矿井通风系统风流参数动态监测及风量调节优化.pdf

博士学位论文 矿井通风系统风流参数动态监测及风量调节优化 Dynamic Monitoring of Airflow Parameters and Air Quantity Regulation Optimization for Mine Ventilation System 作 者司俊鸿 导 师陈开岩 教授 中国矿业大学 二○一二年十月 中图分类号 TD72 学校代码 10290 UDC 622.4 密 级 公开 中国矿业大学 博士学位论文 矿井通风系统风流参数动态监测及风量调节优化 Dynamic Monitoring of Airflow Parameters and Air Quantity Regulation Optimization for Mine Ventilation System 作 者 司俊鸿 导 师 陈开岩 教授 申请学位 工学博士 培养单位 安全工程学院 学科专业 安全技术及工程 研究方向 矿井通风与安全 答辩委员会主席 王德明 评 阅 人 二○一二年十月 论文审阅认定书 论文审阅认定书 研究生 司俊鸿 在规定的学习年限内，按照研究生培养方案的要求， 完成了研究生课程的学习， 成绩合格； 在我的指导下完成本学位论文,经审阅， 论文中的观点、数据、表述和结构为我所认同，论文撰写格式符合学校的相 关规定，同意将本论文作为学位申请论文送专家评审。 导师签字 年 月 日 致 谢 致 谢 论文是在陈开岩教授的悉心指导下完成的， 没有陈开岩教授的指导、 鼓励及大力支持， 论文是不可能得以顺利完成的。师从五载，收获颇丰，感触亦深。陈开岩教授优秀的做人 品质、 严谨的治学态度、 开拓创新的精神、 高屋建瓴把握全局的能力以及忘我的工作精神， 给学生树立了潜移默化的典范作用，这也是导师传授给学生最宝贵的财富。在此，谨向我 的导师致以深深的敬意。 感谢安全工程学院蒋曙光教授、 张人伟副教授、 吴征艳副教授对论文选题提出了宝贵 的意见和建议。 感谢盘江集团张仕和董事长， 盘江精煤股份公司徐再刚总工程师， 通风处麻竹林处长、 旷永华副处长，金佳矿罗卫东总工程师、通风区蔡波区长、余德胜、杨彧、钱燕飞、羊超 等在现场调研、现场实测、数据采集和应用方面的支持和帮助。 感谢同门师兄弟蒋中承、张宝、王军、邵和、郭千枫、黄帅、宋凯、陈辉、赵红梅等 人的大力支持和帮助。 感谢年迈的父母、姐姐以及贤惠的妻子，对我学习的理解、督促和鼓励。 再一次向所有帮助、支持、关心过我的人们致以诚挚的谢意 最后，向百忙之中评审本文并提出宝贵意见的各位专家致以崇高的敬意 I 摘 要 摘 要 合理的矿井通风系统风流参数是保障矿井安全生产、 改善井下人员劳动安全健康条件 和防灾抗灾的最重要环节。 随着矿井监控技术的发展， 通风安全监控设备在煤矿安全生产 中发挥着越来越重要的作用。 针对目前矿井通风系统风流参数监测、 风量调节优化中存在 的问题， 论文系统研究矿井通风网络风流参数动态监测与风量调节优化理论技术， 为实现 矿井自动化进程、保障矿井安全高效集约化开采提供了基础理论依据。 通过对测风求阻的改进方法进行研究，提出了基于最少调节次数的最优回路选择算 法，利用通风网络风流参数监测数据，建立了复杂矿井通风网络测风求阻改进模型，提出 采用贪婪进化算法与广义交叉检验法相结合的 Tikhonov 正则化算法进行求解。根据复杂 矿井通风系统全局调控与局部调控相结合的调节原则， 提出了基于风机工况调节与井下巷 道风阻调节相结合的测风求阻法调节选址方法， 确定测风求阻模型的合理调节位置。 在矿 井通风网络风流参数动态监测研究的基础上， 提出采用冗余传感器平差处理方法和标准差 滤波方法进行监测数据预处理，提高监测数据的准确度，减少扰动对计算结果的影响。 提出了矿井通风网络风流参数无盲区监测时监测点优化选址方法， 以矿井通风风流参 数（风压和风速）监测点最少量、满足通风网络全部分支风阻实时解算需要、风流参数无 盲区监测为目标函数， 以通风网络回路风压平衡方程及节点风量平衡方程为约束条件， 采 用分步优化解算的思想， 建立了基于可变模糊理论的风速传感器优化选址模型和监控系统 线缆优化布置模型，确定了风流参数传感器的安装位置、传感器与监控分站的隶属关系、 监控主线的最优铺设方案。 建立了以调节设施数量最小和主要通风机总功率最小为目标函数， 以非定流余树分支 风量和可调节分支风阻调节量为决策变量，以巷道风量、用风地点需风量、矿井通风总阻 力、可调分支阻力调节量上下限为不等式约束条件，以通风网络回路风压平衡方程、节点 风量平衡方程、 分支阻力方程及风机特性曲线方程为等式约束条件的矿井通风网络风量调 节非线性优化模型，提出了基于贪婪进化算法和关键路径法相结合的求解算法（GECP 算 法） ，获得了矿井通风系统安全、合理、经济运行的优化调控方案。 采用 Microsoft Visual C和 Microsoft SQL Server 数据库混合编程的方法设计编制出 矿井通风系统动态监测及风量调节优化软件， 实现了通风网络监测设备选址、 监测数据查 询、 测风求阻计算以及通风网络风量调节优化等功能。 将上述成果在贵州盘江煤电集团有 限责任公司金佳煤矿进行了试验研究。 该论文有图 61 幅，表 50 个，参考文献 163 篇。 关键词关键词通风网络；风量调节优化；传感器选址；测风求阻；贪婪进化算法 II Abstract The reasonable value of airflow parameters is the most important parts of protecting the mine production safety, improving the underground safety and health conditions of staffs as well as preventing the disasters in mine ventilation system. With the development of mine monitoring technology, the monitoring equipment plays a fatal role in keeping the mine ventilation system operating safely and stably. Aiming at the problems of airflow parameters monitoring and air quantity regulation optimization of mine ventilation system, this paper systematically studies the theory and technology of dynamic monitoring and air quantity regulation optimization. It provides the theoretical basis for the realization of mine automation process, protection of the safety, intensive and efficient exploitation of mine. According to the measuring airflow the constraints obey the Kirchhoff’s current law and Kirchhoff’s voltage law. Adopting the step by step optimization solution , the model of air velocity transducer arrangement based on variable fuzzy optimization theory and the optimal distribution style of monitoring cable are put forward for determining the transducer location of airflow parameters, the affiliation of transducer and monitoring substation, as well as the optimal laying plan of monitoring cable. III This paper establishes