Seismic Vulnerability of Groundwater Resources Based On Failure of Fuel Pipeline Network Using DRASTIC Method and Artificial Neural Network (Case Study: Tehran Plain)

Groundwater is one of the main sources of sustainable development in human societies, and due to the supply of water needed by the drinking, agriculture, and industry sectors, their pollution will have destructive effects. In addition, the fuel transmission network is of great importance due to the storage and transportation of petroleum products. The importance of this system increases in various aspects during events such as earthquakes. Pollution of underground water sources due to leakage from the fuel transmission system is one of the secondary effects of the earthquake and leaves an adverse effect on the environment and human health. This research is focused on evaluating the vulnerability of groundwater due to the failure of the urban fuel transmission network against the occurrence of an earthquake, which was conducted in the form of case studies on the fuel transmission network of Tehran city and the Tehran-Karaj plain aquifer.

In this research, a unified model for seismic damage analysis and risk assessment under conditions of uncertainty in the fuel transmission network of Tehran city has been defined and implemented. In this model, the consequences of damage to the fuel transmission network under three earthquake scenarios (magnitude 5, 6, 7 on the Richter scale) have been evaluated on the pollution of underground water in Tehran. In addition, strategic strategies have been presented in order to reduce the effects of groundwater pollution caused by earthquakes. The proposed model uses a multi-layer perception artificial neural network (ANN). Also, the DRASTIC model has been used to evaluate groundwater pollution based on fuel leakage from the damaged fuel transmission network. In these studies, strategic planning has been done based on the use of robust decision-making techniques and the degree of robustness in order to reduce the probable effects of groundwater pollution by using the theory of minimum-maximum regret.

Results:

 The results of this research showed that the developed artificial neural network model has a high ability to assess damage(failure-leakage) in the pipeline of the fuel transmission network so that the root-mean-square error (RMSE) and the correlation coefficient R are equal to 0.029 and 0.98 respectively. Based on the results, the amount of damage to the pipeline in the first scenario (Mw=5) is equal to 15 leaks and 2 failures, in the second scenario (Mw=6), it is equal to 25 leaks and 7 failures and in the third scenario (Mw=7) the number of leaks was 27 and 9 failures. According to the results obtained from aquifer pollution under three earthquake scenarios, it is clear that in the scenario of an earthquake with Mw=5, 30% of the aquifer has medium pollution potential and 55% of the potential of low pollution, in the scenario of an earthquake with Mw=6, 45% The aquifer has medium pollution potential and 18% has low pollution potential, and in the earthquake scenario with Mw=7, 55% of the aquifer has medium pollution potential and 22% has low pollution potential. In line with the strategic planning of aquifer pollution reduction, different strategies were evaluated against different scenarios with the minimum-maximum regret criterion. Finally, three strategies were presented to reduce the pollution of underground water resources. According to the results, the use of the strategy (insulation of the environment around the pipeline) led to a 70% reduction in groundwater pollution, and the use of the strategy (implementation of an intelligent seismic system to cut off the fluid flow in the event of an earthquake), which was known as a robust strategy, led to aquifer pollution has been reduced by 75%.

Based on obtained results from the performance evaluation of the model developed in this research, it was found that the presented model had an acceptable performance in predicting the seismic vulnerability of the fuel transmission pipeline and assessing the pollution of the aquifer. This model has the ability to be implemented in different urban areas and to evaluate the performance of the fuel transmission system under different earthquake scenarios, as well as to evaluate groundwater pollution. Also, based on the results of the strategic management of groundwater pollution control, implementing an intelligent seismic system to cut off the fluid flow in the event of an earthquake can be used as a comprehensive solution to reduce environmental damage to groundwater resources specifically in the seismic regions.