Electric Bus Fleet Energy Management for SoC-Based Schedule Reliability
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for managing electric vehicle fleets, such as public transportation systems, face challenges due to increased interdependence and lack of flexibility, leading to inaccurate scheduling, excessive energy use, and range anxiety, resulting in inefficiencies and higher CO2 emissions.
Innovation Solution
A computer-implemented method that utilizes a database of historic data to predict battery state of charge (SoC) and energy consumption, adjusting schedules in real-time to prevent failures, using a graphical user interface to visualize vehicle status and allowing for proactive management of electric vehicle fleets.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual calculations and theoretical planning are used for fleet management, then operational simplicity is maintained, but scheduling accuracy and energy management efficiency deteriorate
Solution Approach 1:
The system performs preliminary calculations of expected state of charge and energy consumption before journeys begin, using historical data and converted attributes to predict battery status. This advance planning enables proactive identification of potential failures and schedule adjustments without requiring complex real-time interventions during operations.
Solution Approach 2:
A conversion layer transforms operational attributes (temperature, distance, driver performance) into standardized ranges that can be queried against historical data. This intermediary conversion process enables accurate predictions while maintaining operational simplicity, as the complexity is confined to the conversion mechanism rather than the entire management system.
2Reliability
If conservative schedules with excessive slack are generated, then vehicle reliability is improved, but energy consumption and fleet size requirements worsen
Solution Approach 1:
The system calculates expected state of charge before each journey using historical performance data and converted attributes, enabling reliable schedule generation without excessive slack. By predicting battery status in advance, the system can confidently optimize schedules to minimize energy consumption while maintaining reliability.
Solution Approach 2:
The system uses historical data feedback to continuously improve predictions of energy consumption and state of charge. This feedback mechanism enables the generation of optimized schedules that maintain vehicle reliability while reducing energy consumption, as the system learns from past performance to make more accurate predictions.
3Productivity
If detailed performance tracking and data conversion are implemented, then energy management optimization improves, but system complexity and computational requirements worsen
Solution Approach 1:
A conversion layer transforms diverse operational attributes (temperature, distance, driver performance) into standardized ranges that can be efficiently queried against historical data. This intermediary process enables detailed performance tracking without proportionally increasing system complexity, as the conversion mechanism standardizes data processing.
Solution Approach 2:
The system changes parameters from raw operational attributes to converted ranges with associated performance indices. This parameter transformation enables efficient data processing and querying while maintaining detailed performance tracking, as the converted parameters are optimized for computational efficiency.
Data Source
AI summary
Computer-implemented method for optimizing the energy management of a fleet of electric transportation vehicles, more particularly a fleet of electric buses, including for each vehicle of the fleet, before performing each of the said at least one journey of the schedule: acquiring a set of attributes for the upcoming journey; launching a query in a database in order to obtain at least one performance index in function of the attributes; if the query returns at least one performance index, calculating an expected state of charge of the battery at the end of the upcoming journey; comparing the expected state of charge with a predetermined state of charge limit; and if the expected state of charge is lower than said predetermined state of charge limit, deliver a signal on a human-machine interface to signal a potential failure of the upcoming journey.