本文關(guān)鍵詞： 脈沖害蟲控制 優(yōu)化算法 梯度 時(shí)間縮放變換 混雜優(yōu)化問題 時(shí)滯　出處：《天津工業(yè)大學(xué)》2017年碩士論文　論文類型：學(xué)位論文

【摘要】：害蟲控制在農(nóng)業(yè)生產(chǎn)中具有重要作用,而微分方程數(shù)學(xué)模型在描述害蟲種群動(dòng)力學(xué)行為中起到了非常重要的作用,特別是用脈沖微分方程來描述害蟲種群動(dòng)力學(xué)模型能夠更合理、更精確地反映各種變化規(guī)律,因?yàn)楝F(xiàn)實(shí)世界中害蟲的繁殖和危害、以及人類的控制行為幾乎都是階段性的。最優(yōu)脈沖控制問題一直是一個(gè)熱點(diǎn)研究課題,脈沖控制將化學(xué)防治和生物防治方法結(jié)合到一塊實(shí)現(xiàn)了合理的綜合害蟲治理目的。本文對害蟲治理中的變量優(yōu)化和時(shí)滯優(yōu)化問題進(jìn)行了深入分析和研究。第二章節(jié)我們研究了害蟲增長模型在終端時(shí)刻害蟲數(shù)量最少的最優(yōu)脈沖管理問題。在優(yōu)化過程中,通過時(shí)間縮放和時(shí)間平移技巧,給出了三種情況下目標(biāo)函數(shù)關(guān)于天敵最優(yōu)釋放量和釋放時(shí)間的梯度:優(yōu)化確定釋放時(shí)刻下相同的天敵最優(yōu)釋放量(數(shù)量控制);優(yōu)化天敵的最優(yōu)釋放時(shí)間和相等但卻不確定的最優(yōu)釋放量(時(shí)間控制);優(yōu)化天敵的最優(yōu)釋放時(shí)間和非相等的最優(yōu)釋放量(混合控制)�；谔荻裙胶驮O(shè)計(jì)算法,我們利用數(shù)值模擬找到了天敵的最優(yōu)釋放時(shí)間和釋放量,通過比較顯示混合控制策略效果最好,另外證明了優(yōu)化管理策略可能增加種群的周期震蕩幅度,但最終減少了害蟲的數(shù)量并降低了成本。而且,我們的最優(yōu)生物干擾策略優(yōu)于許多文獻(xiàn)中在確定時(shí)刻釋放確定數(shù)量天敵的策略。第三章節(jié)考慮到目前尚未把寄生卵的滯后寄生、殺蟲劑毒殺作用的滯后效應(yīng)以及投放天敵的年齡等時(shí)滯因素看作控制變量,即狀態(tài)時(shí)滯,結(jié)合農(nóng)作物的生育周期,研究農(nóng)作物在幼苗期、花蕾期、灌漿期等生育期的多特征時(shí)間段上的害蟲控制模型的優(yōu)化問題,我們首先提出了一類具有特征時(shí)間的狀態(tài)時(shí)滯和參數(shù)控制的混雜優(yōu)化問題,接著給出了涉及的求解優(yōu)化算法,最后舉出兩個(gè)害蟲治理領(lǐng)域的例子說明優(yōu)化算法的有效性。
[Abstract]:Pest control plays an important role in agricultural production, and the mathematical model of differential equation plays a very important role in describing the behavior of pest population dynamics. In particular, the use of impulsive differential equations to describe pest population dynamics model can be more reasonable and more accurate to reflect the laws of change, because of the propagation and damage of pests in the real world. The optimal pulse control problem is always a hot research topic. Pulse control combines chemical and biological control methods to achieve a reasonable goal of integrated pest control. In this paper, variable optimization and time-delay optimization in pest management are deeply analyzed and studied. In this paper, we study the optimal impulse management problem of the pest growth model with the least number of pests at the end time. Based on the techniques of time scaling and time translation, the optimal release amount and release time gradient of the objective function are given in three cases: the optimal release amount of the same natural enemy (quantity control) is determined by optimizing the release time; Optimizing the optimal release time of the natural enemy and the equal but uncertain optimal release amount (time control); The optimal release time and non-equal optimal release time of natural enemy are optimized. Based on gradient formula and design algorithm, we find the optimal release time and amount of natural enemy by numerical simulation. The comparison shows that the hybrid control strategy has the best effect. In addition, it is proved that the optimal management strategy may increase the periodic fluctuation of the population, but ultimately reduce the number of pests and reduce the cost. Our optimal biological interference strategy is better than that of releasing a certain number of natural enemies at a given time in many literatures. Chapter 3 takes into account that the lagging parasitic eggs have not yet been parasitized. The lag effect of pesticide poisoning and the time delay factors such as the age when the natural enemy was put into the plant were regarded as the control variables, i.e., the state delay. Combining with the growth period of crops, the crops were studied in the seedling stage and the flowering stage. The optimization problem of pest control model in multiple characteristic periods of grain filling stage is discussed. Firstly, we propose a hybrid optimization problem of state delay and parameter control with characteristic time. Finally, two examples in the field of pest control are given to illustrate the effectiveness of the optimization algorithm.
【學(xué)位授予單位】：天津工業(yè)大學(xué)
【學(xué)位級別】：碩士
【學(xué)位授予年份】：2017
【分類號(hào)】：O175;O232

【參考文獻(xiàn)】

相關(guān)期刊論文 前1條

1 師向云;宋新宇;;一類狀態(tài)依賴脈沖控制的害蟲管理數(shù)學(xué)模型研究[J];系統(tǒng)科學(xué)與數(shù)學(xué);2012年07期

，

本文編號(hào)：1472607