Air overpressure (AOp) produced by blasting is one of the environmental hazards of mining operations. Accordingly, the accurate prediction of AOp is very important, and this issue requires the application of appropriate prediction models. With this in view, this paper aims to propose a new data-driven model in the prediction of AOp using a hybrid model of fuzzy system (FS) and firefly algorithm (FA). This combination is abbreviated as FS-FA model. The used data-sets in the proposed FS-FA model were arranged in a format of three input parameters. In total, 86 sets of the mentioned parameters were prepared. To avoid over-fitting, the data-sets were divided into two parts of training (80%) and test sets (20%). Three quantitative standard statistical performance evaluation measures, variance account for (VAF), coefficient correlation ( R 2 ) and root mean squared error (RMSE), were used to check the accuracy of the FS-FA model. According to the results, the R 2 and RMSE values obtained from the proposed FS-FA model were equal to 0.977 and 1.241 (for testing phase), respectively, which clearly demonstrate the merits of the proposed FS-FA model. In other words, the obtained R 2 and RMSE show that FS-FA model has high prediction level in the modeling of blast-induced AOp.

中文翻译：

---

使用混合模糊系统和萤火虫算法预测爆炸引起的 AOp 的新方法

爆破产生的空气超压 (AOp) 是采矿作业的环境危害之一。因此，AOp的准确预测非常重要，这个问题需要应用适当的预测模型。有鉴于此，本文旨在使用模糊系统 (FS) 和萤火虫算法 (FA) 的混合模型，提出一种新的数据驱动模型来预测 AOp。这种组合简称为FS-FA模型。所提出的 FS-FA 模型中使用的数据集以三个输入参数的格式排列。总共准备了86组上述参数。为避免过拟合，数据集分为训练集（80%）和测试集（20%）两部分。三个定量标准统计绩效评估措施，方差解释（VAF），系数相关性 (R 2 ) 和均方根误差 (RMSE) 用于检查 FS-FA 模型的准确性。根据结果​​，从所提出的 FS-FA 模型中获得的 R 2 和 RMSE 值分别等于 0.977 和 1.241（对于测试阶段），这清楚地证明了所提出的 FS-FA 模型的优点。也就是说，得到的R 2 和RMSE 表明FS-FA 模型在模拟爆炸诱发的AOp 方面具有较高的预测水平。