In order to solve the problem that the sparrow optimization algorithm (SSA) has slow convergence speed and is easy to fall into local optimal solution in the parameter inversion calculation of mining subsidence prediction model, an improved sparrow optimization algorithm (ISSA) is proposed. The algorithm adds Kent mapping to the population initialization process to enhance the uniform distribution of population individuals. The foraging behavior of the parrot optimization algorithm is introduced in the location update of the discoverer, and the safety value is adjusted according to the fitness. In order to improve the performance of the algorithm, the t disturbance distribution and the lens reverse learning strategy are combined. WOA, SSA and ISSA are used to invert the parameters of probability integral method (PIM) respectively. The PIM with inversion parameters is used to simulate the sinking value, anti-rough error interference ability and anti-random error of the experimental working face. The results show that compared with WOA and SSA, the PIM simulation effect using ISSA inversion parameters is the best. The ISSA is applied to an engineering example, and the results show that the calculated value of the PIM model using ISSA inversion parameters is closer to the actual value. The research conclusion provides a reference for improving the accuracy of subsidence prediction and mining area disaster detection.
概率积分法(probability integral method,PIM)基于介质理论发展而来,是矿区开采沉陷研究的重要工具[5-8]。PIM预测变形的准确性依赖于模型参数,不同地质开采环境需针对性地确定参数取值[9]。目前,随着机器学习的不断发展,智能算法被广泛应用于PIM参数的反演中。王正帅等[10]利用粒子群优化算法进行PIM参数反演,使PIM的预测准确性有所提升。乔薄庆等[11]通过改进的蛇优化算法进行PIM参数反演,使PIM的预测精度和准确性均有一定程度提高。孙志豪等[12]通过加权赋值和自适应模拟退火粒子群算法进行沉陷预测,解决了厚松散层边缘处预测模型的快速收敛问题。花朵授粉法[13]、狮群优化算法[14]、蝙蝠算法[15]、人工鱼群算法[16]等智能优化算法,在概率积分参数参演中均取得不错效果。以上研究表明,通过智能优化算法反演PIM的参数,提高了PIM模型对地表变形预测的精度。
LIUZhiqiang, SONGZhaoyang, CHENGShouye,et al. Seventy years innovation development and popularization practice of coal mine construction technology and equipment in China[J].Coal Science and Technology, 2024, 52(1): 65-83.
SUNWenjie, LIWenjie, YANGWenkai,et al.Classification of water environment problems in coal mine and treatment mode[J].Safety in Coal Mines,2023,54(5): 35-41.
SONGShijie, FENGZexu, SUNTao, et al. Loess slope deformation and soil erosion effect in coal mining subsidence area of northern Shanxi[J]. Journal of Xi'an University of Science and Technology,2023,43(2):301-311.
WANGAilin, ZHANGPengfei, ZHANGBiwen,et al.Influence parameter analysis of surface subsidence measurement in mining area and comparison verification of non-principal section monitoring inversion model[J]. Safety in Coal Mines,2020,51(5): 219-224.
ZhengyiTI, LIJiazhen, WANGMeng, et al. Study on coal-mining failure depth of the inclined coal seam floor based on fracture mechanics [J].Journal of Central China Normal University (Natural Science Edition),2020,54(5):787-791, 797.
YANGYu, ZHANGPeilan.Analysis on land reclamation in Niuxintai coal mining subsidence area[J].Journal of Liaoning Technical University (Natural Science),2015,34(8):926-929.
LIUWensheng, ZHANGYanfeng, ZHANGHeran, et al. Application of Grey system model in the deformation prediction of the open-pit mine slope[J].Journal of Liaoning Technical University (Natural Science), 2014, 33(12):1608-1612.
XUKexin, HERong, BAIWeisen.Research on inversion of mining subsidence prediction parameters based on improved SPSO algorithm[J]. Safety in Coal Mines, 2022, 53(8): 218-224.
WANGZhengshuai, DENGKazhong, KANGJianrong. Random PSO embedded cultural framework for parameters inversion of probability-integral method[J].Journal of Liaoning Technical University (Natural Science), 2013, 32(3): 311-315.
QIAOBoqing, GUOQingbiao, LUOJin,et al.Parameter inversion of probability integral method based on improved snake optimization[J].Metal Mines,2025(4):184-194.
SUNZhihao, XULiangji, LIUXiaopeng.A method for predicting subsidence in thick loose layer mines based on segmental weighted parameter assignment[J].Metal Mine,2024(11): 132-141.
LIUShuang, LIUYu.Inversion of probability integral parameters based on improved intelligent algorithm[J].Shandong Coal Science and Technology,2022,40(10): 194-197.
HUANGJinzhong, LIZhong, LIShibao.Probability integral prediction parameter inversion method based on lion swarm optimization[J]. Beijing Surveying and Mapping,2022,36(2):101-108.
LIZhong, WANGLei, TENGChaoqun, et al. Probability integral prediction parameter inversion method based on bat algorithm[J].China Mining Magazine,2021,30(8): 89-96.
ZHANGGuanjin, SONGZijie, ZHOUZiwen, et al. Inversion of probabilistic integral model parameters based on artificial fish school[J]. Journal of Anhui Science and Technology University,2021,35(4):70-76.
WUManyi, XULiangji, ZHANGKun.Estimated parameter extraction research of probabilistic integration method based on SSA-BP neural network[J]. Metal Mine,2022(8):182-189.
WANGShuwei.Faults diagnosis of CNC machine tool by using BP neural network optimized by sparrow algorithm[J]. Journal of Shenyang University of Technology,2023,45(5):546-551.
[36]
ZHANGM L, HAOP.2D and 3D path planning for mobile robots based on improved SSA algorithm[J].International Journal of Intelligent Robotics and Applications,2025,9(1):176-188.
[37]
ZHOUW P, XIONGH J.PV power prediction based on SSA improved BP neural network[J].Journal of Physics: Conference Series,2024,2831(1):012028.
WANGJianmin, LIUZhipeng, HUANGJiapeng,et al.Research on ionospheric total electron content forecasting using VMD-SSA-LSTM[J].Journal of Navigation and Positioning, 2024,12(3): 88-101.
[40]
WANGJ, XUD, YANGW S,et al.Advanced thermal prediction for green roofs: CNN-LSTM model with SSA optimization[J].Energy and Buildings, 2024, 322: 114745.
GUOQinghui, LIYuan, YANGDongsheng.Convergence analysis and application of an improved sparrow search algorithm[J].Control and Decision, 2024, 39(8): 2502-2510.
[43]
LOUJ.A parameter inversion method for the probability integral method based on robust ridge estimation[J].Frontiers in Earth Science,2024,11:1330163.
[44]
GUOQ B, CHENH K, LUOJ, et al. Parameter inversion of probability integral method based on improved crow search algorithm[J].Arabian Journal of Geosciences,2022,15(2):180.