煤炭是我国发展的主要能源，随着计算机技术和人工智能算法在各个领域的应用，结合计算机远程监控和人工干预的煤炭智能化开采已经成为企业煤矿开采的核心。由于工作面开采之前得到的采样数据有限，因此常采用插值方法对工作面内部煤层结构进行预测。此外，在工作面回采过程中，对下一阶段开采到的煤总量进行煤质指标预测有助于指导下一步洗选、配煤方案，而在开采过程中，由于计算量过大和人工操作的误差容易导致煤炭产量预测不准确，从而影响工作面煤质指标预测结果。基于以上问题，论文以克里金（Kriging）插值算法和差分进化（Differential Evolution, DE）算法为理论基础，研究工作面三维模型建立过程，建立工作面煤质指标实时预测过程。主要研究内容如下：

(1) 为求解工程应用上存在的参数确定问题，对元启发式算法进行研究。其中，差分进化算法作为传统的优化算法，参数少，收敛能力强且易于理解，被研究人员广泛采用。针对差分进化算法容易早熟陷入局部最优解的问题，提出改进的自适应指导差分进化算法（Improved Adaptive Guided Differential Evolution, IAGDE），利用新的变异策略、参数自适应调整机制和种群缩减策略对差分进化算法进行改进，并在CEC2013数据集上进行实验。实验结果表明，与SHADE等其他优化算法相比，IAGDE算法的求解精度更高，收敛速度更快，稳定性更好。

(2) 由于综采工作面在开采之前得到的地质数据是有限且稀疏的，故采用插值算法对工作面内部煤层结构高程进行插值。通过分析插值算法的优缺点，选择地质统计学上的Kriging插值法作为理论基础。针对Kriging插值法在变差函数模型拟合过程中存在的过拟合或欠拟合问题，利用IAGDE算法求解Kriging变差函数拟合模型中的参数，提出IAGDE-Kriging插值法。利用某煤矿一工作面在布置过程中得到的运顺和回顺巷道的高程数据，对提出的插值法进行交叉实验验证，并与PSO-Kriging等插值算法比较。实验结果显示，利用IAGDE-Kriging算法得到的插值与实际值之间误差更小，拟合效果更好。

(3) 为更好实现工作面煤质指标预测，建立工作面三维模型和煤质预测系统。首先，对工作面布置过程中得到的勘探数据、巷道信息、小柱状数据、钻探数据等采用多元数据融合技术进行分析和整理，得到可用于插值的基础数据，并运用IAGDE-Kriging法进行对工作面内部顶底板和煤矸层高程进行插值，得到丰富的工作面内部高程数据。其次，利用三维空间数据模型中的规则格网法和数字高程建模方法建立工作面三维模型。最后，在工作面开采过程中，工作人员在浏览器端的交互界面上输入下一阶段综采推进度，计算机根据工作面三维模型计算推进范围内即将开采出来的煤矸石体积。并结合工作面要素信息、基础煤质信息以及工作面毛煤、原煤和商品煤煤质预测公式，进行实时的煤质预测。

Coal is the main energy source for our country's development. With the application of computer technologies and artificial intelligence algorithms in various fields, intelligent coal mining with the combination of computer remote monitoring and manual intervention has become the core of enterprise coal mining. Due to the sampling data obtained before mining the working face is limited, the interpolation method usually be used to predict the coal seam structure inside the working face. In addition, in the mining process, it is necessary to predict the coal quality index of the total amount of coal mined in the next stage and guide the next washing and coal blending scheme according to the predicted coal quality index. However, in the mining process, the coal yield prediction is inaccurate due to the large amount of calculation and the error of manual operation, which affects the prediction result of coal quality index of the working face. To solve the above problems, based on theoretical foundation of kriging interpolation algorithm and differential evolution (DE) algorithm, the paper studies the building process of 3d model of working face and realizes real-time prediction of coal quality indicators of working face. The main research contents are as follows:

(1) In order to solve the problem of parameter determination in engineering applications, this paper studies the meta-heuristic algorithms. The differential evolution algorithm is widely used due to its advantages of less parameters and easy understanding. Aiming at the DE algorithm easily falling into the local optimal solution resulting in the algorithm stagnation, an improved guided differential evolution algorithm (IAGDE) is proposed. The IAGDE algorithm adopts a new mutation strategy, a parameters adaptive tunning scheme and a population reduction strategy to optimize the differential evolution algorithm, and experiments are carried out on the CEC2013 data set. The experimental results show that compared with other optimization algorithms such as SHADE, the IAGDE algorithm has higher solution accuracy, faster convergence speed and better stability.

