【作者】 张伟峰；

【导师】 孔金生；

【作者基本信息】 郑州大学 ， 控制理论与控制工程， 2012， 硕士

【摘要】 焦炭的生产过程是一个严重的时变、非线性、大滞后、多参数和强耦合大工业过程。焦炭作为高炉冶铁的主要原材料和重要燃料,被誉为钢铁企业的“基本粮食”,其质量的优劣对各个用焦行业后续生产的稳定性意义重大。因此,为了保证焦炭质量,建立焦炭质量模型,实施焦炭生产过程产品质量优化控制具有十分的意义。对于焦炭的生产过程,产品影响因素众多,不仅有原料和生产工艺的影响,而且还有生产过程中产生的各种噪声也会对其质量产生影响,因此很难准确的对其进行质量建模与产品质量优化控制。论文在对焦炭产品生产过程及其机理分析的基础上,对焦炭质量指标和影响焦炭质量的因素进行了深入剖析和探讨,提出了焦炭生产过程产品质量建模和质量优化问题。给出了基于GA优化的BP神经网络的焦炭质量模型建模方法。通过实际的生产数据对模型进行仿真和分析,分析结果表明,模型具有很高的预测精度,预测的命中率能够达到焦化厂的实际要求。对焦炭生产过程产品质量优化问题进行深入探讨,在对改进粒子群算法及其优化分析的基础上,给出了基于改进粒子群算法的焦炭生产过程产品质量优化方法。仿真结果表明,所使用的方法有效可靠的解决了这一优化问题,为焦炭行业的优化控制提供了有效的理论依据。更多还原

【Abstract】 The production process of coke is complicated large-scale system composed of multi-subsystem, which has the characteristic of severe non-linearity, uncertainty, time-varying, large-delay, strong coupling and multiple-parameter. its main product is coke. Coke is the main fuel and raw materials in the process of iron production in the blast furnace, plays an important role of a chemical reducing agent and permeable support in the skeleton. Coke is known as "the basic food of iron and steel enterprises" The quality of coke industry in each subsequent production stability has important significance. Therefore in order to ensure the realization of coke quality, coke production process optimization control, establishment of coke quality model is very meaningful. For coke production process, the product of many factors, because of its production process complex together with industrial noise pollution, the quality of modeling and optimization control have great difficulty.The paper is based on the production process and mechanism analysis of coke product, analyses and discusses the indicators of coke quality and the factors that affect the coke quality deeply, and propose the question of the production quality modeling and quality optimization of the coke production process.Then the paper gives BP neural network modeling approach of the coke quality which is based on based on GA optimization. Then simulates and analyses the model through the actual production data, the results show that the model has high prediction accuracy and good predict effectes, and the predicted hit rate can reach the actual requirements of the coking plant.The problem of quality optimization for the production process of coke are discussed deeply. Based on the improved particle swarm algorithm and its optimization analysis, this paper proposes the optimization method, which is based on Improved Particle Swarm Optimization Algorithm, for the coke production process product quality. The simulation results show that the proposed method is an effective and reliable solution to this optimization problem. It is effective theoretical basis to the coke industry optimization control provides an effective theoretical basis更多还原