Models for the prediction of the thermophysical properties of halogenated substances

This entry contains, first, the Matlab functions code for the calculation of the critical temperature, critical pressure, acentric factor, normal boiling temperature, and ideal gas heat capacity by using a classical group contribution approach (GCM functions.p) or a neural-network-based approach (ANN functions.p). The input for their use is anarray containing the number of groups for the definition of the fluid, except for the case of the ideal gas capacity, where the first element of the array will be the temperature in K. An example of use can be found in the related publications.Second, function for the estimation of the uncertainty of the predicted values are also included:uncertainty model.m: obtains the estimated uncertainty for a specified model and property. The input for its use is the array containing the number of groups for the fluid, exceptfor the case of the ideal gas capacity, where the first element of the array will be the temperature in Kelvin. An example of use of this function can be found in the related publication.Tc model.mat: is a structure that contains the array of experimental values of Tc (Tcexp), the fluid groups (groups), the models' parameters (param GCM, param ANN), the predictions of the GCM and the ANN (Tc GCM, Tc ANN), their uncertainties (u GCM, u ANN), and the PDF of both model parameters (pdf GCM, pdf ANN). Pc model.mat, w model.mat, and cp0 model.mat are analogous to Tc model.mat.Tc model.csv, Pc model.csv, w model.csv, and cp0 model.csv: contain the data of the .mat files in csv format.本数据集涵盖两类内容：其一为Matlab函数代码，分别依托经典基团贡献法（GCM）与神经网络法（ANN），用于计算临界温度、临界压力、偏心因子、正常沸点及理想气体热容，对应函数文件为`functions.p`。此类函数的输入为用于定义流体的基团数量数组，仅理想气体热容的计算场景例外：此时数组的首个元素为以开尔文（K）为单位的温度。相关学术出版物中提供了该类函数的使用示例。其二为用于预估预测值不确定度的函数：1. `uncertainty model.m`：可针对指定模型与物性获取对应的预估不确定度，其输入为定义流体所需的基团数量数组，仅理想气体热容的计算场景例外：此时数组的首个元素为以开尔文（K）为单位的温度。该函数的使用示例可参见相关学术出版物。2. `Tc model.mat`为结构体文件，内置临界温度实验值数组（Tcexp）、流体基团信息（groups）、模型参数（param GCM、param ANN）、经典基团贡献法与神经网络法的临界温度预测结果（Tc GCM、Tc ANN）、二者的预测不确定度（u GCM、u ANN），以及两种模型参数的概率密度函数（Probability Density Function, PDF，对应字段为pdf GCM、pdf ANN）。`Pc model.mat`、`w model.mat`与`cp0 model.mat`的文件结构与`Tc model.mat`完全一致。`Tc model.csv`、`Pc model.csv`、`w model.csv`与`cp0 model.csv`均以CSV格式存储了对应`.mat`文件中的全部数据。