在矿井局部通风系统中，局部通风机是不可缺失的重要组成环节，局部通风机一般安装于掘进工作面掘进巷道，它主要为掘进工作面提供新风，降低瓦斯、粉尘等浓度，其稳定、可靠的运行直接影响煤矿的安全生产。目前，我国大部分煤矿井下的局部通风系统，其风机采用定速运行，长时间超负荷运转，未根据煤矿井下掘进工作面瓦斯、粉尘等浓度变化实现实时风量调节，未能实现“风电闭锁、瓦斯电闭锁”自动处理，为煤矿安全生产埋下了隐患，同时也造成了电能的浪费。针对上述局部通风中的突出问题，迫切需要探索一种既能够满足安全通风需求，又能够实现节能降耗的方案，以确保局部通风机高效安全运行。 本文以韩城矿业有限公司象山矿井某一掘进工作面通风情况为研究对象展开研究： （1）针对原海洋捕食者算法(MPA)有着和其它仿生智能算法一样的缺陷，对原算法进行了多策略改进，提出了一种多策略改进的海洋捕食者算法(MMPA)。对MMPA算法在多种测试函数上进行了性能测试。结果表明：改进的MPA算法收敛速度更快，寻优精度更高且不易陷入局部最优解。 （2）针对局部通风系统模型非线性、干扰因素多等特点，运用系统辨识相关理论，选择二阶差分传递函数作为系统的辨识模型，使用带遗忘因子递推的最小二乘算法(FFRLS)和MMPA算法对系统模型进行了辨识，然后对辨识出的系统模型用测试数据进行可靠性分析。可靠性分析结果显示：MMPA算法最大相对误差为2.92%，平均相对误差为0.87%；FFRLS算法最大相对误差为6.50%，平均相对误差为1.84%，MMPA算法辨识模型时误差有明显的下降。这表明MMPA算法辨识的系统可靠有效，需风量经PLC上位机调整后，即可满足安全通风需求。 （3）针对局部通风机风量调节系统控制参数凭人工经验设定的缺点，设计了MMPA-PID控制器和MMPA-RBF-PID控制器，通过对局部通风机风量调节系统一系列设备的数学建模来模拟实际的风量调节系统的工作流程，对调节系统进行阶跃响应、鲁棒实验和抗扰动实验，对不同的控制器性能做出对比和分析。仿真结果表明：MMPA-RBF-PID控制器相较传统的PID控制器和MMPA-PID控制器，其在快速性、鲁棒性和抗扰动性方面均有较大的提升，超调量比常规PID控制减少了26.8%，调节时间减少了5.3s，稳定性更好，减少了电能资源的浪费，达到了精准控制的效果。

In the mine local ventilation system, the local fan is an indispensable and important part. The local fan is generally installed in the tunneling roadway of the tunneling face. It mainly provides fresh air for the tunneling face and reduces the concentration of gas and dust. It is stable and reliable. The operation of coal mines directly affects the safe production of coal mines. At present, the local ventilation systems of most coal mines in my country operate at a constant speed and are overloaded for a long time. The real-time air volume adjustment has not been realized according to the changes in the concentration of gas and dust in the tunneling face of the coal mine. The automatic processing of "gas and electricity lockout" has buried hidden dangers for the safe production of coal mines, and has also caused a waste of electric energy. In view of the outstanding problems in the above local ventilation, it is urgent to explore a solution that can not only meet the safety ventilation requirements, but also achieve energy saving and consumption reduction, so as to ensure the efficient and safe operation of local fans.

This paper takes the ventilation situation of a tunneling face in the Xiangshan Mine of Hancheng Mining Co., Ltd. as the research object:

(1) For the original MPA algorithm has the same defects as other bionic intelligent algorithms, a multi-strategy improvement is made to the original algorithm, and an MMPA algorithm is proposed. The performance test of MMPA algorithm is carried out on various test functions. The results show that the improved MPA algorithm has faster convergence speed, higher optimization precision and is less likely to fall into local optimal solution.

(2) In view of the characteristics of the local ventilation system model such as nonlinearity and multiple interference factors, the second order difference transfer function is selected as the identification model of the system by using the relevant theory of system identification. The system model is identified by using FFRLS algorithm and MMPA algorithm, and then the reliability of the identified system model is analyzed with test data. The reliability analysis results show that the maximum relative error of the MMPA algorithm is 2.92%, and the average relative error is 0.87%; the maximum relative error of the FFRLS algorithm is 6.50%, and the average relative error is 1.84%. This shows that the system identified by the MMPA algorithm is reliable and effective, and the required air volume can be adjusted by the PLC host computer to meet the safety ventilation requirements.