(2) Due to the limited and sparse geological data obtained in the working face before mining, an interpolation algorithm is used to interpolate the elevation of the coal seam structure inside the working face. This paper analyzes the advantages and disadvantages of several interpolation algorithms, and selects the Kriging interpolation method, which is widely used in geostatistics, as the theoretical basis. Aiming at the over-fitting or under-fitting caused by the problem of parameter determination during the fitting process of the variogram model by kriging interpolation, employing the IAGDE algorithm to solve the parameters in the Kriging variogram fitting model and proposing the IAGDE-Kriging interpolation method to interpolate the elevation of the inner roof and floor of the working face and the coal gangue layer. Based on the roadway floor elevation data of the arranged working face, the method is verified by cross-experiment and compared with other interpolation algorithms such as PSO-Kriging. The results show that the interpolation error of the IAGDE-Kriging algorithm is smaller and the fitting effect is better.

(3) Using multivariate data fusion technology to analyze and organize the excavated data, roadway information, small columnar data, drilling data, etc. that obtained during the layout of the working face. Then using the IAGDE-Kriging method to interpolate and rich working face internal elevation data according to the basic data. The three-dimensional model of the working face is realized by using the regular grid method and digital elevation modeling in the three-dimensional spatial data model. According to the established three-dimensional model of the working face, within a certain range of advancement, calculating the volume of coal gangue mined. Combining the coal quality prediction formulas of the rough coal, raw coal and commercial coal of the working face to establish an interactive interface and to realize real-time coal quality prediction process.

[1] 贾建称, 巩泽文, 靳德武, 等. 煤炭地质学“十三五”主要进展及展望[J]. 煤田地质与勘探, 2021, 49(1): 32–44.

[2] 谢和平, 吴立新, 郑德志. 2025 年中国能源消费及煤炭需求预测[J]. 煤炭学报, 2019, 44(7): 1949–1960.

[3] 李鹏, 程建远．回采工作面煤层三维建模技术及其在智能开采中的应[J]．煤矿安全, 2021, 52(8): 156-161.

[4] 廉西猛, 单联瑜, 隋志强, 等. 基于四面体剖分的并行地质块体建模方法[J]. 计算机工程与应用, 2018, 54(21): 246-250.

[5] 刘万里, 张学亮, 王世博. 采煤工作面煤层三维模型构建及动态修正技术[J]. 煤炭学报, 2020, 45(6): 1973-1983.

[6] 周为喜, 陈玉华, 杨永国, 等. 基于角点网格的煤层三维建模与可视化研究[J]. 煤田地质与勘探, 2016, 44(5): 53–57.

[7] 车德福, 陈学习, 吴立新, 等. 基于广义三棱柱体元的三维地层建模方法[J]. 辽宁工程技术大学学报, 2006, 25(1): 36-38.

[8] Manchuk J G, Deutsch C V. Boundary modeling with moving least squares[J]. Computers & Geosciences, 2019, 126: 96-106.

[9] 程建远, 朱梦博, 崔伟雄, 等. 回采工作面递进式煤厚动态预测试验研究[J]. 煤炭科学技术, 2019, 47(1): 237-244.

[10] Che D, Jia Q. Three-Dimensional Geological Modeling of Coal Seams Using Weighted Kriging Method and Multi-Source Data[J]. IEEE Access, 2019, 7: 118037-118045.

[11] 王新苗, 韩保山, 宋焘, 等. 智能开采工作面三维地质模型构建及误差分析[J]. 煤田地质与勘探, 2021, 49(2): 93-101+109.

[12] 荆永滨, 杜学胜, 张瑞林, 等. 复杂地质构造煤层三维模型自动构建技术[J]. 辽宁工程技术大学学报(自然科学版), 2016, 35(3): 243-247.

[13] 张小艳, 朱圣凯, 杨鑫磊. 采煤工作面煤层三维地质建模[J]. 科学技术与工程, 2020, 20(10): 4049-4055.

[14] Jia Q, Che D, Li W. Effective coal seam surface modeling with an improved anisotropy-based, multiscale interpolation method[J]. Computers and Geosciences, 2019, 124:72-84.

[15] Zhang Q, Zhu H. Collaborative 3D geological modeling analysis based on multi-source data standard[J]. Engineering Geology, 2018, 246: 233-244.