the mathematical model of air quantity regulation optimization of mine ventilation network. The objective function is composed of the fewest regulating locations and the smallest air power of main fans; the variables are mixing of the regulating air resistance of branches and the air quantity of the cotree branches which are not the constant branch; the inequality constraint conditions include the lower and upper bound of the air velocity, the regulating air resistance of the roadway, the required air quantity of the chamber, and the fan pressure; the equality constraint conditions obey the Kirchhoff’s current law and Kirchhoff’s voltage law. Furthermore, the combination of greedy evolution and critical path GECP is proposed in for solving the air quantity regulation optimization model. It is feasible to obtain an optimal air quantity regulating scheme which ensures mine ventilation system operating safely, economically and reliably. Using the hybrid programming approach of Microsoft Visual C and Microsoft SQL Server, the software of mine ventilation system dynamic monitoring and optimal regulation is developed, which has the following functions, such as location of airflow monitoring equipment of mine ventilation network, query of the monitoring data, calculation of the MAER model, and air quantity regulation optimization of mine ventilation network. And the practical experiment in Jinjia coal mine of Guizhou Panjiang Refined Coal Co., ltd. is completed. This paper has 61 figures, 50 tables, and 163 references. Keywords ventilation network; air quantity regulation optimization; transducer location; measuring airflow greedy evolution algorithm IV Extended Abstract The coal mine monitoring system is an important measure to protect the coal mine safety in production. With the development of mine monitoring technology, the monitoring equipment plays a fatal role in keeping the mine ventilation system operating safely and stably. However, there still have lots of problems in mine ventilation system, for example, it is complex to obtain the airflow parameters, it is blinding to make the airflow regulating plan, and it is relatively weak to research the airflow parameters monitoring theory, etc. This paper systematically studies the theory and technology of dynamic monitoring and air quantity regulation optimization adopting mine ventilation theory, computational fluid mechanics, graph theory, nonlinear multivariate analysis, software engineering and other related theories, techniques and s, on the basis of summary and analysis of the existing research findings at home and abroad. The main research contents and findings achieved 1 Using the greedy search strategy, GE algorithm is put forward to calculate the constrained nonlinear optimization based on DE algorithm. According to the population classification, GE retains the best vectors in each generation, uses the newly generated vector in the neighborhood to replace the parent vector of the better vectors if the new one is better than the parent vector, and regenerates the poor vectors to ensure the diversity of the population. The mutation operator is introduced in the better and poor vectors of GE algorithm. The objective vector will be replaced immediately only if the new vector is better. The reasonable value range of proper parameters of GE is obtained the best value range of nps is 1 2 times than np, sp is 0.4 0.8 and δ is 0.05 0.3. 