基于电力需求响应的板坯热轧负荷分析与调度

发布时间：2018-04-21 07:16

本文选题：热轧 + 负荷调度　；参考：《湘潭大学》2015年博士论文

【摘要】：电力供需平衡是电力系统运行调度的核心问题,风电、光伏等清洁低碳可再生新能源的快速增长给电力系统传统的运行调度和控制模式带来了严峻的挑战。近年来国际上提出了基于需求响应促进电力平衡和消纳新能源的概念。需求响应模式下,电力用户对电网给出的价格信号或激励信号作出响应,改变其原有的电力消耗模式从而参与电力系统运行。电力用户侧负荷资源具有灵活部署和快速响应的特点,因此能够响应发电侧出力的变化,减小可再生能源接入造成的电网波动,维持电力供需平衡。高耗能工业电力负荷参与需求响应潜力巨大,相关研究与实践逐步得到重视,然而,目前工业电力需求响应相关研究主要集中在连续过程生产上。与连续过程生产相比,间歇过程的生产单元具有强耦合性,生产过程需保证工艺的完整性和物流的连续性,需求响应不能影响正常生产,传统的直接负荷控制、可中断负荷等需求响应负荷控制方法无法直接应用间歇生产过程中。本论文工作将钢铁电力大用户的间歇生产负荷资源纳入电力系统调度体系,通过生产负荷的主动缩减和峰谷转移来减小电网功率波动,促进电力供需平衡,对生产用户而言则可以减小用电成本。具体以板坯热轧这一典型间歇生产过程为研究对象,基于材料加工过程和生产工艺流程分析板坯热轧的负荷特征,建立需求响应负荷调度与热轧生产控制间的关联,在保证生产工艺约束前提下,研究兼顾用电经济性目标与生产工艺目标的热轧负荷调度方法,减小企业用电成本,拓展电力需求响应负荷类型,促进电网稳定和新能源消纳。论文具体研究内容包括:1.分析了板坯热轧加工过程机理及轧制参数数值计算模型,重点研究了板坯热轧过程的轧制形变及工艺参数模型、轧制力模型和温度模型。通过机理分析和数学推导确立轧制负荷与轧机、轧件相关参数间的数值计算关系,建立轧制过程的负荷数值计算模型,并实现Matlab/Simulink环境下的板坯热轧道次负荷仿真计算平台,基于该平台进行热轧过程负荷参数化分析,确定不同工艺参数与轧制负荷的相关关系及其对轧制负荷影响的敏感度。2.研究了以最小化轧制能耗为目标的中厚板轧制规程优化设计方法。基于道次间压下率控制的轧制负荷分配方法,结合轧制过程负荷模型,建立了中厚板最小能耗轧制规程优化设计模型。针对中厚板轧制生产特点及其对算法全局搜索能力和收敛速度的要求,提出一种改进粒子群优化(IPSO)求解算法用于模型求解,实验结果表明,IPSO应用于轧制规程优化,能耗明显降低,算法全局搜索能力强且收敛速度快,综合性能优于遗传算法、标准PSO及几种常用改进算法。方法不仅为轧制过程的电力负荷缩减提供技术指导,还可用于支撑轧制过程电力负荷预测。3.研究了热轧生产控制与需求响应负荷调度的关联及优化调度方法。在价格型需求响应环境下,考虑最小化生产用电成本,基于热轧批量计划问题的车辆路径规划模型,提出一种面向经济负荷调度的热轧调度两阶段优化方法,第一阶段通过热轧批量计划编制确定轧制单元及生产负荷,第二阶段在此基础上实施分时电价下的负荷转移重调度,优化轧制单元加工次序和生产空闲等待时间进行避峰生产。实验结果表明:所提出方法可降低生产用电成本,并通过响应电价参与电网调峰。4.提出了基于价格响应的板坯热轧负荷最优调度模型及求解算法。针对两阶段优化中热轧批量计划与需求响应负荷调度分离,难于实现最优调度结果的问题,建立热轧生产调度多目标集成优化模型,在实施最小化用电成本的经济负荷调度时同步考虑相邻板坯间惩罚值。提出一种基于NSGA-II的多目标生产调度优化算法进行模型求解,针对多目标优化Pareto最优解集,考虑目标偏好性,设计了用于不同生产场景下工程最优解选取的三种决策模式。选取多种对比方法从不同角度与本章所提出方法对比,结果表明所提出模型有效,算法性能良好。5.面向间歇生产过程中普遍存在的多产线并行生产环境,研究了基于价格需求响应的多机并行间歇负荷调度的泛化模型。在产线能力满足生产需求的前提下,以最小化生产用电成本为目标,建立由整数变量和连续时间变量表达的需求分批、产线分配、批量调度间约束关系,并针对模型中的非线性生产约束条件进行线性化转换,构建问题的MILP模型并进行实例求解。将求解结果与广泛采用的最早完工时间模型进行对比,结果表明模型具有较好的用电成本节省和负荷调峰效果。
[Abstract]:Power supply and demand balance is the core problem of power system operation and scheduling. The rapid growth of clean low carbon renewable energy, such as wind power and photovoltaic, has brought serious challenges to the traditional operation scheduling and control mode of power system. In recent years, the concept of demand response to promote power balance and eliminate new energy is proposed in recent years. In response to the mode, the power users respond to the price signals or incentive signals given by the power grid, change their original power consumption mode and participate in the operation of the power system. The power user side load resources have the characteristics of flexible deployment and rapid response, so it can respond to the change of power generation force and reduce the access to renewable energy. The power supply and demand balance of the power grid is maintained. The demand response potential of the high energy consuming industrial power load is great, and the related research and practice are gradually paid attention. However, the related research on the response of industrial electric power demand is mainly concentrated on the continuous process production. The process of production needs to ensure the integrity of the process and the continuity of the logistics. The demand response can not affect the normal production. The traditional direct load control, the interruptible load and other demand response load control methods can not be directly applied in the intermittent production process. This paper puts the batch production load resources of the iron and steel power large users into the power system adjustment in this paper. Through the active reduction of production load and peak valley transfer to reduce the power fluctuation of the power grid and promote the balance of power supply and demand, the cost of electricity consumption can be reduced for the production users. The typical batch process of slab hot rolling is studied, and the load of hot rolled slab is analyzed based on the process of material processing and the process flow. The relationship between demand response load scheduling and hot rolling production control is established. Under the premise of ensuring production process constraints, the hot rolling load scheduling method which takes both electricity economic targets and production process targets is studied, reducing the cost of electricity consumption, expanding the response load type of power demand, and promoting the stability of the power grid and the new energy dissipation. The research contents include: 1. the mechanism of hot rolling process and the numerical calculation model of rolling parameters are analyzed. The rolling deformation and process parameters model, rolling force model and temperature