Mobile EV Charging Unit Deployment Using Demand Clustering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The infrastructure for electric vehicle charging stations is inadequate compared to traditional gas stations, leading to longer queue times and more frequent recharging needs for electric vehicles, necessitating a more efficient deployment of mobile charging units (EVCUs) to meet varying power demands.
Innovation Solution
A system utilizing a machine learning model trained on attributes of regions with low electric vehicle charge to predict the probability of EVCU demand, employing clustering to determine optimal deployment locations based on vehicle density, congestion, and other factors, enabling efficient power distribution.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If electric vehicle charging stations are established to accommodate the increase of electric vehicles, then the availability of charging infrastructure is improved, but the queue time for recharging increases and the frequency of recharging needs increases
Solution Approach 1:
The patent implements dynamic deployment of mobile charging units (MCUs) that can move between different locations based on real-time demand. The system continuously monitors charging requests and repositions MCUs to high-demand areas, transforming the static charging infrastructure into a dynamic system that adapts to changing spatial and temporal demand patterns, thereby reducing queue times while maintaining infrastructure availability.
Solution Approach 2:
The system predicts future charging demand using historical data and machine learning models, then proactively positions mobile charging units in anticipated high-demand areas before the actual demand occurs. This preliminary positioning action reduces waiting times when vehicles arrive at charging locations, as the charging infrastructure is already in place rather than being deployed reactively.
2Loss of time
If more electric vehicle charging stations are deployed to reduce queue time, then the waiting time decreases, but the infrastructure cost and complexity increases
Solution Approach 1:
The mobile charging units are designed as multi-functional assets that can serve multiple locations and purposes. Each MCU can charge multiple vehicles simultaneously, operate at different sites throughout the day, and adapt to various charging需求的. This universality allows a smaller number of mobile units to replace what would otherwise require many fixed charging stations, reducing overall infrastructure complexity while maintaining service coverage.
Solution Approach 2:
The system uses dynamic allocation and routing of mobile charging units based on real-time demand signals, vehicle locations, and unit availability. This dynamic management approach allows the same physical infrastructure to serve varying spatial and temporal demands, eliminating the need for static over-provisioning of charging stations and reducing the overall complexity of the charging network.
3Reliability
If fixed charging stations are used, then the infrastructure is stable and reliable, but the ability to adapt to varying power demands across different regions is limited
Solution Approach 1:
The mobile charging units provide inherent adaptability through their ability to move between locations and adjust their deployment based on varying power demands. The system dynamically repositions MCUs to regions with higher charging needs, whether due to geographic factors, temporal patterns, or unexpected demand surges, thereby achieving adaptability that fixed infrastructure cannot provide while maintaining operational reliability.
Solution Approach 2:
The system incorporates autonomous decision-making capabilities where the mobile charging units and management system automatically respond to demand signals without requiring centralized manual intervention. The system self-adjusts its deployment pattern based on real-time data, enabling it to adapt to varying power demands across different regions while maintaining stable and reliable charging service delivery.
Data Source
AI summary
An apparatus, method and computer program product are provided for determining a location for deploying an electric vehicle charging unit (EVCU). In one example, an apparatus divides a zone into a plurality of subregions and causes a machine learning model to output a probability of which an EVCU is needed at each of the plurality of subregion based on one or more attributes associated with the subregion. The apparatus generates one or more clusters within the zone, where each of the one or more clusters include one or more of the plurality of subregions. The apparatus calculates a value for each of the one or more clusters based on the probability associated with each subregion within said cluster. The apparatus selects one of the one or more clusters based on the value and assigns a location within the one of the one or more clusters for deploying the EVCU.