2 According to the dynamic and complex characteristics of mine ventilation system, the algorithm of choosing the optimal circuit for the minimum regulation number of MAER model is put forward. Combining with the real-time dynamic monitoring data of the air quantity, this paper establishes a modified MAER model of mine ventilation network. Because it is easily construct the ill-posed coefficient matrix while the model applied to the complex network, the Tikhonov regularization is proposed to revise the model based on the combination of the greedy algorithm and generalized cross-validation . In order to improve the accuracy of the monitoring data and reduce the disturbance to the measurement and calculating results, the adjustment of redundant air velocity transducer measurement and the standard deviation filter are put forward based on the theoretical study of the dynamic airflow parameters of mine monitoring system. Using the hybrid programming approach of Microsoft Visual C and Microsoft SQL Server, the software of V mine ventilation system monitoring data query and calculating software is developed. 3 Analysis of the interaction of diagonal structure and airflow stability, and the multi-fan combined operation in complex ventilation system, this paper puts forward the classification and hierarchical principles of the diagonal branch, independent the constraints obey the Kirchhoff’s current law and Kirchhoff’s voltage law. Adopting the step by step optimization solution , the model of air velocity transducer arrangement based on variable fuzzy optimization theory and the optimal distribution style of monitoring cable are put forward for determining the transducer location of airflow parameters, the affiliation of transducer and monitoring substation, as well as the optimal laying plan of monitoring cable. Furthermore, the reasonable air velocity transducer arrangement based on variable fuzzy optimization theory are put forward on the basis of the MAER . Then, the solution of the model as well as computer calculating software is developed. 5 This paper establishes the mathematical model of air quantity regulation optimization of mine ventilation network. The objective function is composed of the fewest regulating locations and the smallest air power of main fans; the variables are mixing of the regulating air resistance of branches and the air quantity of the cotree branches which are not the constant branch; the inequality constraint conditions include the lower and upper bound of the air velocity, the regulating air resistance of the roadway, the required air quantity of the chamber, and the fan pressure; the equality constraint conditions obey the Kirchhoff’s current law and Kirchhoff’s voltage law. Furthermore, s of reducing dimensions of nonlinear VI optimization model are analyzed for increasing the convergence rate and convergence degree of the algorithm. Based on the greedy search strategy and critical path regulation strategy, the combination of greedy evolution and critical path GECP is proposed in for solving the air quantity regulation optimization model. On this basis, the computer software of air quantity regulation optimization is developed, which automatically generates the model by using the relevant standards and rules of mine air quality requirement calculating , the upper and lower of the air quantity and the regulating air resistance of the different roadway. 