Abstract In Brazil, big consumers can choose the free market to purchase their energy. The main advantage is the possibility of negotiating a lower price and the main risk is the difference between the energy contracted and consumed in each month, which should be settled by the spot price. This paper proposes to optimize the parameters of a contract by a statistical and optimization model. In the statistical part, the parameters for energy and peak demand time series are estimated in a correlated way with the simulated scenarios of spot price, which is an input in the proposed methodology. In the optimization part, a stochastic model is applied using a convex combination of the Expected Value and Conditional Value-at-Risk to find the optimum contract parameters, which are the energy contracted, the upper and lower bounds of the contract and peak demand contracted. To summarize, the main contribution of the proposed approach is to provide a methodology for big electricity consumers in the free energy market in Brazil, which considers a statistical model, that correlates and simulates scenarios of energy and peak demand with the spot price scenarios, and optimizes it based on a stochastic optimization model combining the Expected Value and Conditional Value at Risk as risk metrics. The results indicate that the methodology can be a powerful tool for consumers in the free market and, due to the nature of the model, it can be generalized for different energy markets.

中文翻译：

---

考虑能源、电力需求和现货价格的相关情景，巴西大型电力消费者的自由合同环境

摘要 在巴西，大消费者可以选择自由市场购买能源。主要优势是可以协商更低的价格，主要风险是每个月合同和消耗的能源之间的差额，应以现货价格结算。本文提出通过统计和优化模型来优化合约的参数。在统计部分，能源和峰值需求时间序列的参数与现货价格的模拟场景相关联，这是所提出的方法的输入。在优化部分，使用期望值和条件风险价值的凸组合应用随机模型来找到最佳合同参数，即收缩的能量，合约上下限和高峰需求收缩。总而言之，所提出方法的主要贡献是为巴西自由能源市场的大电力消费者提供一种方法论，该方法考虑了一种统计模型，该模型将能源和峰值需求情景与现货价格情景相关联并进行模拟，以及基于随机优化模型对其进行优化，该模型结合了预期值和风险条件值作为风险指标。结果表明，该方法可以成为自由市场中消费者的强大工具，并且由于模型的性质，它可以推广到不同的能源市场。它考虑了一个统计模型，将能源和高峰需求场景与现货价格场景相关联并模拟，并基于结合预期值和条件风险值作为风险指标的随机优化模型对其进行优化。结果表明，该方法可以成为自由市场中消费者的强大工具，并且由于模型的性质，它可以推广到不同的能源市场。它考虑了一个统计模型，将能源和高峰需求场景与现货价格场景相关联并模拟，并基于结合预期值和条件风险值作为风险指标的随机优化模型对其进行优化。结果表明，该方法可以成为自由市场中消费者的强大工具，并且由于模型的性质，它可以推广到不同的能源市场。