This paper presents the development of a model to predict the behavior of turbogeneration systems using curve affinity methods and data analysis. For this purpose, a hydroelectric turbogenerator with a capacity of 35 kW, manufactured by the Spanish company Perga Ingenieros was used. This turbogenerator was installed in the water distribution network of Bucaramanga, Colombia. Given that this type of system is not widely documented in scientific literature, it is a challenge to characterize it in detail, which provides a novel element to this research. This study compares the reliability of the proposed model with existing models in specialized literature to predict the characteristic and efficiency curves of this type of device. The results indicate that the proposed model has a satisfactory predictive capacity and fits well with the experimental data according to the goodness-of-fit parameters: Mean Absolute Percentage Error (MAPE), Sum of Squares Error (SCE), Coefficient of Determination R and Root Mean Square Error (RMSE). The model generally presented an improvement in goodness of fit with respect to the other models studied: 65 %, 76 %, 28 %, and 55 % for MAPE, SCE, R, and RMSE, respectively. To achieve a better fit in the model, the experimental dataset should be expanded, and the number of turbogenerators should be increased. Finally, hydroelectric turbogeneration systems are presented as an option for energy recovery in water distribution networks.

中文翻译：

---

用于评估从配水管网进行能量回收的潜水式水轮发电机性能的数学模型


本文介绍了使用曲线亲和方法和数据分析来预测涡轮发电系统行为的模型的开发。为此，使用了西班牙 Perga Ingenieros 公司制造的容量为 35 kW 的水力涡轮发电机。该涡轮发电机安装在哥伦比亚布卡拉曼加的供水网络中。鉴于这种类型的系统在科学文献中并未广泛记录，因此详细描述它是一个挑战，这为这项研究提供了一个新颖的元素。本研究将所提出模型的可靠性与专业文献中的现有模型进行了比较，以预测此类器件的特性和效率曲线。结果表明，该模型具有令人满意的预测能力，并且根据拟合优度参数：平均绝对百分比误差（MAPE）、误差平方和（SCE）、决定系数 R 和均方根误差 (RMSE)。与所研究的其他模型相比，该模型的拟合优度总体上有所提高：MAPE、SCE、R 和 RMSE 分别为 65%、76%、28% 和 55%。为了更好地拟合模型，应扩大实验数据集，并增加涡轮发电机的数量。最后，水电涡轮发电系统作为配水管网能量回收的一种选择。