Fleet Charging Scheduling Using Telematics Arrival Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing charging optimization techniques for electric vehicles are inadequate in dynamic environments, failing to account for vehicles that are likely to arrive or depart from charging stations, leading to suboptimal performance in terms of cost, readiness, and reliability.
Innovation Solution
A charging management system that predicts vehicle arrival times and states based on telematics data, including route information and vehicle attributes, to optimize future charging sessions and integrate them with current sessions using a charging management algorithm, improving scheduling and resource allocation at charging stations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If static optimization techniques are used for charging station management, then the optimization problem can be solved for known vehicles and chargers, but the solution becomes rapidly invalidated in dynamic environments where vehicles continuously arrive and depart
Solution Approach 1:
The patent transforms the static optimization approach into a dynamic one by continuously updating the charging optimization problem as vehicles arrive and depart. The system re-solves the optimization problem in real-time based on current vehicle队列, charger availability, and predicted future arrivals, ensuring the solution remains valid in dynamic environments.
Solution Approach 2:
The patent applies preliminary action by predicting future vehicle arrivals and pre-calculating charging schedules before vehicles actually arrive. This allows the system to prepare optimized charging assignments in advance, reducing wait times and improving overall system efficiency when vehicles do arrive.
2Productivity
If charging optimization considers only current vehicles at charging stations, then the optimization is simpler to compute, but it fails to account for vehicles likely to arrive or depart, leading to suboptimal performance
Solution Approach 1:
The system performs preliminary calculations by predicting future vehicle arrivals and their charging requirements before they occur. This advance planning allows the optimization to consider future demand without significantly increasing computational complexity during real-time operations, as the heavy lifting is done in advance.
Solution Approach 2:
The patent implements feedback mechanisms where the system continuously monitors actual vehicle arrivals and departures, compares them with predictions, and uses this information to refine future predictions and optimization decisions. This closed-loop approach improves productivity while managing complexity through learned patterns.
3Measurement precision
If telematics data is integrated into charging optimization, then vehicle arrival predictions and charging needs are more accurate, but data processing requirements and system complexity increase
Solution Approach 1:
The patent introduces an intermediary layer that processes telematics data and translates it into meaningful predictions about vehicle arrivals and charging needs. This intermediary processing layer filters and structures raw telematics data before feeding it to the optimization algorithm, improving prediction accuracy while managing data processing complexity through standardized data formats and aggregation.
Data Source
AI summary
Certain aspects of the present disclosure provide techniques for a method of managing charging of vehicles, comprising: estimating a vehicle return state based on telematics data associated with a vehicle; determining a future charging session for the vehicle based on the vehicle return state and one or more of: a vehicle attribute; a job attribute; and a station attribute; generating a hybrid set of charging sessions by adding the future charging session to a set of one or more current charging sessions; and processing the hybrid set of charging sessions with a charging management algorithm to determine one or more charging session attributes for each charging session in the hybrid set of charging sessions.