發(fā)布時間：2018-05-30 12:12

  本文選題：缽苗移栽機 + 末端執(zhí)行器�。� 參考：《浙江大學(xué)》2017年博士論文

【摘要】：工廠化育苗技術(shù)因具有生產(chǎn)效率高、成本低、可連續(xù)生產(chǎn)等諸多優(yōu)點,已在我國得到了普遍推廣和應(yīng)用。隨著育苗技術(shù)的發(fā)展,溫室自動化移栽設(shè)備的需求也越來越大。針對國內(nèi)溫室缽苗移栽裝備作業(yè)效率低和可靠性差的問題,以浙江大學(xué)原有移栽設(shè)備為基礎(chǔ),本文首先進行了補苗移栽策略的改進、苗盤分送裝置的設(shè)計、機械間苗裝置的設(shè)計、末端執(zhí)行器的優(yōu)化設(shè)計及測試等研究,從硬件系統(tǒng)方面提高移栽系統(tǒng)的可靠性和效率,接著基于設(shè)計的末端執(zhí)行器搭建實時測力系統(tǒng),分析基質(zhì)含水率、基質(zhì)體積配比、單孔基質(zhì)量等育苗農(nóng)藝參數(shù)對移栽作業(yè)過程中拉拔力和夾緊力的影響,探討夾緊力和拉拔力與移栽效果之間的關(guān)系,為末端執(zhí)行器的安裝調(diào)試提供理論指導(dǎo),進一步提高移栽作業(yè)的可靠性,最后采用蟻群算法對末端執(zhí)行器補苗移栽作業(yè)的路徑進行規(guī)劃,通過縮短補苗路徑,在不增加硬件成本的前提下進一步提高移栽作業(yè)的效率。主要研究內(nèi)容和結(jié)論如下:(1)提出了新的補苗策略,并且設(shè)計了一套苗盤分送裝置及控制系統(tǒng),可以根據(jù)穴盤內(nèi)缺苗數(shù)的多少對穴盤進行歸類,減少了補苗作業(yè)的工作量,提高了補苗作業(yè)的效率。設(shè)計了一種穴盤機械分隔裝置及控制系統(tǒng),裝置的列分隔組件的核心為一組卷曲布簾,解決了原有裝置傷苗的問題,增加了設(shè)備的可靠性。(2)設(shè)計了一種單氣缸驅(qū)動的插拔夾取式末端執(zhí)行器,不僅保證了缽苗的主動分離,還可以在氣缸的一個推程內(nèi)連續(xù)完成插入并夾緊育苗基質(zhì)的動作。根據(jù)指針尖點的運動軌跡,基于編寫的VB程序?qū)δ┒藞?zhí)行器的關(guān)鍵結(jié)構(gòu)尺寸進行了綜合確定。移栽測試結(jié)果顯示:設(shè)計的末端執(zhí)行器對不同育苗農(nóng)藝條件下Q72穴盤中黃瓜苗的移栽成功率達到100%,基質(zhì)塊的平均損傷率為17%,且損傷程度均較小;對Q72穴盤中番茄、辣椒缽苗的移栽成功率同樣達到100%,基質(zhì)塊的嚴重損傷率分別為2.5%和4.5%;對Q128穴盤中黃瓜苗的移栽成功率也達到100%,基質(zhì)塊的損傷率為23.8%。該末端執(zhí)行器對育苗農(nóng)藝、作物品種和穴盤規(guī)格都具有較強的適應(yīng)性,且在保證缽苗基質(zhì)塊完整性方面具有較大的突破。與原有第三代末端執(zhí)行器相比,該末端執(zhí)行器的可靠性和效率均有較大提高。(3)基于設(shè)計的末端執(zhí)行器搭建了一套實時測力系統(tǒng),其核心部件為兩個高精度、快速響應(yīng)的力學(xué)傳感器,可實時測量移栽過程中的拉拔力和夾緊力。通過力學(xué)分析和幾何關(guān)系分別建立了拉拔力和夾緊力的計算方式,并采用萬能試驗機對傳感器進行了標定。為驗證該實時測力系統(tǒng)的有效性,進行了移栽試驗并分析夾緊力和拉拔力隨育苗基質(zhì)含水率的變化。試驗共得到6個梯度的含水率,且結(jié)果表明,拉拔力隨含水率的降低先增大后降低,夾緊力隨含水率的降低而增大。含水率76%左右時,同試驗組內(nèi)拉拔力的標準偏差較小;含水率74%左右時,同試驗組內(nèi)夾緊力的標準偏差較小。(4)通過三因素三水平的全面交互試驗研究3個主要農(nóng)藝育苗參數(shù)(基質(zhì)含水率、基質(zhì)體積配比和單孔基質(zhì)量)對移栽作業(yè)過程中夾緊力和拉拔力的影響,并探討了夾緊力、拉拔力與移栽效果之間的關(guān)系。試驗中共有9盤缽苗,每盤缽苗被分為3個區(qū)域,每個區(qū)域為1個試驗組,一共27組。每個試驗組隨機選取10株缽苗進行重復(fù)移栽試驗。采用多因素方差分析法研究了 3個主要農(nóng)藝育苗參數(shù)對夾緊力和拉拔力影響的顯著性。對夾緊力的分析表明,基質(zhì)含水率、基質(zhì)體積配比和單孔基質(zhì)量對夾緊力的影響均十分顯著(P0.05),且影響的顯著性依次為:基質(zhì)含水率單孔基質(zhì)量基質(zhì)體積配比。3個育苗參數(shù)兩兩之間的交互作用以及3個參數(shù)之間的交互作用對夾緊力的影響均十分顯著(P0.05)。其中,基質(zhì)含水率與基質(zhì)體積配比、基質(zhì)含水率與單孔基質(zhì)量之間的交互作用最為顯著。對拉拔力的分析表明,基質(zhì)含水率、基質(zhì)體積配比和單孔基質(zhì)量對拉拔力的影響均十分顯著(P0.05),且影響的顯著性依次為:單孔基質(zhì)量基質(zhì)含水率基質(zhì)體積配比。只有基質(zhì)體積配比與單孔基質(zhì)量之間的交互作用對拉拔力有顯著影響(P0.05),其余參數(shù)之間的交互作用均不顯著。進一步分析表明,基質(zhì)含水率在82.5%-78.5%范圍內(nèi)逐漸下降時,夾緊力和拉拔力的均值呈線性增長。基質(zhì)體積配比為7:2:1水平時,夾緊力的均值最大;基質(zhì)體積配比為6:2:2水平時,拉拔力的均值最大。夾緊力均值隨著單孔基質(zhì)量的增加先增大后減小,拉拔力均值隨著單孔基質(zhì)量的增加而增大,但是增大的趨勢變緩。結(jié)合移栽效果分析發(fā)現(xiàn),移栽效果的好壞并不直接與夾緊力或者拉拔力相關(guān)。進一步分析夾緊力與拉拔力的和、差、乘積、比值,發(fā)現(xiàn)移栽效果與夾緊力和拉拔力的比值之間存在一定的關(guān)系。當(dāng)夾緊力與拉拔力的比值在5.99～8.67的范圍內(nèi)時,移栽效果相比會更好。該試驗結(jié)果可以為末端執(zhí)行器的設(shè)計、安裝和調(diào)節(jié)提供指導(dǎo),提高末端執(zhí)行器移栽的可靠性和效率。(5)基于蟻群算法實現(xiàn)了末端執(zhí)行器補苗移栽作業(yè)的路徑規(guī)劃。根據(jù)輸出路徑的長度,通過交互試驗確定算法的關(guān)鍵參數(shù)取值,遺傳算法中取pc=0.8、pm = 0.3,蟻群算法中取α=1,β=20,ρ=0.]。與遺傳算法相比,蟻群算法輸出結(jié)果的收斂速度較慢,且波動較大,但蟻群算法第一次迭代運行后的輸出結(jié)果要遠遠優(yōu)于遺傳算法。對10種隨機的穴盤狀態(tài)進行路徑規(guī)劃,遺傳算法和蟻群算法的平均運行時間分別為0.32 s和0.94 s。相比于傳統(tǒng)的CSM(Common Sequence Method,固定順序掃描法)方法,遺傳算法和蟻群算法都可以顯著縮短補苗路徑的長度。然而遺傳算法由于算法本身的局限性,不能處理只有]個空穴的情況。通常情況下,蟻群算法輸出結(jié)果的標準差要小于遺傳算法。待補穴盤中空穴數(shù)量和供苗穴盤中健康缽苗的數(shù)量之和(p+q)對遺傳算法和蟻群算法優(yōu)化效果的相對優(yōu)劣影響顯著,其和越大,蟻群算法的優(yōu)勢越明顯。使用20×10穴盤進行優(yōu)化效果對比,蟻群算法完成40個空穴的補苗路徑比遺傳算法的路徑短6000.9 mm(33.2%),比傳統(tǒng)的CSM方法短13058.4mm(51.9%)。整體比較發(fā)現(xiàn),蟻群算法更適合用于末端執(zhí)行器補苗移栽作業(yè)的路徑規(guī)劃,可在不增加硬件成本的條件下進一步提高設(shè)備的作業(yè)效率。
