基于粒子群優(yōu)化算法.doc
約72頁DOC格式手機打開展開
基于粒子群優(yōu)化算法,摘要近年來，隨著海上船舶數(shù)量的增加和船舶噸位的增大，船舶航行安全問題日益重要，如何保證海上船舶的航行安全是一個迫切需要解決的問題，同時也是許多專家和學(xué)者研究的重點和熱點。合理的船舶避碰方案的確定和船舶碰撞危險度的確定是保證海上船舶航行安全的重要問題，本文首先通過粒子群優(yōu)化算法及改進的兩種粒子群優(yōu)化算法來進行船舶避碰方案...
內(nèi)容介紹
此文檔由會員 違規(guī)屏蔽12 發(fā)布
摘 要
近年來，隨著海上船舶數(shù)量的增加和船舶噸位的增大，船舶航行安全問題日益重要，如何保證海上船舶的航行安全是一個迫切需要解決的問題，同時也是許多專家和學(xué)者研究的重點和熱點。
合理的船舶避碰方案的確定和船舶碰撞危險度的確定是保證海上船舶航行安全的重要問題，本文首先通過粒子群優(yōu)化算法及改進的兩種粒子群優(yōu)化算法來進行船舶避碰方案的確定。然后考慮到船舶碰撞危險度的確定是一個很復(fù)雜的過程，受很多因素的影響，具有很強的非線性特征，本文基于粒子群算法和神經(jīng)網(wǎng)絡(luò)的特點，構(gòu)建了粒子群神經(jīng)網(wǎng)絡(luò)模型，并通過函數(shù)擬合、分類和廣義異或問題進行驗證，最后將粒子群神經(jīng)網(wǎng)絡(luò)模型應(yīng)用到船舶碰撞危險度的確定。
論文的主要研究成果可歸納如下：
(1) 將粒子群優(yōu)化算法、改進的混沌粒子群優(yōu)化算法和免疫粒子群優(yōu)化算法三種算法應(yīng)用到船舶避碰方案的確定上，通過本船分別與一個目標(biāo)船、兩個目標(biāo)船和三個目標(biāo)船形成的各種會遇態(tài)勢進行仿真，并與窮舉法的結(jié)果相比較，證明了這三種算法可以取得比較好的效果，可以應(yīng)用到船舶避碰方案的確定上。
(2) 構(gòu)建了粒子群神經(jīng)網(wǎng)絡(luò)模型。針對神經(jīng)網(wǎng)絡(luò)結(jié)構(gòu)的不確定性和權(quán)閾值的隨機性，本文首先通過混合粒子群優(yōu)化算法來同時確定神經(jīng)網(wǎng)絡(luò)的網(wǎng)絡(luò)結(jié)構(gòu)和權(quán)閾值，然后再通過BP算法進行訓(xùn)練。主要表現(xiàn)在通過二進制粒子群算法確定各個隱含層的神經(jīng)元數(shù)和十進制粒子群優(yōu)化算法確定網(wǎng)絡(luò)的權(quán)閾值。
(3) 通過函數(shù)擬合數(shù)值實驗、Iris花分類、Wine數(shù)據(jù)集分類、LED分類和廣義異或問題，來驗證粒子群神經(jīng)網(wǎng)絡(luò)模型的性能，結(jié)果表明粒子群神經(jīng)網(wǎng)絡(luò)模型可以取得較好的效果。
(4) 將粒子群神經(jīng)網(wǎng)絡(luò)模型應(yīng)用到船舶碰撞危險度的確定上。分別通過具有影響碰撞危險度的兩個因素和六個因素的樣本數(shù)據(jù)進行碰撞危險度的確定，取得了較好的效果。
關(guān)鍵詞 粒子群優(yōu)化算法；神經(jīng)網(wǎng)絡(luò)；避碰幅度；船舶碰撞危險度
Abstract
In recent years, with the number of ships increasing and ships’ weights growing, navigation security issues are increasingly important. How to ensure the safety of ships sailing is an urgent need to resolve. At the same time, many experts and scholars are studying the emphasis.
Reasonable ways to ships collision avoidance and the suitable determines of ships collision avoidance are the guarantee of the safety navigation. Firstly, this thesis gives the ways of ships collision avoidance by particle swarm optimization (PSO) algorithm and other two improved PSO algorithm. Then considering the determination of ships collision risk is a very complex process, and it is affected by many factors. Also, it has a strong feature of the nonlinear. Based on the features of PSO algorithm and the neural network, this thesis constructs the model of neutral network. It is verified through function fitting, classification and general XOR problem. At last, this model is applied to determination of ships collision avoidance.