[16] Wu Q, Xu H. Three-dimensional geological modeling and its application in Digital Mine[J]. Science China Earth Sciences, 2014, 57(3):491-502.

[17] Renaudeau J, Malvesin E, Maerten F, et al. Implicit structural modeling by minimization of the bending energy with moving least squares functions[J]. Mathematical geosciences, 2019, 51(6): 693-724.

[18] 王震. 基于自适应差分退火算法的作业车间调度问题研究[D]. 武汉：武汉纺织大学, 2020.

[19] 荆涛, 田锡天. 基于蒙特卡洛-自适应差分进化算法的飞机容差分配多目标优化方法[J]. 航空学报, 2022, 43(03): 577-588.

[20] 张洪杰. 改进差分进化算法在电力系统经济调度中的应用研究[D]. 秦皇岛：燕山大学, 2020.

[21] 李甜甜. 基于改进粒子群算法的超参数优化问题的研究[D]. 西安：西安电子科技大学, 2019.

[22] Zhu Z, Tang B, Yuan J. Multirobot task allocation based on an improved particle swarm optimization approach[J]. international Journal of Advanced robotic systems, 2017, 14(3): 1729881417710312.

[23] Yu M. A solution of TSP based on the ant colony algorithm improved by particle swarm optimization[J]. Discrete & Continuous Dynamical Systems-S, 2019, 12(4&5): 979.

[24] 刘林, 王朋, 翟永杰, 等. 基于算法的煤质发热量预测[J]. 热力发电, 2015, 44(02): 47-51+57.

[25] 武秋芳. 煤炭质量预测方法研究及信息管理系统开发[D]. 郑州：华北水利水电大学, 2017.

[26] 谭鹏, 李鑫, 张小培, 等. 基于工业分析的煤质发热量预测[J]. 煤炭学报, 2015, 40(11): 2641-2646.

[27] 胡涛, 茅大钧, 程鹏远, 等. 基于煤质预测的多目标优化配煤方法研究[J]. 热能动力工程, 2021, 36(12): 151-156.

[28] 李海涛, 邵泽东. 空间插值分析算法综述[J]. 计算机系统应用, 2019, 28(7): 1–8.

[29] 王玉璟. 空间插值算法的研究及其在空气质量监测中的应用[D]. 郑州：河南大学, 2010.

[30] 杨鑫磊. 采煤工作面煤质核心指标估算模型研究[D]. 西安：西安科技大学, 2020.

[31] Pant M, Zaheer H, Garcia-Hernandez L, et al. Differential Evolution: A review of more than two decades of research[J]. Engineering Applications of Artificial Intelligence, 2020, 90: 103479.

[32] Y. Lu, Y. Lin, Q. Peng, and Y. Wang. Summary of simulated annealing algorithm improvement and parameter exploration[J]. College Mathematics, 2015, 31(06): 96-103.

[33] 闫群民, 马瑞卿, 马永翔, 等. 一种自适应模拟退火粒子群优化算法[J]. 西安电子科技大学学报, 2021, 48(4): 120-127.

[34] Storn R, Price K. Differential evolution–a simple and efficient heuristic for global optimization over continuous spaces[J]. Journal of global optimization, 1997, 11(4): 341-359.

[35] Sun Z, Wang N, Bi Y, et al. Parameter identification of PEMFC model based on hybrid adaptive differential evolution algorithm[J]. Energy, 2015, 90: 1334-1341.

[36] Sun Z, Ling Y, Qu H, et al. An adaptive DE algorithm based fuzzy logic anti-swing controller for overhead crane systems[J]. International Journal of Fuzzy Systems, 2020, 22(6): 1905-1921.

[37] Deng W, Xu J, Song Y, Zhao H. Differential evolution algorithm with wavelet basis function and optimal mutation strategy for complex optimization problem[J]. Applied Soft Computing Journal, 2021, 100:106724.

[38] Zhang J, Sanderson AC. JADE: adaptive differential evolution with optional external archive[J]. IEEE Transactions on Evolutionary Computation, 2009, 13(5): 945-58.

[39] Sun G, Yang B, Yang Z, Xu G. An adaptive differential evolution with combined strategy for global numerical optimization[J]. Soft Computing, 2020, 24: 627796.

[40] Tanabe R, Fukunaga AS. Improving the search performance of SHADE using linear population size reduction[C]. 2014 IEEE congress on evolutionary computation (CEC) 2014, 1658-65.

