agricultural modernization. This thesis investigated the disipline of tomato fruit growth and development and analyzed the relationship between fruit growth and environmental factors. In additon to this, we have established three yield formation models based on product of temperature and light and another one based on time of physiological development. As a result of good validation, concise and practical feature, the models are able to be applied to precisely agricultural management as a technical support. In the light of the thesis, some conclusions can be made as follows:1. The development rule of both transverse and veritcal diameter of tomato was accelerating first and then decelerating under spring growing season in plastic shed as well as autumn growing season. It appeared a apparent"S"curve of tomato volume and fresh weight development under spring growing season in plastic shed. However, there was straight line trend in autumn solar greenhouse.There was a uniformation growth speed of fruit from 1 to 5 tassel at spring crop and a gradually declination from 1 to 4 tassel at autumn crop. The relationship among transverse and vertical diameter to fresh weight was quantity in the exponent and the connection bteween volum and fresh weight is straight line.2. The changing trend of 7 variables under two palstic sheds in spring is similar. The envrionmental factors'change under solar greenhouse in autumn was that average temperature of air declined gragually, the average temperature of soil was higher than air and the change of average temperature difference was not significant. There was a greatly mutual reliance among 7 variables. The fruit growth rate of Jin Peng and Fen Guan variety displayed an acceleration first and then deceleration.The fruit growth rate of Zhen Qi variety fluctuated greatly.3. In the three yield formation models per plant, the yield increased following accumulation of produt of temperature and light and appeared a quantity in the exponent. And the regression coefficient approached to 1, which meant better relativity. The regression coefficient of three validation lines were 0.9782, 0.9533 and 0.9800 respectively, which meant that the actual value and predicted value were similar and the difference between them was small. The value of RMSE were all small, which meant that three models had high accuracy and could be used to simulate yield per plant.4. The distribution of tomato dry matter showed that the overground part was higher than underground part. The distribution in nutritorium decreased and the distribution in generative organ increased gradually and finally came to a certain value. We adopted time of physiology as independent variable t and yield per plant as induced variable W to simulate the two by means of Logistic regression in DPS. The prospective accuracy and correctness was good and outdoor validation of this model was carried out and the result was better.