基于自适应变异 PSO-BP算法的船舶横摇运动预测

为了准确高效预测船舶在海上的航行状态，以保证人员、货物和船舶的安全，提出一种自适应变异的粒子群优化算法（self-adapting particle swarm optimization algorithm，SAPSO），将该算法与误差反传（back propaga-tion，BP）神经网络结合。SAPSO-BP预测模型使用SAPSO算法优化BP网络的网络参数。克服传统BP神经网络对初始权值阈值敏感，容易陷入局部极小值的缺点，同时也克服了传统PSO算法早熟收敛、搜索准确度低及迭代效率低等缺点。运用该模型对科研教学船“育鲲”轮在海上航行的横摇情况进行实时预测实验，验证该方法的可行性与有效性具有较高的预测精度。

Ship rolling forecasting based on self-adapting PSO-BP algorithm

In order to predict the navigation state of ship in the wind and waves accurately, timely and efficient and en-sure the safety of personnel, cargo and ship. We propose a self-adapting particle swarm optimization (SAPSO) algorithm to optimize the back propagation (BP) neural network model. The proposed model is referred to as SAPSO-BP model which employs PSO to adjust control parameters of BP network. This method overcomes the shortcomings of traditional BP neural network, which is sensitive to the threshold value of the initial value and is easy to fall into local minimum. At the same time, it also overcomes the shortcomings of the traditional PSO algorithm, such as premature convergence, low accuracy, and low efficiency and so on. The measurement data from scientific research and training ship Yukun was chosen as the test database. Simulation results have demonstrated that the proposed method can give predictions for ship rolling motion in real time with high accuracy and satisfactory stability.