EXTENDED ABSTRACT
This study aims to develop an integrated model for optimal smart electric vehicle parking management, addressing multiple challenges including behavioral uncertainties of owners, market price fluctuations, and network operational constraints. To this end, a two-stage stochastic programming model is formulated that aims to maximize the expected profit of the parking owner by comprehensively considering revenue sources such as charging services, vehicle-to-grid V2G revenue, and flexibility services, and costs including electricity purchase, battery degradation compensation, and operational costs. Uncertainties are modeled in the form of possible scenarios for parameters such as vehicle entry and exit times, initial charge level, and electricity price. The proposed model simultaneously considers the technical limitations of batteries, the constraints of the distribution network using linearized AC load shedding, and the physical constraints of parking. The model implementation was performed using the CPLEX solver in Python and its performance was evaluated by comparing four different scenarios including base case, non-smart charging, V2G activation and full participation on a 50-space parking lot connected to the IEEE 33-bus experimental network. The simulation results show that the proposed model in the full scenario, in addition to a 47% increase in owner profit, leads to a 26% improvement in network losses and an improvement in voltage profile. As a flexible and implementable framework, this model can serve as a decision-making basis for investors and grid operators in effectively integrating electric vehicle fleets with the power system.
Introduction
As environmental concerns grow, the transition to electric vehicles has emerged as a key solution. However, the rapid spread of these vehicles has created new challenges for power grids, including uncertainty in owner behavior and unpredictable charging patterns. Smart electric vehicle parking lots, by playing the role of energy aggregator, have high potential to participate in the energy market and provide ancillary services to the grid. However, optimal management of these parking lots requires advanced mathematical models that consider technical, economic, and operational aspects in an integrated manner.
The main issue of this research is to address the multiple challenges of managing these parking lots. These challenges include: 1) the complexity of planning due to uncertainty in parking times and vehicle charging needs, 2) the risk of managing energy purchases and sales due to fluctuations in electricity prices, 3) the lack of a comprehensive model that simultaneously covers pricing, capacity management, and service delivery issues while considering network and parking constraints, and 4) the low implementability of existing solutions due to neglect of operational constraints.
This research is necessary from several perspectives. From a scientific perspective, providing a unified model helps fill a research gap. From an economic perspective, profit optimization models strengthen the incentive for private investment. From an environmental perspective, efficient parking management leads to the integration of renewable resources and reduced energy waste. From a technical perspective, the development of robust algorithms for complex problems is a step forward in power systems engineering. Finally, from a social perspective, improved service delivery can help to achieve wider adoption of this technology and accelerate the transformation of transportation.
Materials and Methods
The research methodology of this paper is based on the development of an integrated mathematical optimization model for smart parking management of electric vehicles. The model is formulated in a two-stage stochastic programming framework to handle key uncertainties such as vehicle entry and exit times, initial charge levels, and electricity price fluctuations. The modeling begins with defining a set of basic assumptions, including full access to vehicle information, rational behavior of owners, and discrete time partitioning. Then, the technical behavior of the system is modeled in detail. This section includes dynamic equations and constraints of the vehicle batteries such as the permissible range of charge level and charge/discharge power, the model of vehicle presence in the parking lot, and the calculation of battery wear cost. In the next step, the technical constraints of the power distribution network are integrated into the model using a linearized AC load distribution model to maintain voltage stability and not exceed the line capacity. To deal with uncertainties, a scenario-based approach is used. Uncertain parameters such as vehicle travel times are modeled with probability distributions such as normal and beta and then transformed into a set of discrete scenarios. The ultimate objective function is to maximize the expected profit of the parking lot owner considering all sources of revenue and cost under these scenarios. Finally, the model is transformed into a linear mixed-integer programming problem and implemented and solved using the commercial solver CPLEX in the Python environment.
Results and Discussion
The simulation results indicate the success of the proposed model in achieving the economic and technical objectives. From an economic perspective, the full flexible management scenario using smart charging and V2G capability increased the average daily profit of the owner, which shows a growth of 47% compared to traditional management. The revenue structure in this scenario consisted of 55% charging revenue, 30% V2G revenue, and 15% flexibility services, emphasizing the critical role of V2G in profitability. Dynamic pricing also shifted demand to off-peak hours and attracted owner participation. From a technical perspective, the integration of grid constraints with the linear AC load shedding model led to significant improvements in grid performance. Active power losses were reduced by 26% and the voltage profile during peak hours was improved to a safe level. Also, the maximum load of the lines was reduced from 95% to 78%, which provides a capacity to accommodate 34% more electric vehicles without the need for immediate infrastructure expansion. The network reliability index also improved significantly. Comparison with previous studies shows that the present model has superior performance in terms of increasing profits and improving network stability due to simultaneous consideration of network constraints and uncertainties. Overall, the integrated approach of this research has become a win-win solution for parking lot owners and network operators.
Conclusion
The results of this study show that the integrated flexible power management model for smart parking lots simultaneously achieves both economic and technical objectives. From an economic perspective, the owner's daily profit increased with charging optimization and V2G activation. From a technical perspective, the integration of grid constraints reduced energy losses, improved voltage quality, and increased electric vehicle acceptance capacity. Comparison with previous research confirms the superiority of this comprehensive approach. As an operational framework, this model can be the basis for policy making and investment in the development of sustainable urban infrastructure.