(3) Aiming at the shortcomings of setting the control parameters of the local fan air volume adjustment system based on manual experience, the MMPA-PID controller and MMPA-RBF-PID controller are designed. Through the mathematical construction of a series of equipment in the local fan air volume adjustment system the working process of the actual air volume adjustment system is simulated by the model, and the step response, robustness and anti-disturbance experiments are carried out on the adjustment system, and the performance of different controllers is compared and analyzed. The simulation results show that compared with the traditional PID controller and MMPA-PID controller, the MMPA-RBF-PID controller has greatly improved in terms of rapidity, robustness and anti-disturbance, and the overshoot is higher than that of the conventional The PID control is reduced by 26.8%, the adjustment time is reduced by 5.3s, the stability is better, the waste of power resources is reduced, and the effect of precise control is achieved.

[1]梁壮, 邸帅, 赵东波. 煤炭行业高质量发展存在的问题及对策研究[J]. 中国煤炭, 2022, 48(11): 9-14.

[2]李华. 通风机监控系统的设计与研究[J]. 煤矿机械, 2010, 31(01): 150-152.

[3]侯鹏伟. PLC技术在矿山机电控制中应用分析[J]. 矿业装备, 2022(06): 166-168.

[4]程健维, 陈祖云, 郑万成, 等. 智能矿井通风调节演示实验系统的设计与实验研究[J]. 中国现代教育装备, 2021(15): 20-23+34.

[5]张超林, 王恩元, 王奕博, 等. 近20年我国煤与瓦斯突出事故时空分布及防控建议[J]. 煤田地质与勘探, 2021, 49(04): 134-141.

[6]成连华, 郭阿娟, 郭慧敏, 等. 煤矿瓦斯爆炸风险耦合演化路径研究[J]. 中国安全科学学报, 2022, 32(04): 59-64.

[7]王海军, 曹云, 王洪磊. 煤矿智能化关键技术研究与实践[J]. 煤田地质与勘探, 2023, 51(01): 44-54.

[8]黄山峰, 秦冠军, 姜凯, 等. 基于多元线性回归算法的风电变桨电机温度预警方法[J]. 能源研究与利用, 2022(05): 39-41.

[9]Shi Y. Developmental swarm intelligence: developmental learning perspective of swarm intelligence algorithms[J]. International journal of swarm intelligence research (IJSIR), 2014, 5(1): 36-54.

[10]王鑫恒, 王佐勋, 吕国辉. 输电线路上行波故障信号的去噪及仿真[J]. 齐鲁工业大学学报, 2020, 34(02): 57-62.

[11]程适, 王锐, 伍国华, 等. 群体智能优化算法[J]. 郑州大学学报(工学版), 2018, 39(06): 1-2.

[12]冯增喜, 李诗妍, 赵锦彤, 等. 基于精英反向学习策略的麻雀搜索算法[J]. 计算机仿真, 2023, 40(01): 378-381+393.

[13]圣文顺, 徐爱萍, 徐刘晶. 基于蚁群算法与遗传算法的TSP路径规划仿真[J]. 计算机仿真, 2022, 39(12): 398-402+412.

[14]Wei D, Wang Z, Si L, et al. Preaching-inspired swarm intelligence algorithm and its applications[J]. Knowledge-Based Systems, 2021, 211(1): 55-63.

[15]何小锋. 量子群智能优化算法及其应用研究[D]. 上海理工大学, 2014.

[16]Issa M. Enhanced arithmetic optimization algorithm for parameter estimation of PID controller[J]. Arabian Journal for Science and Engineering, 2023, 48(2): 2191-2205.

[17]Loucif F, Kechida S, Sebbagh A. Whale optimizer algorithm to tune PID controller for the trajectory tracking control of robot manipulator[J]. Journal of the Brazilian Society of Mechanical Sciences and Engineering, 2020, 42(1): 1-5.

[18]李昌聪. 矿井瓦斯抽采泵稳定运行控制系统的研究[D]. 西安科技大学, 2021.

[19]卢晓玲, 马平. 基于粒子群算法的超临界机组给水系统模型辨识[J]. 华电技术, 2015, 37(01): 9-12+76.

[20]Ren H, Shen X, Jia X. A novel dual-biological-community swarm intelligence algorithm with a commensal evolution strategy for multimodal problems[J]. The Journal of Supercomputing, 2021, 77: 10850-10895.

[21]Kusuma P D, Adiputra D. Hybrid Marine Predator Algorithm and Hide Object Game Optimization[J]. Engineering Letters, 2023, 31(1): 1-9.

[22]张长有, 张文宇, 袁永斌, 等. 基于改进MPA优化的高斯混合模型算法[J]. 科技管理研究, 2022, 42(23): 199-208.

[23]张孟健, 王德光, 汪敏, 等. 求解工程约束问题的新型智能优化算法及展望[J]. 计算机应用, 2022, 42(02): 534-541.

[24]吴延凯, 张伟, 王亚刚. 基于GA-PSO的印版滚筒温度二自由度PID参数整定[J]. 包装工程, 2020, 41(05): 185-191.