The main research results in this thesis can be summarized as follows:
(1) The ways of ships collision avoidance are, respectively, determined by PSO algorithm, improved chaotic PSO algorithm and improved immunity PSO algorithm, and the simulation is done with a target ship, two and three goals in various encounter posture. In comparison with the exhaustive law, it is improved that the three algorithms can achieve better results, and it can be applied to determination of ways of ships collision avoidance.
(2) The model of neural network based on PSO algorithm is constructed. Considering the random of the number of the hidden layers and the determination of weight for BP neural network, firstly, this model utilizes hybrid particle swarm optimization to optimize the structure and initial weights for the neural network, and then training by BP. Its performance is mainly through the binary PSO algorithm to determine the threshold of the neutral network and decimal PSO algorithm to determine the neurons number of the hidden layers.
(3) Function fitting, Iris classification, Wine classification, LED classification and general XOR problem are used to verify the performance of neural network based on PSO algorithm. The computing results show that the model can achieve better results.
(4) Determines the ships collision risk through the model of neural network based on PSO algorithm. Simulated by the sample data of two actors and six actors respectively, the results show that the model can achieve better results.
Key Words PSO algorithm; neural network; collision avoidance amplitude; ships collision risk
目 錄
摘 要 I
Abstract III
第1章 緒論 1
1.1 課題的研究背景和現(xiàn)狀 1
1.1.1 船舶避碰的研究背景和現(xiàn)狀 1
1.1.2 選題的背景 3
1.2 課題研究內(nèi)容及主要成果 5
1.3 本文的章節(jié)安排和結(jié)構(gòu) 5
第2章 船舶避碰基礎(chǔ)知識 7
2.1 船舶避碰階段的劃分 7
2.2 船舶避碰過程 8
2.3 船舶會遇局面劃分 10
2.4 安全會遇距離的確定 11
2.5 避碰行動方式 12
2.6 避讓行動時機和幅度 12
2.7 復(fù)航 14
2.8 多船會遇 15
2.9 本章小結(jié) 15
第3章 基于粒子群優(yōu)化算法的船舶避碰決策 17
3.1 粒子群優(yōu)化算法 17
3.1.1 基本粒子群優(yōu)化算法原理 17
3.1.2 基本粒子群優(yōu)化算法描述 17
3.1.3 基本粒子群算法參數(shù)的設(shè)置 18
3.2 混沌粒子群優(yōu)化算法 18
3.3 免疫粒子群優(yōu)化算法 21
3.4 船舶避碰方案的確定 23
3.4.1 船舶..
近年來，隨著海上船舶數(shù)量的增加和船舶噸位的增大，船舶航行安全問題日益重要，如何保證海上船舶的航行安全是一個迫切需要解決的問題，同時也是許多專家和學(xué)者研究的重點和熱點。
合理的船舶避碰方案的確定和船舶碰撞危險度的確定是保證海上船舶航行安全的重要問題，本文首先通過粒子群優(yōu)化算法及改進的兩種粒子群優(yōu)化算法來進行船舶避碰方案的確定。然后考慮到船舶碰撞危險度的確定是一個很復(fù)雜的過程，受很多因素的影響，具有很強的非線性特征，本文基于粒子群算法和神經(jīng)網(wǎng)絡(luò)的特點，構(gòu)建了粒子群神經(jīng)網(wǎng)絡(luò)模型，并通過函數(shù)擬合、分類和廣義異或問題進行驗證，最后將粒子群神經(jīng)網(wǎng)絡(luò)模型應(yīng)用到船舶碰撞危險度的確定。
論文的主要研究成果可歸納如下：
(1) 將粒子群優(yōu)化算法、改進的混沌粒子群優(yōu)化算法和免疫粒子群優(yōu)化算法三種算法應(yīng)用到船舶避碰方案的確定上，通過本船分別與一個目標(biāo)船、兩個目標(biāo)船和三個目標(biāo)船形成的各種會遇態(tài)勢進行仿真，并與窮舉法的結(jié)果相比較，證明了這三種算法可以取得比較好的效果，可以應(yīng)用到船舶避碰方案的確定上。
(2) 構(gòu)建了粒子群神經(jīng)網(wǎng)絡(luò)模型。針對神經(jīng)網(wǎng)絡(luò)結(jié)構(gòu)的不確定性和權(quán)閾值的隨機性，本文首先通過混合粒子群優(yōu)化算法來同時確定神經(jīng)網(wǎng)絡(luò)的網(wǎng)絡(luò)結(jié)構(gòu)和權(quán)閾值，然后再通過BP算法進行訓(xùn)練。主要表現(xiàn)在通過二進制粒子群算法確定各個隱含層的神經(jīng)元數(shù)和十進制粒子群優(yōu)化算法確定網(wǎng)絡(luò)的權(quán)閾值。
(3) 通過函數(shù)擬合數(shù)值實驗、Iris花分類、Wine數(shù)據(jù)集分類、LED分類和廣義異或問題，來驗證粒子群神經(jīng)網(wǎng)絡(luò)模型的性能，結(jié)果表明粒子群神經(jīng)網(wǎng)絡(luò)模型可以取得較好的效果。
(4) 將粒子群神經(jīng)網(wǎng)絡(luò)模型應(yīng)用到船舶碰撞危險度的確定上。分別通過具有影響碰撞危險度的兩個因素和六個因素的樣本數(shù)據(jù)進行碰撞危險度的確定，取得了較好的效果。
