ABSTRACT

Peak particle velocity (PPV) is one of the most-used parameters for assessment of ground vibration resulting from blasting and associated damage to nearby areas. To prevent and control the damages might be caused by PPV, its amount should be predicted and controlled before conducting blasting operation. This paper analyzes results of this environmental issue of blasting using a combination of predictive and probabilistic models (prediction and simulation phases). To get the right patterns, a new intelligent model (regression tree-based), known as gene expression programming (GEP), is applied and developed. Considering various conditions of data, GEP model was able to propose a predictive model that consists of mathematical relations between input and output parameters. The design of this model was carried out with different conditions, and the most optimal model with the lowest error was selected based on several performance indices. Using the selected model of GEP model, the Monte Carlo simulation technique was implemented to examine the possible risks and control them. The results showed that there is a need to have a better controlling of conditions of explosive operations before blasting operations. Using the developed models, the blast-safety-area can be obtained/determined and all workers and equipment are in safe side during blasting operations.

摘要

峰值粒子速度（PPV）是评估爆破和对附近区域的相关破坏所引起的地面振动的最常用参数之一。为防止和控制PPV可能造成的损坏，在进行爆破作业之前应预测和控制其数量。本文使用预测模型和概率模型（预测和模拟阶段）的组合分析了爆破这一环境问题的结果。为了获得正确的模式，应用并开发了一种新的智能模型（基于回归树），称为基因表达编程（GEP）。考虑到各种数据条件，GEP模型能够提出一种预测模型，该模型由输入和输出参数之间的数学关系组成。该模型的设计是在不同条件下进行的，根据几个性能指标选择误差最小的最佳模型。使用所选的GEP模型模型，实施了蒙特卡洛模拟技术，以检查可能的风险并加以控制。结果表明，需要在爆破操作之前更好地控制爆炸操作的条件。使用开发的模型，可以获取/确定爆破安全区域，并且在爆破过程中所有工人和设备都处于安全位置。结果表明，需要在爆破操作之前更好地控制爆炸操作的条件。使用开发的模型，可以获取/确定爆破安全区域，并且在爆破过程中所有工人和设备都处于安全位置。结果表明，需要在爆破操作之前更好地控制爆炸操作的条件。使用开发的模型，可以获取/确定爆破安全区域，并且在爆破过程中所有工人和设备都处于安全位置。