model are emphatically studied. Through mechanism analysis and mathematical deduction, the numerical calculation between rolling load and rolling mill and the relative parameters of rolling parts is established. Set up the load numerical calculation model of rolling process and realize the simulation platform for the secondary load of slab hot rolling under the Matlab/Simulink environment. Based on this platform, the load parameterization analysis of the hot rolling process is carried out, and the relationship between the different process parameters and the rolling load and the sensitivity of the rolling load to the rolling load are determined to the minimum.2.. The optimization design method for the rolling regulation of medium and thick plates is aimed at the energy consumption of the rolling mill. The rolling load distribution method based on the rate controlled by the inter channel compression rate and the rolling process load model are combined to establish the optimal design model of the minimum energy consumption rolling regulation for medium and thick plates. The characteristics of the rolling production and the global searching ability and convergence speed of the medium and thick plate rolling are given. An improved particle swarm optimization (IPSO) algorithm is proposed to solve the model. The experimental results show that IPSO is applied to the rolling regulation optimization, the energy consumption is obviously reduced, the global search ability is strong and the convergence speed is fast. The comprehensive performance is superior to the genetic algorithm, the standard PSO and several commonly used improved algorithms. The force load reduction provides technical guidance, and can also be used to support the rolling process power load forecasting.3. to study the relationship between the hot rolling production control and demand response load scheduling and the optimal scheduling method. Under the price demand response environment, the vehicle path planning model based on the problem of hot rolling batch planning is proposed to minimize the cost of production power consumption. A two stage optimization method for hot rolling scheduling is developed for economic load scheduling. The first stage is to determine the rolling unit and production load through the hot rolling batch plan. The second stage is based on the load transfer redispatch under the time-sharing price, optimizing the processing order of the rolling unit and producing the idle waiting time to avoid the peak production. The results show that the proposed method can reduce the cost of production power consumption, and the optimal scheduling model and solving algorithm of slab hot rolling load based on price response are proposed by the response to electricity price in the power grid peak.4.. The problem of optimal scheduling results is difficult to achieve in the two stage optimization of hot rolling batch planning and demand response load adjustment. A multi-objective integrated optimization model for hot rolling production scheduling is established. A multi objective scheduling optimization algorithm based on NSGA-II is proposed to solve the penalty value between adjacent slab in the implementation of economic load scheduling to minimize the cost of electricity consumption. A multi objective optimization algorithm is proposed to optimize the Pareto optimal solution set and consider the preference of the target. In different production scenarios, three decision models are selected for the optimal solution of engineering. A variety of contrast methods are selected from different angles to the proposed method. The results show that the proposed model is effective and the performance of the algorithm is good.5. is oriented to the multi production line parallel production environment which is common in the batch production process, and the response of the price demand response is studied. A generalization model of multi machine parallel intermittent load scheduling. On the premise of line capacity satisfying production demand, in order to minimize the cost of production power consumption, the constraint relation between demand batch, line distribution and batch scheduling is established, which is expressed by integer variables and continuous time variables, and the nonlinear production constraints in the model are linear. The MILP model of the problem is constructed and the case is solved. The results are compared with the earliest completed time model. The results show that the model has better cost saving and load peak adjustment effect.

【学位授予单位】：湘潭大学
【学位级别】：博士
【学位授予年份】：2015
【分类号】：TG335.11

【参考文献】