This paper proposes several data-based models for solar fields (SF) of concentrating solar power (CSP) plants. The proposed models estimate SF thermal output from solar resource data and aim to replace first-principles-based SF models typical in CSP generation schedulers. The advantages of the first models as opposed to the second ones are the ease in their development and their adaptability to changes in the plant, facilitating the forecast of SF thermal output. This information is very useful in order to take advantage of the operational flexibility of CSP plants, which is the main attribute that differentiates them from other renewable sources. The applicability of these models is studied in the optimal generation self-scheduling problem of a CSP plant participating in a day-ahead energy market. A simulation-based economic study has been performed using realistic data, where several scheduling strategies are evaluated in a 50 MW parabolic-trough plant with thermal storage. The scheduling strategies differ only in the model used for the SF. The study shows that the data-based models provide results very similar to those obtained with a first-principles-based SF model. Therefore, they can be used as a viable alternative, while significantly simplifying the design and development of the generation scheduler.

本文为聚光太阳能 (CSP) 发电厂的太阳能场 (SF) 提出了几种基于数据的模型。所提出的模型根据太阳能资源数据估计 SF 热输出，旨在取代 CSP 发电调度程序中典型的基于第一性原理的 SF 模型。第一个模型相对于第二个模型的优点是它们的开发容易和它们对工厂变化的适应性，促进了 SF 热输出的预测。这些信息对于利用 CSP 工厂的操作灵活性非常有用，这是将它们与其他可再生能源区分开来的主要属性。这些模型的适用性在参与日前能源市场的 CSP 电厂的最优发电自调度问题中进行了研究。已使用实际数据进行了基于模拟的经济研究，其中评估了一个 50 MW 抛物槽式蓄热电厂的几种调度策略。调度策略仅在用于 SF 的模型中有所不同。研究表明，基于数据的模型提供的结果与使用基于第一原理的 SF 模型获得的结果非常相似。因此，它们可以用作可行的替代方案，同时显着简化发电调度程序的设计和开发。