蒸汽动力系统优化设计与运行集成建模及求解策略的研究

【作者】 罗向龙；

【导师】 尹洪超；

【作者基本信息】 大连理工大学 ， 热能工程， 2004， 硕士

【摘要】 过程工业蒸汽动力系统是过程工业的重要组成部分，它的安全、稳定运行是企业安全、稳定、长周期运行的基础。蒸汽动力系统的设计水平、运行和控制性能对过程工业的能量利用率和经济性具有重要影响，研究蒸汽动力系统设计、运行、控制各层次的集成建模优化对于指导蒸汽动力系统全面优化具有重要意义。 (1)本文在总结前人在蒸汽动力系统优化的工作成果的基础上，针对蒸汽动力系统的特点，将过程能量系统多层次集成建模的思想应用到蒸汽动力系统，阐述蒸汽动力系统的物理结构、流结构、功能层次结构模型，对蒸汽动力系统中包括锅炉、汽轮机、燃气轮机、余热锅炉等主要设备建立适用于集成优化求解的数学模型。 (2)在研究蒸汽动力系统最优设计和优化运行模型的基础上，建立蒸汽动力系统设计和运行同步优化的多周期集成模型，模型综合考虑了设计阶段投资折旧费用函数的非线性、各设备运行过程中的启停费用、锅炉效率的变化以及汽轮机做功过程的非线性等因素。 (3)随着周期和设备的增多，蒸汽动力系统多周期最优设计与运行集成模型的求解更加困难，传统的各种求解算法均不能有效的求解。本文对遗传算法基本原理、优缺点等进行了讨论，针对蒸汽动力系统集成模型自身的多周期和非线性的特点，提出了改进的遗传算法，改进后的遗传算法在初始解群的生成、约束条件的处理、遗传算子等方面都做了相应的改进，并引入了边界搜索的方法来优化搜索路径。改进后的遗传算法更加简单实用，能够快速求得最优解。 (4)采用改进的遗传算法分别对蒸汽动力系统优化调度算例和设计与运行同步优化算例进行计算，并将优化调度算例结果与不考虑锅炉效率变化和汽轮机做功非线性的优化调度线性模型计算结果进行了比较，比较结果表明蒸汽动力系统优化调度过程中考虑锅炉效率的变化和汽轮机做功的非线性因素是非常有必要的；利用线性规划软件LINGO5.0对中石油某炼化公司蒸汽动力系统运行计划进行了优化，得到了令人满意的优化运行计划。通过以上优化计算，证明了本文提出的蒸汽动力系统集成模型及求解策略的有效性和实用性。更多还原

【Abstract】 Steam power system is an important part of the process industry, whose secure and steady operation is the foundation of security, stabilization and long period operation for the whole corporation. The design, operation and control of steam power system affect the effective utilization of energy and economic performance in the process industry. Therefore, the study on the design, operation and control is very important to direct the general optimization of steam power system.(1) In the dissertation, the literature about optimal design, optimal operation and optimal control of steam power system were summarized. The physical structure, flow structure and hierarchical structure of steam power system were expatiated. Besides, the performance models of boiler, steam turbine, gas turbine, steam converter valve and heat recovery steam generator (HRSG) were gotten.(2) Based on the design model and operation model of steam power system, the integrated model of design level and operation level was constructed according to the integrated modeling theory. In this model, the non-linear performance of equipment investment cost, the changes of boiler efficiency and the non-linear performance of steam turbine working were considered. Besides, the cost of changeover between periods of operation was considered in the model.(3) In the model mentioned above, the possible combination dimensions of the system will show exponential growth with the number of unit and periods. Usual algorithm cannot solve the problem effectively. In order to overcome this difficulty, the improved genetic algorithm was introduced. The improved genetic algorithm was improved on many aspects and it was much more simple than before. Using this method to solve the multi-period MINLP problem, the optimal result could be achieved within reasonable time.(4) Based on the integrated modeling theory, two examples were studied. One was the short-term optimal multi-period operational planning problem. The other was the simultaneous optimization of the design and operation for a steam power system. Two MINLP models were constructed and solved by the improved genetic algorithm, which indicated that the integrated model and solving strategy were effective. In the last part of the dissertation, a MILP optimal operational planning model was constructed and solved by LINGO5.0 for an actual steam power system. The satisfying one-year operational planning of the steam power system was gotten, which testified that the integrated model was practicable.更多还原