[25]Padhy S, Sahu P R, Panda S, et al. Marine predator algorithm based PD-(1+ PI) controller for frequency regulation in multi‐microgrid system[J]. IET Renewable Power Generation, 2022, 16(10): 2136-2151.

[26]顾寄南, 潘甜. 基于表面粗糙度的缸体打磨控制专家系统研究[J]. 机电工程, 2018, 35(03): 219-223.

[27]陈勇, 胡进杰, 赵文华, 等. 流平设备的温度专家PID控制系统[J]. 自动化与仪表, 2020, 35(01): 14-17.

[28]张玉涵, 蓝益鹏. 磁悬浮平台直线同步电动机TS型模糊控制的研究[J]. 电机与控制应用, 2021, 48(10): 1-6+13.

[29]王定康, 于丰华, 许童羽, 等. 基于单神经元PID的变量喷雾系统精准控制方法研究[J]. 中国农机化学报, 2022, 43(01): 61-66.

[30]曹江卫, 魏霞. 基于RBF-PID控制器的带式输送机自适应调速系统[J]. 煤矿机械, 2020, 41(05): 203-205.

[31]Moetamedzadeh H R, Khanmirza E, Pourfard A, et al. Intelligent nonlinear model predictive control of gas pipeline networks[J]. Transactions of the Institute of Measurement and Control, 2019, 41(16): 4569-4589.

[32]崔心惠, 李文萱, 朱山川, 等. 多目标海洋捕食者算法的污水处理优化控制[J]. 宜宾学院学报, 2022, 22(06): 13-21+66.

[33]Ma D, Song M, Yu P, et al. Research of RBF-PID control in maglev system[J]. Symmetry, 2020, 12(11): 1780-1787.

[34]任海军, 邓广, 吉昊, 等. 基于人工蜂群-RBF-PID的风力机液压变桨距控制系统设计[J]. 可再生能源, 2019, 37(07): 1078-1083.

[35]丁芳, 李荣荣, 丁政委. 基于同步控制的行李电机ACA-RBF-PID控制[J]. 中国民航大学学报, 2015, 33(01): 41-44+64.

[36]张星原, 龙伟, 万里霞, 等. 基于Smith-RBF-PID控制在台车式电阻炉温控系统中的应用[J]. 制造业自动化, 2015, 37(03): 26-29+46.

[37]顾士成, 袁树杰, 马瑞峰. 矿井通风智能化管控系统设计[J]. 河南科技, 2021, 40(10): 45-47.

[38]王国法. 煤矿智能化最新技术进展与问题探讨[J]. 煤炭科学技术, 2022, 50(01): 1-27.

[39]金科, 黄建功. 《煤矿安全规程》及相关法规[M]. 北京: 中国言实出版社, 2005: 68-72.

[40]邓振宇. 矿井通风系统优化研究[J]. 能源技术与管理, 2020, 45(03): 55-57.

[41]Faramarzi A, Heidarinejad M, Mirjalili S, et al. Marine Predators Algorithm: A nature-inspired metaheuristic[J]. Expert systems with applications, 2020, 152: 113377.

[42]张达敏, 徐航, 王依柔, 等. 嵌入Circle映射和逐维小孔成像反向学习的鲸鱼优化算法[J]. 控制与决策, 2021, 36(05): 1173-1180.

[43]张琳, 汪廷华, 周慧颖. 一种多策略改进的麻雀搜索算法[J]. 计算机工程与应用, 2022, 58(11): 133-140.

[44]Xu F, Ma T. Modeling and studying acceleration-induced effects of piezoelectric pressure sensors using system identification theory[J]. Sensors, 2019, 19(5): 1052-1059.

[45]蔡宇, 刘旭, 程英豪. 线性变参数振动系统的全局辨识[J]. 机械工程学报, 2022, 58(23): 114-122.

[46]李军, 周永言, 刘哲, 等. 在工业过程控制领域:基础控制技术的进展与展望[J/OL]. 自动化学报: 1-18[2023-04-16].

[47]高晴, 张莉, 高田田, 等. 基于PCS7的反应釜炉温控制系统PID参数整定仿真与分析[J]. 计算机与数字工程, 2017, 45(08): 1607-1610+1624.

[48]滕瑞. 高动态热压焊电源及其温度智能控制方法研究[D]. 贵州大学, 2022.

[49]陈铁军, 彭皎龙, 杨阳. 改进ACC算法优化的RBF神经网络研究及其应用[J]. 计算机测量与控制, 2013, 21(05): 1323-1326.

[50]Ahmad Z, Li J, Mahmood T. Adaptive Hyperparameter Fine-Tuning for Boosting the Robustness and Quality of the Particle Swarm Optimization Algorithm for Non-Linear RBF Neural Network Modelling and Its Applications[J]. Mathematics, 2023, 11(1): 242-250.

[51]李丹, 翟震. 改进的RBF神经网络PID算法在电磁振机中的应用[J]. 包装工程, 2019, 40(07): 192-196.