灾害气体在风网中传播的时间-事件混合步长法

Mixed time-event-step method of disaster gas migration in ventilation network

为仿真描述有源风网上外源气体涌入运移过程,创新引入事件步长法,把风网中外源气体组分运动到达某一节点作为一个事件,同时取同期其他分支事件所经历的最短时间作为下一个时间步长,考察记录网络分支中外源气体到达的位置.为减小截断误差,对过大的事件时间步长作出限定,形成所谓"时间-事件混合步长"法.使每个迭代时间步长都能至少有外源气体到达某一个节点,采用节点分支汇流法计算节点浓度,使网络节点上气体浓度得到及时更新.结合矿井瓦斯突出流在网络中的扩散给出了算例,展示了不同时刻瓦斯达到网络各个节点的过程.研究结果表明:新方法能够自动截取网络中组分气体流动的微小事件变化,防止因过大的时间步长导致的截断误差;与单纯的时间步长法相比,计算精度高,模拟仿真更准确.

To describe the migration process of external gas in source ventilation network, the event-step method is introduced. The situation that when external gas arrives at a certain point is pointed as an event, the shortest time that others branches experience the event during the same period is set to be the next time step, and the positions external gas arrives at in ventilation network are recorded. At the same time, to reduce the truncation error, the excessive event-time-step is limited and this forms the ＂mixed time-event-step＂ method. The external gas can arrive at a certain point in every iteration time, the point concentration is calculated with point branch converge method, so the gas concentration at point can be updated timely. The process of gas arriving at every point at different time is given out through the combination of the example of mine outburst gas in ventilation network. The results show that the new method can get the micro event change of gas flow in network automatically, the truncation error caused by excessive time step can be prevented. Compared with the single time-step method, the calculation accuracy is higher and the simulation is more accurate.