论文部分内容阅读
运用训练反向传播神经网络的方法对月度竞价市场历史数据进行学习,由网络模拟结果得出成交电价的概率分布函数,它与预测电价的差值服从正态分布。在此基础上建立机会约束规划模型,求解此模型得到以一定置信水平满足目标函数的投标报价。将其用于月度竞价市场,运用实际数据分析结果表明本方法是可靠的,可供发电公司投标报价参考。
Using the method of training backpropagation neural network to study the historical data of monthly bidding market, the probability distribution function of transactional electricity price is obtained from the result of network simulation. The difference between it and the forecast electricity price obeys the normal distribution. On the basis of this, a chance constrained programming model is established, and the model is solved to get the bidding price which satisfies the objective function with a certain level of confidence. Using it in the monthly bidding market, the actual data analysis results show that the method is reliable and can be used as reference for bidding of power generation companies.