[Abstract]:Because of the advantages of high production efficiency, low cost and continuous production, the plant cultivation technology has been widely popularized and applied in China. With the development of the seedling raising technology, the demand for the automatic transplanting equipment of greenhouse is becoming more and more large. The problem of low efficiency and poor reliability of the domestic greenhouse seedling transplanting equipment is Zhejiang big. On the basis of the original transplanting equipment, this paper first carried out the improvement of the strategy of replanting and transplanting, the design of the seedling plate distribution device, the design of the mechanical thinning device, the optimization design and testing of the end effector, and the improvement of the reliability and efficiency of the transplanting system from the hardware system, and then built the real-time measurement based on the designed terminal actuator. The relationship between the clamping force and the pulling force and the effect of the transplanting is discussed. The relationship between the clamping force and the pulling force and the effect of the transplanting is discussed, and the theoretical guidance for the installation and debugging of the terminal actuator is provided, and the reliability of the transplanting operation is further improved. An ant colony algorithm is used to plan the path of the transplant operation of the end effector. By shortening the route of the supplement, the efficiency of the transplanting operation is further improved without increasing the cost of hardware. The main contents and conclusions are as follows: (1) a new strategy is put forward, and a set of seedling disk distribution device and control system are designed. In order to classify the hole disc according to the number of the holes in the hole, it reduces the workload of the remedial work and improves the efficiency of the remedial work. A device and control system are designed. The core of the column dividing component is a set of curled cloth curtains, which solves the problem of the original plant injury and increases the reliability of the equipment. 2) a single cylinder driven plug and fetching end actuator is designed, which not only ensures the active separation of the bowl seedlings, but also can continuously insert and clamp the nursery matrix in a push range of the cylinder. Based on the moving trajectory of the pointer point, the key structure size of the terminal executor is synthesized based on the written VB program. The results of transplanting test showed that the successful rate of transplant success of the designed end actuators to the cucumber seedlings of Q72 acupoints under different cultivation conditions was 100%, the average damage rate of the matrix block was 17%, and the damage degree was small; the successful rate of transplant success of Tomato in Q72 and the serious damage rate of the matrix block was respectively 100%. For 2.5% and 4.5%, the successful rate of transplant success of Cucumber Seedlings in Q128 is 100%. The damage rate of matrix block is 23.8%., the end actuator has strong adaptability to seedling farming, crop variety and point plate specification, and it has a great breakthrough in guaranteeing the integrity of the matrix block. Compared with the original third generation end actuator, The reliability