The invention discloses a curved surface fitting algorithm taking a radial basis network as a core. The curved surface fitting algorithm comprises the following steps: designing a radial basis network model for fitting; according to a model learning algorithm, solving a velocity distribution curved surface in a current time-distance space; calculating the space-time average speed of the road section. A research object is limited to a road section, a spatial and temporal distribution rule of traffic flow speed on the road section is fully utilized, real speed distribution is approached by using discrete sampling points through a modeling method, so that the method is more suitable for urban road network traffic state calculation, analysis and demonstration are carried out by using experiments, and the method aims at an existing traffic state calculation method, in order to overcome the defects that the calculation is inaccurate and the generalization ability is not strong when the short-time traffic problem emerges, the algorithm is improved, so that the algorithm has stronger robustness.
本发明公开了一种以径向基网络为核心的曲面拟合算法,包括以下步骤:设计用于拟合的径向基网络的模型;根据模型学习算法,求解当前时间‑距离空间上的速度分布曲面;计算路段时空平均速度。将研究对象限定到路段上,充分利用了车流速度在路段上的时空分布规律,通过建模的方法,利用离散采样点逼近真实的速度分布,使之更适应城市路网交通状态计算,并利用实验加以分析论证,针对了现有的针对现有交通状态计算方法,在短时交通问题突发时计算不准确泛化能力不强的缺点,对算法进行改进,使其具有更强的鲁棒性。
Curved surface fitting algorithm taking radial basis network as core
一种以径向基网络为核心的曲面拟合算法
2021-10-22
Patent
Electronic Resource
Chinese
Curved surface fitting drum design method and curved surface fitting drum
European Patent Office | 2021
|Curved mesh generation using radial basis functions
AIAA | 2016
|Curved Mesh Generation Using Radial Basis Functions (AIAA 2016-3179)
British Library Conference Proceedings | 2016
|Application of Artificial Fish Swarm Algorithm in Radial Basis Function Neural Network
BASE | 2016
|