论文部分内容阅读
本文以在浙江省宁波市镇海区进行的１２年棉红铃虫（ＰｅｃｔｉｎｏｐｈｏｒａｇｏｓｓｙｐｉｅｌｌａＳａｕｎｄｅｒｓ）性信息素诱蛾试验为基础，通过相关分析，选取第一代蛾高峰期前累计诱蛾，棉花现蕾期和６月中旬到７月上旬的降水量３个参数，构建对第二代平均卵量进行综合评判的模型。由于多种因子对诱蛾量的影响具有不确定性和模糊性，采用ＦＵＺＺＹ综合评判模型建模。以回报准确率为标准，对９种权重、４类评判矩阵和６种广义算子以及它们的特性作了比较和讨论。在构建的２１６个模型中，有９４个模型的历史符合率达到１００％。在这些模型的基础上，得到了一个优化评判模型。该模型与常规线性回归模型比较具有准确、方便、所含信息多等待点。在评判模型的基础上，增加第一次发现卵量的日期，构建了预报模型，该模型加强了预报的功能。模拟运行，可得到“预报速查表”；预测结果与大田相符。该研究为性信息素和其它行为调节物质在测报上的应用以及动态经济阈值的研究提供了一些新的思路。
Based on the 12-year-old sex pheromone trapping test of Pectinophora gossypiella Saunders in Zhenhai District, Ningbo City, Zhejiang Province, the correlation analysis was used to select the first-generation cumulative moth, cotton budding stage and June to mid-July rainfall in early July three parameters to build a second generation of the average number of eggs to conduct a comprehensive evaluation of the model. Due to the uncertainty and fuzziness of the influence of many factors on the amount of trapping moths, the FUZZY comprehensive evaluation model was used for modeling. With the accuracy of return as the standard, 9 kinds of weights, 4 types of evaluation matrix and 6 kinds of generalized operators and their characteristics are compared and discussed. Of the 216 models that were constructed, 94 models met historical rates of 100%. Based on these models, an optimal evaluation model is obtained. Compared with the conventional linear regression model, the model is accurate, convenient and contains more information waiting points. Based on the evaluation model, the date of first finding of eggs was increased, and a forecasting model was constructed, which enhanced the forecasting function. Simulation run, available “forecast table”; forecast results in line with the field. The research provided some new ideas for the application of sex pheromone and other behavioral regulators in the field of forecasting and the study of dynamic economic threshold.