【作者】 李晓军；

【导师】 谭忠富；

【作者基本信息】 华北电力大学（北京） ， 技术经济及管理， 2007， 博士

【摘要】 打破垄断，建立竞争性的电力市场是我国电力体制改革的目标。市场化的交易机制将改变传统的电力工业运营模式，发电商、供电公司、电网公司和用户都将成为独立的市场主体参与市场交易。电力体制改革的目的就是为了建立竞争性的电力市场，通过各种交易方式，以电价为调节手段，实现电力资源的优化配置。本文针对市场环境下电力产业中的交易问题及其优化模型进行研究，本文主要工作如下:研究了不同电力产业运营模式下的电力产业价格链形态以及相应的价格形成机制。我国电价形成机制由过去单一的政府定价逐渐发展成政府监管定价、市场竞价和合同定价的多种定价机制。针对我国电力产业价格链和电力改革的现状，提出了完善我国电力产业价格链，降低我国电价水平的一些措施和途径。介绍了目前发电企业与电网企业的主要交易方式和结算方式，介绍了机组的发电成本构成，并分析了影响发电成本的主要因素。针对单一买方的电力市场建立了发电企业与电网企业交易的主从二级优化模型。提出了电力市场交易协调优化的框架，并针对长期、中期、短期和超短期电力市场建立优化模型进行分市场协调优化。对发电企业非递增报价进行了经济学分析，得出发电企业非递增报价是有其发电成本理论基础的。据此本文设计了发电企业非递增报价的形式，建立了差别定价方式下的电力交易优化模型研究发电企业非递增报价下的电力交易优化问题。由于交易模型中存在离散和连续两类不同性质的变量，其目标函数为非连续、不可微的函数，本文提出了一种启发式优化算法，算法计算简单方便，收敛速度快。论述了市场环境下水电参与竞价需要考虑的问题。针对水电参与竞价下电网企业购电的交易优化问题，建立了电网企业购电优化的离散混合规划模型。优化原则是在水火电满足负荷功率的条件下追求购买火电的费用最小，水电机组不参与竞价且水电的购买费用在模型体现为零，这样就保证了水电优先上网发电。为了解决多时段、多类变量的非线性离散混合规划问题，对负荷和机组根据负荷类型进行分类以简化模型，并提出了分解协调的求解方法，从而简化了模型的复杂性，避免在求解过程中的出现“维数灾”的问题，提高了算法的有效性。针对电网企业如何购买备用容量才能使自身利益最大化问题，利用会计成本法对发电机组的备用报价进行分析，建立发电侧备用容量的优化模型;同时根据用户上报的可中断负荷和可中断电价进行排序，建立以用户可中断负荷为备用容量的离散优化模型，最终构建了双侧开放的备用市场联合优化模型。由于电网企业对可中断负荷用户承担备用的处理方式只能是停止供电。因此电网企业最优购用备用容量优化模型中含有离散变量，为此根据最速下降原理提出了一种启发式算法，使模型更加实用，符合实际情况。研究表明与单独利用发电侧备用或用户侧备用相比，联合优化备用优于单独发电侧备用和单独的用户可中断负荷备用。结合世界各国放开用户选择权的经验，研究了我国目前发电企业向大用户直接供电急需解决的问题，并提出了目前发电企业向大用户直接供电交易的适当模式，即经公共电网进行直供电交易。针对大用户直供电价格设计问题，建立了用户用电量与电价之间的分段线性函数关系，同时将发电企业的发电成本分成容量成本与备用成本，分别在用户功率段上分摊，保证了发电成本的成功回收。建立了双目标优化的电价设计模型兼顾双方的利益，实现发电企业与用户效益的最大化。更多还原

【Abstract】 Electricity industry reform aims to break monopoly and build competitive electricity market. Marketlized transaction mechanism makes generators, power-supplying companies, power grid companies and customers independent participants to take part in transaction. In electricity market, many kinds of trade mode are used to optimize the electricity resource based on the price signal. Electricity transaction optimization is researched in this dissertation as follows:Firstly the formation of price chain and pricing mechanism in different electricity industry modes are demonstrated. The electricity pricing mechanism has been transformed from single government pricing mechanism into multiplied pricing mechanism including government regulation, market competition and contract pricing mechanisms. Aiming the unreasonable electricity price chain, some solutions are presented to perfect electricity pricing mechanism and effectively reduce the level of electricity price in China.In chapter three, main electricity transaction modes and settlement mechanism are introduced. The generation cost structure on thermal unit and the factors affecting generation cost are analyzed. Principal-subordinate two level optimization model is presented based on single-buyer market. To coordinate different electricity transaction, a coordinating optimization framework is given. Some models are formulated to optimize the transaction in long-term, medium-term, short-term and super short-term electricity market.From the economics analysis, the foundation of non-incremental bidding is discussed. Based on above, in chapter four the form of non-incremental bidding is designed. To study the transaction optimization mathematics model is built in PAB mode. Involving discrete and continuous variables, the goal function of the model is a discontinuous and differential function. A heuristic algorithm is presented to solve the model. This algorithm has advantages of easily calculation and fast constringency speed.The principal and problems of hydropower market transaction are discussed in chapter five. To solve the electricity transaction optimization combined with hydropower unit, a discrete-mixed model of electricity transaction is described. The optimizing principal is to minimize the total cost of purchasing the thermal-power in precondition of hydropower and thermal-power meeting the load demand, and the hydropower unit need not bid and the cost of hydropower in the model is zero, so the hydropower can be utilized with priority. To solve the non-linear discrete-mixed programming problem, load and unit are classified based on load types. A decomposing-coordinating algorithm is presented to simplify the model, avoiding the curse of dimensionality, increasing the calculation efficiency.Aiming to how the power grid company purchase the reserve capacity to maximize its benefit, chapter six analyzes the reserve bidding of generation unit with accountant cost theory, the reserve capacity optimization model in generation side is demonstrated. And according to the interruptible load and interruptible price submitted by customers, the reserve capacity discrete optimization model taking the interruptible load as reserve capacity is built. Based on the above models, reserve market joint optimization model considering generation reserve and interruptible load is presented. To use the interruptible load, power company can only stop the power supply, so the optimization model includes discrete variable. A heuristic algorithm based on the steepest descent method is used to solve the above optimization model. Compared with the reserve transaction mode only having generation reserve or interruptible load, the joint reserve transaction is more beneficial for the power grid company.Combined with the experience of electricity industry reform over the world, chapter seven analyzes the problems in larger users directly trading with power plant, and brings forward the appropriate transaction mode direct trade through public power grid. For the designing power price, it formulate the step-wise linear function between electricity consumption and power price, and share the capacity cost and reserve cost of unit on the power block of large-user to recover the generation cost. A double-goal optimization model of designing power price is discussed to consider the benefits of power plant and large-user together.更多还原