这是一个包含基本PSO算法和改进PSO算法的程序，包括带压缩因子的PSO，权重改进的PSO，变学习因子PSO，混沌PSO等(This is one that contains the basic PSO algorithm and PSO algorithm to improve the procedures, including with a compression factor of the PSO, the weight of the improved PSO, variable learning factor PSO, PSO and other chaos)

从网上下载了一个GA－PSO算法，试着利用GA和PSO组合的策略进行优化，结果算法很问题，效率和不错。我下载原始算法，有一个问题就是它是针对所有的设计变量上下限都是一样的，所以我对程序进行了修改与改进，现在可以处理上下限不一致的问题，同时fix了一些bug。(GA GA and PSO algorithm matlab program combined ion group When doing optimization, first choose the GA algorithm, but the instability of the GA (or into a local optimum) it drives people crazy, even after twice the difference results can sometimes reach 30 or more, but because of time reason, I had to choose the optimal 1000, then the minimum of which the most optimal solution. Although the problem is considered solved, but will be from an academic, which makes Ben is very embarrassing. So, I downloaded a GA-PSO algorithm, try to use a combination of GA and PSO optimization strategy, the results of algorithm is the problem, efficiency and good. I downloaded the original algorithm, there is a problem is that it is the design variables for all upper and lower limits are the same, so I had to modify and improve the program, can now handle upper and lower limits inconsistencies, and fix some bug.)

粒子群算法，也称粒子群优化算法（Particle Swarm Optimization），缩写为 PSO， 是近年来发展起来的一种新的进化算法(Evolutionary Algorithm - EA)。PSO 算法属于进化算法的一种，和模拟退火算法相似，它也是从随机解出发，通过迭代寻找最优解，它也是通过适应度来评价解的品质，但它比遗传算法规则更为简单，它没有遗传算法的“交叉”(Crossover) 和“变异”(Mutation) 操作，它通过追随当前搜索到的最优值来寻找全局最优。这种算法以其实现容易、精度高、收敛快等优点引起了学术界的重视，并且在解决实际问题中展示了其优越性。粒子群算法是一种并行算法。 BP（Back Propagation）神经网络是1986年由Rumelhart和McCelland为首的科学家小组提出，是一种按误差逆传播算法训练的多层前馈网络，是目前应用最广泛的神经网络模型之一。BP网络能学习和存贮大量的输入-输出模式映射关系，而无需事前揭示描述这种映射关系的数学方程。它的学习规则是使用最速下降法，通过反向传播来不断调整网络的权值和阈值，使网络的误差平方和最小。BP神经网络模型拓扑结构包括输入层（input）、隐层(hidden layer)和输出层(output layer)。(Particle swarm optimization, also known as particle swarm optimization (Particle Swarm Optimization), abbreviated as PSO, is a new evolutionary algorithm developed in recent years (Evolutionary Algorithm- EA). Kind, and simulated annealing algorithm PSO algorithm is similar evolutionary algorithms, it is also starting a random solution, through an iterative search for the optimal solution, which is also used to uate the quality through fitness solution, but it is simpler than genetic algorithm rules It has no genetic algorithm " crossover" (Crossover) and " variant" (Mutation) operation, which by following the current search to find the optimal value to the global optimum. This algorithm is its easy implementation, high accuracy, fast convergence, etc. attracted academic attention and show its superiority in solving practical problems. PSO algorithm is a parallel algorithm. BP (Back Propagation) neural network is a 1986 team of scientists headed by Rumelhart and McC)

各种粒子群或改进型粒子群算法 1）粒子群优化算法(求解无约束优化问题) 1>PSO(基本粒子群算法) 2>YSPSO(待压缩因子的粒子群算法) 3>LinWPSO(线性递减权重粒子群优化算法) 4>SAPSO(自适应权重粒子群优化算法) 5>RandWSPO(随机权重粒子群优化算法) 6>LnCPSO(同步变化的学习因子) 7>AsyLnCPSO(异步变化的学习因子)(算法还有bug) 8>SecPSO(用二阶粒子群优化算法求解无约束优化问题) 9>SecVibratPSO(用二阶振荡粒子群优化算法求解五约束优化问题) 10>CLSPSO(用混沌群粒子优化算法求解无约束优化问题) 11>SelPSO(基于选择的粒子群优化算法) 12>BreedPSO(基于交叉遗传的粒子群优化算法) 13>SimuAPSO(基于模拟退火的粒子群优化算法) (Various PSO or improved particle swarm algorithm 1) Particle Swarm Optimization (unconstrained optimization problems) 1> PSO (PSO) 2> YSPSO (PSO algorithm to be compression factor) 3> LinWPSO (linearly decreasing weight particle swarm optimization) 4> SAPSO (adaptive weight particle swarm optimization) 5> RandWSPO (random weight particle swarm optimization) 6> LnCPSO (synchronous changes in learning factor) 7> AsyLnCPSO (asynchronous learning factor changes) (algorithm also bug) 8> SecPSO (particle swarm optimization algorithm with a second-order problem solving unconstrained optimization) 9> SecVibratPSO (with a second-order oscillating particle swarm optimization algorithm for solving constrained optimization problems five) 10> CLSPSO (chaotic particle swarm optimization algorithm for solving unconstrained optimization problems) 11> SelPSO (selection-based particle swarm optimization) 12> BreedPSO (genetic optimization algorithm based )

基于粒子群优化算法(PSO)确定参数的脉冲欧和神经网络滤波方法。利用粒子群优化算法(PSO)确定pcnn图像滤波参数，对图像进行滤波(Based on Particle Swarm Optimization (PSO) to determine the parameters of the pulse in Europe and neural network filtering method. The use of particle swarm optimization (PSO) to determine PCNN parameters of image filtering, image filtering)

PSO与GA混合优化算法，可实现比单纯使用PSO或者GA算法更好的优化效果(PSO and GA hybrid optimization algorithm, can achieve more than a simple algorithm using PSO or GA optimization results better)

用PSO优化LSSVM的参数，得到最优参数，使得分类更加准确。(Using the PSO the LSSVM the parameter, the optimal parameters, more accurate classification. )

一个利用PSO粒子群优化算法训练BP神经网络的程序，在matlab环境中进行操作，代码简便可行。(The use of a PSO particle swarm optimization algorithm to train BP neural network procedures, operating in the MATLAB environment, the code is simple and feasible.)

两个文件 主程序 微粒群优化的神经网络 子程序 适应度函数，可修改

论文研究-供水管网爆管故障诊断的PSO-SVM方法.pdf,  根据供水管网的实际水压监测数据, 采用粒子群算法优化反演管道的海曾-威廉斯系数, 并通过在管道中间加入虚节点来模拟爆管故障, 进而基于节点水压法建立了管网在爆管故障情况下的水力计算模型. 由计算模型计算出一组不同爆点、不同爆管程度组合下的监测点处的水压值, 以此反向训练支持向量机(SVM)模型, 并通过粒子群算法(PSO)对优化支持向量机模型的核参数, 建立了基于PSO-SVM方法的供水管网爆管诊断模型. 最后, 通过一个供水管网的室内实验模型验证了上述诊断模型的有效性.