The intertwining of technology and energy transition is reshaping the landscape of global energy systems, where policies are understood within the context of evolutionary change to identify fundamental functions of the system. To design those policies, we need to understand technical progress, which is a continuous process where new knowledge arises from prior knowledge. We start from the idea that the information contained in patent data can be used as a proxy for knowledge. Thus, we consider a network that is made up of patents (nodes), which are in turn interconnected when they are part of another patent. Our analysis is based on identifying the theoretical model behind the evolution of the observed patent network associated with the energy storage industry. An important network evolution mechanism is preferential attachment. This mechanism can be summarized simply by the statement that the more connections a node has, the more connections it gets over time. Differently put, preferential attachment describes a situation where the new connection is not entirely random, but it depends on the degree of existing nodes (the number of links that the nodes have with other nodes). This allows more connected nodes to gain more connections if compared to less-connected nodes. For instance, consider the following algorithm for network formation: 1) growth-we begin with n_0 nodes, and we add a new node each point in time; and 2) preferential attachment-the probability that a new node will be connected to another node depends on its degree, i.e., the number of links with other nodes. The project aims at building a database of patents files since the 1980s associated with energy storage systems. With it, using non-parametric data science methods, we will study the evolution mechanisms and geographical structure of knowledge over time, represented by the patent network. The statistical method that will be used is developed and described in detail in (Pham, Sheridan, and Shimodaira 2015), and implemented in the PAFit R Package.
