Italian experience on electrical storage ageing for primary frequency regulation

The paper describes the results of different types of ageing tests performed by Terna (the Italian Transmission System Operator) applied to several electrochemical technologies, namely lithium-based and sodium-nickel chloride based technologies. In particular, the tested lithium-based technologies exploit a graphite-based anode and the following cathode electrochemistries: lithium iron phosphate, lithium nickel cobalt aluminium, lithium nickel cobalt manganese and lithium titanate. These tests have been performed in the Storage Labs located in the two most important Italian islands, i.e. Sardinia (Codrongianos) and Sicily (Ciminna). The aim of the storage labs is intended to give the electrical grid ancillary services, e.g. primary frequency regulation, secondary frequency regulation, voltage regulation, synthetic rotational inertia provision and many more. For these applications, the ageing of the batteries is difficultly foreseeable since ageing tests are not standardized. In fact, the network services in which the storage technologies can make a significant contribution are very different from each other and imply different effects on the battery aging and performances. Hence, standard tests are required to verify the suitability of a particular storage technology for certain grid services by means of tailored tests, which may be different based on the specific grid service. Authors proposed some novel cycle types, which showed that, in several cases, frequency regulation cycle ages the batteries much more than the "standard" cycle. The "standard" cycle definition has been adopted in the paper to identify a battery cycle test which was carried out to uniformly compare and rank the different technologies Moreover, sodium-nickel chloride batteries are unaffected by the types of cycle and have negligible ageing. In addition, lithium manganese oxide and lithium titanate batteries show a very good behaviour with a slight degradation of the dischargeable energy irrespectively of the type of cycle. Moreover these two lithium –based technologies have been subjected to the greater number of tests (5000-6000 cycles). Differently, lithium nickel cobalt aluminium technology shows a considerable ageing and a strong dependence on the cycle types. Even if the theoretical explanations of such aging behaviours need time to be understood and expounded, the authors are convinced that the scientific community should become aware of these experimental results.