[41] Mohamed AK, Mohamed AW. Real-Parameter unconstrained optimization based on enhanced AGDE algorithm[J]. Machine Learning Paradigms: Theory and Application, 2019, 431-50.

[42] Gämperle R, Müller SD, Koumoutsakos P. A parameter study for differential evolution[J]. Advances in Intelligent Systems, Fuzzy Systems, Evolutionary Computation, 2002, 10(10): 29398.

[43] Das S, Abraham A, Chakraborthy UK. Differential evolution using a neighborhood-based mutation operator[J]. IEEE Transactions on Evolutionary Computation, 2009,13: 526–553.

[44] Iorio AW, Li X. Solving rotated multi-objective optimization problems using differential evolution[C]. Australian Joint Conference on Artificial Intelligence, 2004, 86172.

[45] Mallipeddi R, Suganthan PN, Pan QK, Tasgetiren MF. Differential evolution algorithm with ensemble of parameters and mutation strategies[J]. Applied Soft Computing, 2011, 11(2): 1679-96.

[46] Wang Y, Cai Z, Zhang Q. Differential evolution with composite trial vector generation strategies and control Parameters[J]. IEEE Transactions on Evolutionary Computation, 2011, 15(1): 55-66.

[47] Wu G, Shen X, Li H, Chen H, Lin A, Suganthan PN. Ensemble of differential evolution variants[J]. Information Sciences, 2018, 423: 172-186.

[48] Tan Z, Li K, Wang Y. Differential evolution with adaptive mutation strategy based on fitness landscape analysis[J]. Information Sciences, 2021, 549: 142-163.

[49] Chen X. Novel dual-population adaptive differential evolution algorithm for large-scale multi-fuel economic dispatch with valve-point effects[J]. Energy, 2020, 203: 117847

[50] Sun G, Yang B, Yang Z, Xu G. An adaptive differential evolution with combined strategy for global numerical optimization[J]. Soft Computing, 2020, 24: 6277-6296.

[51] Yi W, Chen Y, Pei Z, Lu J. Adaptive differential evolution with ensembling operators for continuous optimization problems[J]. Swarm and Evolutionary Computation, 2021, 100994.

[52] Ronkkonen J, Kukkonen S, Price KV. Real parameter optimization with differential evolution[C]. Proceedings of the IEEE congress evolutionary computation (CEC-2005), 2005, 1: 506–513.

[53] Wang Y, Cai Z, Zhang Q. Differential evolution with composite trial vector generation strategies and control parameters[J]. IEEE Transactions on Evolutionary Computation, 2011, 15(1): 55–66.

[54] Liang JJ, Qu BY, Suganthan PN, Hernandez-Diaz AG. Problem Definitions and Evaluation Criteria for the CEC 2013 Special Session on Real-Parameter Optimization[C]. Zhengzhou University, China, and Nanyang Technological University, Singapore, 2013.

[55] AGDE Mohamed AW, Mohamed AK. Adaptive guided differential evolution algorithm with novel mutation for numerical optimization[J]. International Journal Machine Learing and Cybernetics, 2019, 10(2): 253-277.

[56] Tanabe R, Fukunaga A. Success-history based parameter adaptation for differential evolution[C]. 2013 IEEE congress on evolutionary computation, 2013, 71-78.

[57] Feng X, Zou R, Yu H. A novel optimization algorithm inspired by the creative thinking process[J]. Soft Computing, 2015, 19(10):295572.

[58] Caraffini F, Neri F, Cheng J, Zhang G, Picinail L, Iacca G, Mininno E. Super-fit multicriteria adaptive differential evolution[C]. 2013 IEEE Congress on Evolutionary Computation (CEC), 2013, 167885.

[59] Mohamed AW, Sabry HZ, Farhat A. Advanced differential evolution algorithm for global numerical optimization[C]. 2011 IEEE International Conference on Computer Applications and Industrial Electronics (ICCAIE)[C], 2011, 15661.

[60] Hansen N, Mller SD, Koumoutsakos P. Reducing the time complexity of the derandomized evolution strategy with covariance matrix adaptation (CMA-ES) [J]. Evolutionary Computation, 2003, 11(1):1-18.

[61] Feng X, Zou R, Yu H. A novel optimization algorithm inspired by the creative thinking process[J]. Soft Computing, 2015, 19(10):295572.

[62] Li W, Lu J, Dong M, et al. Quantitative analysis of calorific value of coal based on spectral preprocessing by laser-induced breakdown spectroscopy (LIBS)[J]. Energy & Fuels, 2018, 32(1): 24-32.