6 The practical experiment in Jinjia coal mine of Guizhou Panjiang Refined Coal Co., ltd. shows that using the theory of optimal transducer location of airflow parameters for mine monitoring system and the modified MAER model for complex mine ventilation system, it realized the monitoring of airflow parameters dynamically without blind monitoring zone. According to the basic data of the airflow parameters for mine ventilation network in Jinjia coal mine, the air quantity regulation optimization model is established. The plan of air quantity regulation optimization is acquired by using the GECP algorithm. It is successful to demonstrate the correctness and feasibility of dynamic monitoring of airflow parameters for mine monitoring system and air quantity regulation optimization of mine ventilation network that ensures mine ventilation system operating safely, economically and reliably. Keywords ventilation network; air quantity regulation optimization; transducer location; measuring airflow greedy evolution algorithm VII 目 录 目 录 摘 要..................................................................... I摘 要..................................................................... I 目 录................................................................... VII目 录................................................................... VII 图清单................................................................... XII图清单................................................................... XII 表清单................................................................... XVI表清单................................................................... XVI 变量注释表............................................................... XIX变量注释表............................................................... XIX 1 绪论..................................................................... 11 绪论..................................................................... 1 1.1 研究的背景及意义........................................................ 1 1.2 国内外研究现状.......................................................... 3 1.3 主要研究内容........................................................... 13 1.4 论文研究的关键技术路线................................................. 13 2 基于贪婪搜索和邻域搜索策略的改进 DE 算法研究 ............................. 162 基于贪婪搜索和邻域搜索策略的改进 DE 算法研究 ............................. 16 2.1 微分进化算法基本原理................................................... 16 2.2 贪婪进化算法基本原理及程序设计......................................... 17 2.3 基于马尔可夫链分析的 GE 算法收敛性证明.................................. 21 2.4 GE 算法仿真实验 ........................................................ 22 2.5 GE 算法与 DE 算法特性比较 ............................................... 25 2.6 GE 算法控制参数取值范围 ................................................ 25 2.7 本章小结............................................................... 28 3 复杂通风网络测风求阻改进模型及求解算法 .................................. 293 复杂通风网络测风求阻改进模型及求解算法 .................................. 29 3.1 通风网络测风求阻基本模型............................................... 29 3.2 测风求阻模型病态特性判别............................................... 29 3.3 测风求阻模型最优回路选择............................................... 30 3.4 矿井通风系统调节位置与风流稳定性分析................................... 32 3.5 矿井通风网络测风求阻法调节位置......................................... 43 3.6 基于 Tikhonov 正则化方法的测风求阻模型求解.............................. 47 3.7 测风求阻法计算机程序设计及实例分析..................................... 48 3.8 本章小结............................................................... 53 4 矿井通风网络风流参数监测设备选址及动态监测 .............................. 544 矿井通风网络风流参数监测设备选址及动态监测 .............................. 54 4.1 矿井监控系统分站与线缆优化布置......................................... 54 4.2 矿井监控系统风流参数传感器选址的基本原则............................... 58 VIII 4.3 基于测风求阻模型的风流压差传感器优化选址............................... 58 4.4 矿井监控系统风速传感器优化选址......................................... 59 4.5 监测数据滤波处理技术................................................... 67 4.6 矿井风流多参数无盲区监测传感器优化布置................................. 69 4.7 求解巷道风阻的通风网络风流参数动态监测实例分析 ......