Energy storage systems—Characteristics and comparisons

Abstract

Electricity generated from renewable sources, which has shown remarkable growth worldwide, can rarely provide immediate response to demand as these sources do not deliver a regular supply easily adjustable to consumption needs. Thus, the growth of this decentralized production means greater network load stability problems and requires energy storage, generally using lead batteries, as a potential solution. However, lead batteries cannot withstand high cycling rates, nor can they store large amounts of energy in a small volume. That is why other types of storage technologies are being developed and implemented. This has led to the emergence of storage as a crucial element in the management of energy from renewable sources, allowing energy to be released into the grid during peak hours when it is more valuable.

The work described in this paper highlights the need to store energy in order to strengthen power networks and maintain load levels. There are various types of storage methods, some of which are already in use, while others are still in development. We have taken a look at the main characteristics of the different electricity storage techniques and their field of application (permanent or portable, long- or short-term storage, maximum power required, etc.). These characteristics will serve to make comparisons in order to determine the most appropriate technique for each type of application.

Introduction

Electrical energy is an invisible, omnipresent commodity that is readily available at the lowest possible cost in most cases. It has long been considered a common consumer good [1]. Today, it makes up 12% of the total energy processed by humanity, a proportion that is expected to grow over the next few years (34% predicted for 2025) in a context of diminishing fossil fuels, growing use of renewable energy, and greater respect for the environment [2].

At present, the production of electricity is highly centralized and, often, a long distance away from its end users. Load levelling is initially based on the prediction of daily and seasonal needs, but also, when production is not sufficient, on the contribution of secondary modes like hydraulic and thermal plants. In fact, these plants also use stored energy: water for the pumped storage plants, and fossil fuels for the thermal plants.

Delocalized electricity production and the introduction of variable, fluctuating sources (renewable energy: solar, wind turbines, etc.) increase the difficulty of stabilizing the power network, mainly due to a supply–demand imbalance. It is therefore convenient to generate the energy, transmit it, convert it, and then store it if need be. More than ever then, the storage of electrical energy has become a necessity. But electricity is difficult to store as this requires bulky, costly equipment.

It may be useful to keep in mind that centralized production of electricity has led to the development of a complex system of energy production–transmission, making little use of storage (today, the storage capacity worldwide is the equivalent of about 90 GW [3] of a total production of 3400 GW, or roughly 2.6%). In the pre-1980 energy context, conversion methods for the “storage of alternate current” were extremely costly, unreliable, or simply were not being used. This, along with the fact that electricity is mass produced, transmitted, and used in AC, has led to the belief that electricity cannot be stored. However, high-performance, inexpensive power electronics able to handle very high power levels have changed all that. It can now be asserted that electricity can be stored, even if it is indirect storage. But this requires that investment and operating costs be kept to an acceptable level, and that the environmental issues be considered.