关键词: 开采沉陷, 概率积分法, 参数反演, 粒子群优化算法, 量子粒子群优化算法

Abstract: In process of parameter inversion of probability integral method predicts that there are problems such as large calculation amount and complicated process in the process of inverting parameters of the model.Existing intelligent optimization algorithms can make up for these deficiencies，but there are some defects such as easy to fall into premature convergence，poor global search effect of particles，and slow convergence speed.Through experiments，it was found that the quantum-behaved particle swarm optimization algorithm （QPSO） can greatly reduce the running time of the algorithm on the basis of ensuring the accuracy is unchanged， and reduce the probability of particles falling into premature convergence，expanding the particles to be globally unique Solution space.The QPSO algorithm is introduced into the solution of the prediction parameters of mining subsidence.The cumulative sum of the absolute value of the difference between measured values and predicted values of the sinking and moving deformation is the minimum cost function，and the probability integral method parameters inversion model based on QPSO algorithm is constructed.The study results show that:①in the simulation test，under the same operating environment，the QPSO algorithm and the particle swarm optimization algorithm （PSO） have the same precision，and the QPSO algorithm has a slightly higher stability，and the efficiency of the parameter is greatly improved （the running time of the QPSO algorithm is reduced by nearly 90% compared to the PSO algorithm），which verifies the validity and reliability of the parameter inversion model based on the QPSO algorithm；②the probability integral method parameters of of Guqiao South Mine 1414（1） working face are calculated based on the QPSO parameter inversion model，the results show that q=1.041 5，tan[β]=1.910 8，b=0.374 2，[θ]=85.086 9，[S1]=55.663 5 m，[S2]=37.161 8 m，[S3]=-0.667 0 m，[S4]=-9.798 0 m，the error in the fitting of sinking and horizontal movement is 72.04 mm，which meets the engineering application standards，although the precision of QPSO algorithm is similar to that of PSO algorithm，its operational efficiency is significantly improved.The model established in this paper has certain reference value for the accurate inversion of mining subsidence prediction parameters.

Key words: Mining subsidence, Probability integral method, Parameter inversion, Particle swarm optimization algorithm, Quantum-behaved particle swarm optimization algorithm