面向大工业用户的售电公司购售电策略研究

发布时间：2018-05-27 15:40

本文选题：购售电综合策略模型 + 随机价格模型　；参考：《华北电力大学(北京)》2017年硕士论文

【摘要】：我国新一轮电改指出要有序向社会资本开放配售电业务,售电试点地区的售电公司和用户积极开展购售电交易,电力改革发展态势良好。售电商作为电力市场的主体,可以选择不同的市场购电,再将电力出售给终端用户,它的利润是售电收入与供电成本之差。但由于电力市场中价格是不断变动的,用户的购电需求量也不确定,因此售电商在交易中要承担一定的风险。在目前的改革形势下,本文针对面向大工业用户的售电公司的购售电策略的研究有着重要的研究价值和实际意义。本文的研究对象为不拥有自身输配电资产的独立售电商,其客户目标群体为大工业用户,从购电和售电两方面研究其在市场中的最优购售电策略。在购电方面,考虑售电公司从中长期、日前和实时三个市场中购电,主要研究其购电成本,包含在各个市场中的购电量及相应价格。其中,日前和实时市场为现货市场,设为每小时结算一次电价和电量,而中长期市场中的合约依据其负荷特点的不同分为工作日峰时、非工作日峰时和谷时三类,具有不同的价格和购买量限制。售电方面,主要研究了销售电价的定价方法,考虑了固定电价和峰谷分时电价两种定价法,同时在定价过程中考虑了输配电成本的回收,并将其体现在销售电价上。在购售电交易过程中,售电商所承担的风险采用CVaR风险约束条件来控制。结合购电和售电这两方面,本文建立了售电公司的购售电综合策略模型。此外,该购售电模型涉及到各个市场的购电价格,其中,中长期合约是在时间跨度一开始就签订的,其电价可视为已知;而日前市场和实时市场的电价是时刻波动的,视为随机参数,本文建立了一种随机价格模型,结合拉丁超立方抽样法来获得随机价格样本,可以模拟现货市场价格的不确定性。最后,基于美国德州电力市场数据,采用遗传算法,在销售电价采用固定电价形式时,对售电公司代理不同负荷特点的大工业用户可获最大利润做出对比分析,得出不同负荷的供电成本差异;在销售电价采用峰谷分时电价形式时,求解模型得到售电公司的最优购售电策略。
[Abstract]:The new round of electricity reform in China points out that it is necessary to open the distribution business to the social capital in an orderly manner, and the power sale companies and customers in the pilot area of electricity sale have actively carried out the purchase and sale of electricity, and the electric power reform has developed in a good situation. As the main body of electricity market, electricity sellers can choose different markets to buy electricity, and then sell electricity to end users. Its profit is the difference between electricity sales income and power supply cost. However, because the price in the electricity market is constantly changing and the demand for electricity purchase is uncertain, the seller has to bear certain risks in the transaction. Under the current situation of reform, this paper has important research value and practical significance to study the strategy of purchasing and selling electricity to large industrial customers. The research object of this paper is the independent power seller who does not own its own transmission and distribution assets, and the target group of its customers is the large industrial users. The optimal strategy of purchasing and selling electricity in the market is studied from the two aspects of electricity purchase and power sale. In the aspect of purchasing electricity, we consider that the company buys electricity from three markets: medium and long term, day before and real time. It mainly studies the cost of purchasing electricity, including the purchase quantity in each market and the corresponding price. Among them, the pre-day and real-time market is the spot market, which is set to settle the electricity price and electricity quantity once per hour, while the contracts in the medium and long term markets are divided into three categories according to their load characteristics: working peak, non-working peak and valley time. There are different price and purchase limits. In the aspect of electricity sale, this paper mainly studies the pricing method of sale electricity price, considering two pricing methods: fixed electricity price and peak-valley time-sharing price. At the same time, the recovery of transmission and distribution cost is considered in the process of pricing, and it is reflected in the sale price. In the process of purchase and sale of electricity, the risk borne by the seller is controlled by the CVaR risk constraint. Combining the two aspects of electricity purchase and power sale, this paper establishes a comprehensive strategy model of power purchase and sale. In addition, the model relates to the purchase price of electricity in various markets, in which medium- and long-term contracts are signed at the beginning of the time span, and the price of electricity can be considered as known, while the price of electricity in the pre-day market and in the real-time market is always fluctuating. As a random parameter, this paper establishes a stochastic price model, which combines Latin hypercube sampling method to obtain the random price sample, which can simulate the uncertainty of the spot market price. Finally, based on the data of Texas Electric Power Market and genetic algorithm, when the electricity price is fixed, the paper makes a comparative analysis on the maximum profit of the large industrial customers who represent the different load characteristics of the power sale company. When the price of electricity is in the form of peak-valley time-sharing price, the model is solved to get the optimal purchasing and selling strategy of the power company.
【学位授予单位】：华北电力大学(北京)
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：F416.61

【参考文献】