and efficiency of the terminal actuator have been greatly improved. (3) a set of real-time force measuring system is set up based on the design - based terminal actuator. The core component is two high precision and fast response mechanical sensors, which can measure the pulling force and clamping force in the transplanting process in real time. Through the mechanical analysis and the geometric relationship, the pulling force is established respectively. In order to verify the effectiveness of the real time force measuring system, a transplanting test was carried out and the change of the water content of the clamping force and drawing force with the seedling matrix was analyzed to verify the effectiveness of the real time force measuring system. The test results showed that 6 gradient water content was obtained in the test, and the results showed that the drawing force increased with the decrease of water content first. When the water content is about 76%, the standard deviation of drawing force in the test group is smaller when the water content is about 76%. When the water content is about 74%, the standard deviation of the clamping force is smaller in the test group. (4) through the comprehensive interactive test of three factors and three levels, the parameters of the 3 main agronomic seedlings (matrix water content, matrix body) are studied. The influence of product ratio and single hole base mass on the clamping force and pulling force in the process of transplanting, and the relationship between the clamping force, drawing force and the effect of transplanting are discussed. There are 9 plates of bowl seedlings in the experiment. Each bowl is divided into 3 regions, each area is 1 test groups and a total of 27 groups. 10 seedlings of each test group are randomly selected to repeat the shift. The effects of 3 main agronomic parameters on the clamping force and pulling force are studied by the multi factor variance analysis. The analysis of the clamping force shows that the influence of matrix water content, matrix volume ratio and single hole base mass on the clamping force is very significant (P0.05), and the significance of the influence is the matrix water content single hole in turn. The interaction between the base mass matrix volume ratio.3 breeding parameters 22 and the interaction between the 3 parameters and the interaction effect on the clamping force are all very significant (P0.05). Among them, the interaction between matrix water content and matrix volume, the interaction between the matrix water content and the single pore base mass is the most significant. The analysis of drawing force shows that matrix The effect of water content, matrix volume ratio and single hole base mass on drawing force is very significant (P0.05), and the significant effect is the matrix volume ratio of the mass matrix of single hole base mass. Only the interaction between the matrix volume ratio and the mass of single pore base has significant influence on drawing force (P0.05), and the interaction between the other parameters is made. It is not significant for use. Further analysis shows that the mean of clamping force and drawing force increases linearly when the matrix moisture content decreases gradually within the range of 82.5%-78.5%. When the matrix volume ratio is 7:2:1, the mean of the clamping force is the largest. When the matrix volume ratio is 6:2:2 