Modeling of a vanadium redox flow battery electricity storage system

Abstract

Today, the electricity industries are facing new challenges as the market is being liberalized and deregulated in many countries. Electricity storage is undoubtedly a disruptive technology that will play, in the near future, a major role in the fast developing distributed generations network. Indeed, electricity storage has many potential applications: management of the supply and demand of electricity, power quality, integration of renewable sources, improvement of the level of use of the transport and distribution network, etc. Over the years, many storage technologies have been investigated and developed, some have reached the demonstrator level and only a few have become commercially available. The pumped hydro facilities have been successfully storing electricity for more than a century; but today, appropriate locations are seldom found. Electrochemical storage is also an effective means to accumulate electrical energy; among the emerging technologies, the flow batteries are excellent candidates for large stationary storage applications where the vanadium redox flow battery (VRB) distinguishes itself thanks to its competitive cost and simplicity. In this ambitious work that encompasses the domains of electricity, electrochemistry and fluid mechanics, we have proposed a novel multiphysics model of the VRB. This model describes the principles and relations that govern the behaviour of the VRB under any set of operating conditions. Furthermore, this multiphysics model is a powerful means to identify and quantify the sources of losses within the VRB storage system; indeed, one of the purposes of this study is to propose strategies of control and operation for a greater effectiveness of the overall storage system. The electrochemical model is based on the electrochemical principles and the study of the VRB chemistry; this model determines the equilibrium voltage from the vanadium concentrations, and the associated activation, concentration, ohmic and ionic overpotentials. Furthermore, the vanadium concentrations within the tank and the stack are constantly determined as a function of the current and the electrolyte flowrate. A simplified model of the internal loss is also proposed. The electrochemical performance was then established through the simulation of a stand alone system composed of a solar source, a VRB and a load. The model determines the stack voltage, the power flows and the vanadium concentrations over a 24 h period. Furthermore, the model was successfully compared with experimental data through a series of charge and discharge cycles at constant currents. Thereafter, the properties of the electrolyte are briefly investigated: in particular their dependence upon the electrolyte composition. Indeed, the viscosity and the density are important parameters of the mechanical model. In order to analyse the battery performance, a mechanical model has been proposed to determine the mechanical power required to flow the electrolytes. This model based on fluid mechanics has an analytical part that predicts the pressure drop within the pipes and the tanks, and a numerical part. Indeed, the stack geometry is so complex that it can not be described analytically; therefore, a numerical model based on finite element method (FEM) is proposed. Hence, the mechanical power necessary to the battery operation is obtained at any operating conditions. The electrochemical and the mechanical models are finally assembled to form the original multiphysics model of the VRB. This model provides a good insight of the battery operation and offers a powerful means to enhance the battery performance. Indeed, there is at constant current an optimal flowrate that maximizes the efficiency. A second series of charge and discharge cycles has determined the efficiency of different control strategies. Finally, the battery operations at constant power were also discussed in details and an optimal operating point has been highlighted.