【作者】 钟炜；

【导师】 钟登华；

【作者基本信息】 天津大学 ， 工程管理， 2007， 博士

【摘要】 本论文以网络环境下构建流域梯级水电站联合优化调度辅助决策系统为研究对象,将智能优化和网络通讯等技术应用于流域梯级水电站的信息管理、水文预报及优化调度决策中,研制开发了较为实用的流域梯级水电站联合优化调度辅助决策系统。系统可对生成的调度方案进行合理评价和优化决策,为实现梯级水电站调度方案实施的智能化、信息化和网络化管理提供了理论与技术支持。论文主要研究内容及成果如下:(1)针对流域内降水及下垫面条件时空分布不均匀的特点,提出了水库入库径流分级预测方法。通过对采集的水文序列进行时空分析,建立了分级预报因子集特征值变量与观测值对应的小波神经网络径流预测模型。设计了预测模型分类识别器,根据不同的径流特征值自动搜寻相适应的网络模型进行预测,最终将分类预测模型合并为组合预测模型。(2)研究了市场机制下梯级水电站优化调度模型的建模方法及针对梯级水电站厂内经济运行优化调度模型的建模方法,并提出相应的模型求解算法。其中,厂内经济运行调度模型求解采用本论文提出的基于遗传算法的双层耦联算法,设计上层为求解机组负荷分配问题模块,下层为求解机组组合问题模块。该算法具有可同时求得电厂负荷分配问题和机组组合问题的最佳调度方案的优点。(3)建立了基于多目标决策模糊优选理论的梯级水电站调度决策方案评估模型。方案评估以调度优化目标相应的评估指标为依据,采用多目标模糊优选法对生成各调度方案进行排序,模型中引入调度规则集,通过规则重要性标度计算得到方案优劣性评价成果,为梯级水电站调度方案评估模型提供评价依据,并提出了针对水电站调度决策的加权推理机制指导生成最优调度方案的算法和程序。(4)研制开发了网络环境下的流域梯级水电站优化调度辅助决策系统软件。系统基于上述所提出的理论方法和技术,结合具体软件研发目标,并对系统功能需求进行了深入的分析;采用B/S网络构架模型和软件分层服务模式,设计开发出了系统的总体框架和各个功能模块,并对各个模块的实现过程进行了详细的说明。所研制的系统软件在实际调度工程中的应用,可为决策人员提供完善的信息数据和科学的调度方案,提高企业的信息综合管理水平和调度自动化实施能力。更多还原

【Abstract】 The thesis takes the establishment of united optimal operation aid decision system with web-based method for a basin cascade plants as the objective of this research. With the technologies of intelligence optimization and network information applied into the aspect of the basin cascade hydropower plants, such as information management, hydrological forecasting and optimal operation, a comparatively practical united optimal operating aid decision system of the basin cascade hydropower plants was developed. This system can carry on rational appraisal and optimal decision to the producing operation schemes, and offers theoretical foundation and technical support for the operation scheme of the basin cascade hydropower plants, which realized intelligence, informationization and network management.The main research contents and achievements of the thesis are as follows:(1) Aiming at the heterogeneous distribution characteristic of the time-space rainfall and the underground cushion condition in the basin, a classified forecasting method for putting the runoff into the reservoir was proposed. Through carrying on the time-space analysis to the gathered hydrology series, the wavelet neural network forecasting model corresponding to the characteristic value variable of the classified forecasting factor collect with the observed values was established, and the classified recognizer of model for according to the different runoff characteristic value searching automatically adapting network model to forecast was designed. Finally, inserted the classified forecasting model merges into mixed model to set up the united runoff forecasting model of the reservoir.(2) Studied systematically the building-model methods of the optimal operation model for the basin cascade hydropower plants preceded under the condition of market mechanism and a hydropower plant produced in the economical operation, and presented a set of efficient algorithm to solve the corresponding model. Among them, the solution of economical optimal operation model in a hydropower plant was applied in this text putting forwarded that the double-deck coupling algorithm which based on the improving genetic algorithm, where the upper-deck module was designed as to solve the load optimal distributing question among the cascade hydropower plants and the lower-deck module was designed as to solve the generator group optimal-uniting question among the generators in a hydropower plant. This algorithm has advantages in getting the optimal operation schemes of the load distribution, as well as the generator group unitization at the same time.(3) Established an optimal operation evaluation model of the basin cascade hydropower plants based on the theory of multi-objective fuzzy optimal selection. In this evaluation model that the scheme evaluation indicator is corresponding with the optimal operation object, the produced operation schemes was applied to the multi-objective fuzzy optimal selection method to proceed the orders, a number of the characteristic values was used to describe the merits of the operating results and the evaluation indicators including the introduction of Weight Evaluation Model. The evaluation model offers an advanced method to decide the optimal operation scheme of the basin cascade hydropower plants presented the reasoning mechanism derived by the weighted method, so as to guide the optimal operating model generating the actual optimal operating program.(4) Developed a united optimal operating aid decision system software with web-based method for the basin cascade hydropower plants. This system software was designed according to the theoretical method and technology mentioned above, combined the concrete optimal operating goals to deeply analyze the functional demands of the system, used the B/S network architecture model and the software layered service modes to build the systems overall framework and the various functional modules, also carried on the detailed explanation to the realization course of each module. Apply the developed system software to the realistic project that can provide perfect information data and the scientific operation scheme for policy-makers, and improve the comprehensive management level of information and the implement competence of operation automation for the enterprise.更多还原