level, the mean of drawing force is the largest. The mean of clamping force is with the base mass of the single hole. The increase first increases and then decreases, and the Rabari mean increases with the increase of the mass of the single hole base, but the increasing trend slows down. It is found that the effect of transplanting is not directly related to the clamping force or drawing force. Further analysis of the effect of the clamping force and drawing force, the difference, the product and the ratio, and the discovery of the effect of transplanting and the clamping force There is a certain relationship between the ratio of pulling force and the drawing force. When the ratio of the clamping force to drawing force is within the range of 5.99 ~ 8.67, the effect of transplanting is better. The result of the experiment can provide guidance for the design, installation and adjustment of the end actuators, and improve the reliability and efficiency of the transplanting of the end effector. (5) the end of the experiment is based on the ant colony algorithm. According to the length of the output path, according to the length of the output path, the key parameters of the algorithm are determined by the interactive experiment. Pc=0.8, PM = 0.3 in the genetic algorithm are taken. The ant colony algorithm takes the alpha =1, beta =20, and P =0.]. as the genetic algorithm, and the ant colony algorithm has a slower convergence rate and a larger fluctuation, but the ant colony algorithm is the first one. The output results after one iteration are far superior to the genetic algorithm. The path planning of 10 random cavern States is carried out. The average running time of the genetic algorithm and the ant colony algorithm is 0.32 s and 0.94 S., respectively, compared to the traditional CSM (Common Sequence Method, fixed sequence scanning) method, and the genetic algorithm and the ant colony algorithm are all obvious. However, due to the limitations of the algorithm itself, the genetic algorithm can not handle only a hole. In general, the standard deviation of the output results of the ant colony algorithm is less than that of the genetic algorithm. The number of holes in the pending hole and the sum of the health bowl seedlings in the supply of the hole discs (p+q) are the genetic algorithm and the ant colony algorithm The relative advantages and disadvantages of the optimization effect are significant, the more obvious it is, the more obvious the advantage of the ant colony algorithm is. Using the 20 x 10 hole disk to make the optimization effect comparison, the ant colony algorithm completes the 40 hole filling path shorter than the genetic algorithm's path is 6000.9 mm (33.2%), and is shorter than the traditional CSM method (51.9%). The overall comparison shows that the ant colony algorithm is more suitable for use. The path planning of the end effector seedling transplant operation can further improve the operation efficiency of the equipment without increasing the cost of hardware.
【學(xué)位授予單位】：浙江大學(xué)
【學(xué)位級別】：博士
【學(xué)位授予年份】：2017
【分類號】：S223.9

【相似文獻】

相關(guān)期刊論文 前10條

1 鮑官軍;高峰;荀一;都明宇;楊慶華;;氣動柔性末端執(zhí)行器設(shè)計及其抓持模型研究[J];農(nóng)業(yè)工程學(xué)報;2009年10期

2 楊慶華;金寅德;錢少明;鮑官軍;;基于氣動柔性驅(qū)動器的蘋果采摘末端執(zhí)行器研究[J];農(nóng)業(yè)機械學(xué)報;2010年09期

3 馮良寶;姬長英;田光兆;黃浩乾;;基于穩(wěn)定抓取力學(xué)模型的末端執(zhí)行器控制器設(shè)計[J];江西農(nóng)業(yè)學(xué)報;2010年07期

4 孫國祥;汪小e，

本文編號：1955259