關(guān)鍵詞 粒子群優(yōu)化算法；神經(jīng)網(wǎng)絡(luò)；避碰幅度；船舶碰撞危險度
Abstract
In recent years, with the number of ships increasing and ships’ weights growing, navigation security issues are increasingly important. How to ensure the safety of ships sailing is an urgent need to resolve. At the same time, many experts and scholars are studying the emphasis.
Reasonable ways to ships collision avoidance and the suitable determines of ships collision avoidance are the guarantee of the safety navigation. Firstly, this thesis gives the ways of ships collision avoidance by particle swarm optimization (PSO) algorithm and other two improved PSO algorithm. Then considering the determination of ships collision risk is a very complex process, and it is affected by many factors. Also, it has a strong feature of the nonlinear. Based on the features of PSO algorithm and the neural network, this thesis constructs the model of neutral network. It is verified through function fitting, classification and general XOR problem. At last, this model is applied to determination of ships collision avoidance.
The main research results in this thesis can be summarized as follows:
(1) The ways of ships collision avoidance are, respectively, determined by PSO algorithm, improved chaotic PSO algorithm and improved immunity PSO algorithm, and the simulation is done with a target ship, two and three goals in various encounter posture. In comparison with the exhaustive law, it is improved that the three algorithms can achieve better results, and it can be applied to determination of ways of ships collision avoidance.
(2) The model of neural network based on PSO algorithm is constructed. Considering the random of the number of the hidden layers and the determination of weight for BP neural network, firstly, this model utilizes hybrid particle swarm optimization to optimize the structure and initial weights for the neural network, and then training by BP. Its performance is mainly through the binary PSO algorithm to determine the threshold of the neutral network and decimal PSO algorithm to determine the neurons number of the hidden layers.
(3) Function fitting, Iris classification, Wine classification, LED classification and general XOR problem are used to verify the performance of neural network based on PSO algorithm. The computing results show that the model can achieve better results.
(4) Determines the ships collision risk through the model of neural network based on PSO algorithm. Simulated by the sample data of two actors and six actors respectively, the results show that the model can achieve better results.
Key Words PSO algorithm; neural network; collision avoidance amplitude; ships collision risk
目 錄
摘 要 I
Abstract III
第1章 緒論 1
1.1 課題的研究背景和現(xiàn)狀 1
1.1.1 船舶避碰的研究背景和現(xiàn)狀 1
1.1.2 選題的背景 3
1.2 課題研究內(nèi)容及主要成果 5
1.3 本文的章節(jié)安排和結(jié)構(gòu) 5
第2章 船舶避碰基礎(chǔ)知識 7
2.1 船舶避碰階段的劃分 7
2.2 船舶避碰過程 8
2.3 船舶會遇局面劃分 10
2.4 安全會遇距離的確定 11
2.5 避碰行動方式 12
2.6 避讓行動時機和幅度 12
2.7 復(fù)航 14
2.8 多船會遇 15
2.9 本章小結(jié) 15
第3章 基于粒子群優(yōu)化算法的船舶避碰決策 17
3.1 粒子群優(yōu)化算法 17
3.1.1 基本粒子群優(yōu)化算法原理 17
3.1.2 基本粒子群優(yōu)化算法描述 17
3.1.3 基本粒子群算法參數(shù)的設(shè)置 18
3.2 混沌粒子群優(yōu)化算法 18
3.3 免疫粒子群優(yōu)化算法 21
3.4 船舶避碰方案的確定 23
3.4.1 船舶..
