Intelligent Control of Groundwater in Slopes with Deep Reinforcement Learning

Abstract

The occurrence of landslides has been increasing in recent years due to intense and prolonged rainfall events. Lowering the groundwater in natural and man-made slopes can help to mitigate the hazards. Subsurface drainage systems equipped with pumps have traditionally been regarded as a temporary remedy for lowering the groundwater in geosystems, whereas long-term usage of pumping-based techniques is uncommon due to the associated high operational costs in labor and energy. This study investigates the intelligent control of groundwater in slopes enabled by deep reinforcement learning (DRL), a subfield of machine learning for automated decision-making. The purpose is to develop an autonomous geosystem that can minimize the operating cost and enhance the system’s safety without introducing human errors and interventions. To prove the concept, a seepage analysis model was implemented using a partial differential equation solver, FEniCS, to simulate the geosystem (i.e., a slope equipped with a pump and subjected to rainfall events). A Deep Q-Network (i.e., a DRL learning agent) was trained to learn the optimal control policy for regulating the pump’s flow rate. The objective is to enable intermittent control of the pump’s flow rate (i.e., 0%, 25%, 50%, 75%, and 100% of the pumping capacity) to keep the groundwater close to the target level during rainfall events and consequently help to prevent slope failure. A comparison of the results with traditional proportional-integral-derivative-controlled and uncontrolled water tables showed that the geosystem integrated with DRL can dynamically adapt its response to diverse weather events by adjusting the pump’s flow rate and improve the adopted control policy by gaining more experience over time. In addition, it was observed that the DRL control helped to mitigate slope failure during rainfall events.