Development of a Python tool based on model predictive control for an optimal management of the Calais canal

Abstract

Model predictive control (MPC) has been widely employed to control a large variety of water systems, such as dams, irrigation canals, inland waterways, drinking water networks and wastewater treatment plants. Its predictive capabilities and the possibility to incorporate constraints make MPC well suited to address several, and sometimes opposite, management objectives linked to water systems. The design of MPC for water systems is usually performed via dedicated software (e.g., Matlab) and tested in simulation using dedicated hydraulic software. However, the implementation of MPC strategies in real systems requires additional development to allow for its embedding within the information systems that are used by system managers. A possible solution is to create a tool based on Python that can be easily integrated with the information systems of managers, and within which existing Matlab solutions can be incorporated. In this paper, the development a ready-to-use Python tool using a hierarchical MPC approach designed for the management of the Calais Canal is presented.