Electric Vehicles Charging Infrastructure: Challenges & Solutions

In recent years, it has been calculated and analysed the reason behind the slow acceptance rate of plug-in electric vehicles. The most prominent reasons came out to be our distress level and inquietude of EV drivers’ in addition to high cost.

The recent technological advancements helped in the reduction of distance and time requirements to reduce the recharging time of the batteries at charging stations to almost 30 minutes in a single charge and increased the mileage. The main focus is to catalyse EV adoption, which is considered the prominent challenge restraining the increasing EV ecosystem. Now, analysing the trends and scenarios in EV implementation, the question arises to deploy a structured network of public charging infrastructure to minimise the distance traveled in excess to complete the journey to charging stations in the downtown and civic areas and how the size of urban and civic data should be taken to make this planning highly effective and productive.

Numerous studies have been done on classical methodologies of optimisation frameworks for electric charging ports and stations deployment which have been developed to modern approaches due to major constraints and restrictions. The research work uses methodologies developed in the scenario of facility location-based optimisation based on simulations and sensor-based datasets of traffic flows and related surveys. There are several research works done that describe the large-scale datasets gathered by urban sensing appliances. Despite all these works done and explored for the optimal location of EV charging stations based on the trajectory of large-scale data, the most appropriate approach has not been yet examined and scrutinised. This leads to a motivation to choose data-driven optimisation on the locations of EV charging ports in which a discrete optimisation problem is formulated having focused on covering the entire sample region.

The method is conducted with the dataset of one million phone users taken into consideration over 6 months through which mobility patterns of original motion were recorded in the specific region. Later on, the second step focus on developing a discrete optimisation technique for sorting out the effective configuration for the network of charging stations.

The solution to the problem optimally leads to computationally infeasible, a near-optimal solution based on a genetic algorithmic approach is suggested to move the solution being computationally efficient and later on, re-optimised using the Particle Swarm Optimisation

Ravi Kishor Ranjan

Guest Author The Author is the Academic Associate at Great Lakes Institute of Management, Gurgaon.