基于ICGSO算法的瓦斯突出预测研究

Research of gas outburst prediction model based on ICGSO algorithm

针对萤火虫群优化(GSO)算法在解决全局优化问题时出现的易陷入局部最优、收敛速度慢、求解精度不高等问题,提出一种改进的混沌萤火虫群优化(ICGSO)算法,修改了GSO算法动态决策域半径更新公式,并采用自适应动态步长,引入混沌优化算法提高局部搜索能力。实验结果表明将ICGSO算法应用于建立在神经网络预测模型上的瓦斯突出预测中的有效性。

Glowworm Swarm Optimization（GSO） algorithm has some drawbacks such as getting in local minima,convergence rate is slow and the precision of the solution is not high.In order to solve this problem,an improved Chaos Glowworm Swarm Optimization algorithm was presented in this paper.The new algorithm with a new formula to replace the standard dynamic decision domain radius updating formula of GSO algorithm adopted adaptive dynamic step length,and chaotic optimization algorithm is introduced to improve the ability of the local search.The new algorithm is introduced into the construction of gas outburst prediction model,experimental results prove the validity of the model.