There appears to be increasing policy emphasis globally on supporting the development of electric vehicle (EV) technologies linked to industrial growth. The key reason for this is that the electrification of the vehicle drivetrain offers a viable solution for the sustainability requirements of the transport sector and to achieve emission reduction targets. However, different countries use different innovation policies to support the local development of EV technologies and domestic EV industry. To support national governments in making informed decisions, this paper seeks to develop a framework providing an ex-ante impact of various innovation decisions. This framework is based on “adaptive neuro-fuzzy inference system” (ANFIS). The necessary data for ANFIS framework is generated by analysing EV innovation policies of United States, Japan, European Union, Germany, France and United Kingdom and comparing them with the actual EV technology development that is measured by patent filings on those regions. Next, an ANFIS model has been constructed by specifying an equation and transforming the generated dataset into input-output data pairs. Finally, the data pairs are used for training and validating the ANFIS framework by using MATLAB software. The training results indicate that such framework can be applied for evaluating current EV innovation policies and predicting the future technology development with the introduced set of inputs owing to ANFIS`s backward and forward-pass mechanism.
|Publication status||Published - Dec 2014|
|Event||European Electric Vehicle Congress (EEVC2014) - Brussels, Belgium|
Duration: 3 Dec 2014 → 5 Dec 2014
http://www.egvi.eu/calendar/53/46/European-Electric-Vehicle-Congress-EEVC-2014 (Link to conference website)
|Conference||European Electric Vehicle Congress (EEVC2014)|
|Period||3/12/14 → 5/12/14|
- Electric Vehicle, Public Policy, Technology Development, International Policy Review, ANFIS-Based Modelling