optimal solution considering constraints of both the EV user and the utility grid. The solution can be related to the direction of power flow, electricity cost, allowable charging rate, scheduling of charging and discharging of EVs, and power management. The central control, in a few cases, is supported by the necessary algorithms that process the data. The processing of data includes error check, relevant parameter estimations, data storage, and analysis. Nonetheless, the centralized control system determines solutions or makes decisions considering information from the entire system [9, 10]. A schematic of the centralized controller is shown in Figure 1.5. Each of the entities shown connected by dotted lines depicts communication links.
The major drawback of the central controller in a smart charging system is an optimization problem. The optimization problem becomes very large and complex as it involves numerous parameters from different entities. The controller’s failure in the centralized control system will result in a complete halt in operation or incur huge losses to the connected components. Further, scalability is another challenge when the optimization problem exceeds the constraints, such as the maximum number of EVs or charging stations [72-74]. The drawbacks of the centralized controller are outfitted by adopting hierarchical control architecture. Several controllers are deployed to administer a particular function. In contrast, the central controller is given the responsibility to monitor and perform load demand response. The hierarchical architecture resulted in reduced computational requirements [75, 76]. However, the risk of a negative impact on the smart charging system due to centralized control is not largely reduced.
Figure 1.5 Schematic of centralized controller in smart charging architecture.
1.7.2 Decentralized
Decentralized control, contrary to centralized, has distributed control and optimization modules. Charging of EVs takes place spatially in a distributed manner. Hence, the planning of decentralized control in smart charging systems is considered to be safe and reliable. In decentralized control, decision making takes place locally, where the EV charging takes place. The requirement of extensive and reliable communication systems, large and complete optimization, and the risk of damage due to a controller’s incorrect decision is readily reduced [74, 77]. The only challenge is performing load management. The data exchange between the utility grid and EV users still demands communication systems. The schematic of decentralized control is shown in Figure 1.6. Each entity has a local level controller and aggregator that are connected to a central controller. The distribution of tasks assigned to each local controller reduces the burden of the central controller to a large extent as compared to the centralized controller scheme.
Figure 1.6 Schematic of decentralized controller in smart charging architecture.
The simplicity in the implementation of the decentralized controller is leading to an increase in demand. Further, the coherency, like EV operation (spatially distributed), reduces deployment complexity. Further, decentralized control architectures are seemingly practical and scalable, considering their computational complexity [78].
Table 1.3 briefly presents the difference between the control architectures: centralized and decentralized. Based on the merits and demerits, the required architecture can be selected for the design of smart charging systems.
1.7.3 Comments on Suitability
The drawbacks and benefits of centralized and decentralized control architecture infer a requirement for maturity in the smart charging system.
Table 1.3 List of Differences between control architectures in a smart charging system.
| Control architecture | Merits | Demerits |
|---|---|---|
| Centralized | Better voltage and frequency regulationBetter utilization of network capacityCan be used with provisions for ancillary servicesEase of control and operation inclined towards the PSO | System design and deployment are complexDemands huge capital investment in developing robust communication architectureDifficult to scale due to predefined constraints in the optimization problemHigh computational requirements to process and analyze a large amount of dataRobust error correction of data protection is required |
| Decentralized | Ease in the tracking of fault in the systemMore control to the EV users and higher acceptance rateLess capital investment in deploying communication architectureEasily scalableEase of renewable energy integration | The impact on the utility grid is tough to determineThe use of EVs as ancillary services is difficult to implement |
During the initial push for smart charging systems for EVs, decentralized architecture looks more acceptable and easier to implement. Further, the demography and topography of the area also significantly impact the selection of control architecture. For an area with higher demography, the decentralized control architecture will resemble centralized control.
For example, consider a densely populated building where the people reside on multiple floors. The parking spot where the charging or discharging of EVs will occur is obviously very compact. Hence, a single controller will be making decisions for multiple EVs. Assume another scenario where the buildings are spatially distributed and the number of persons living in each building is less than the previous scenario. In the second scenario, the parking spot where the charging or discharging of EVs occurs will also not be dense, so single controller can make decisions for multiple buildings. Hence, the demography of an area plays a critical role. On the other side, when topography is looked into, the decentralized controller is preferred, practically, for hilly or mountainous regions. The deployment of communication systems for hilly regions is challenging compared to the plane area.
1.8 Outlook towards Smart Charging
EVs are looked upon as a key to unleashing the potential of clean transportation and low-carbon emission electricity. The push for electrification of the transport sector has brought changes in the operations of current utility grids with a rise in the integration of renewable energy sources. Renewable energy sources are spatially distributed in nature; EVs’ mobility and the capability to smart charge and discharge are seen as impactful in integrating renewables to the grid. The outlook of smart charging infrastructure has a wide perspective, which is drawn based on the geography of the land where the infrastructure will be developed, the system analysis time frame, the focus of the impact study, and the society.
The geography of the land helps decide the type of control architecture to be deployed for smart charging. Apart from control architecture, the availability of renewables is also considered. Consider a remotely located region with hilly terrain. The control architecture for such a region is preferred to be distributed, due to capital investment in developing communication architecture. The availability of renewables introduces another opportunity to develop an isolated grid rather than connecting to a larger grid. Hence, the potential of renewables in generating electricity is analyzed and,