MVB Periodic Message Scheduling via Multi-Objective Evolutionary Algorithm
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current algorithms for constructing periodic scheduling tables in multifunction Vehicle Bus (MVB) networks, such as RM, EDF, DM, and LSF algorithms, are limited by low convergence rates and fail to meet the real-time requirements of periodic data transmission in actual scheduling scenarios.
Innovation Solution
A method and system for train periodic message scheduling based on a multi-objective evolutionary algorithm, which involves binary encoding an MVB periodic message table, performing crossover and mutation operations using a genetic algorithm, and calculating fitness values through multi-objective functions to generate an optimal periodic scheduling table that meets real-time requirements.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If traditional algorithms (RM, EDF, DM, LSF) are used for constructing periodic scheduling tables, then the scheduling process is simple to implement, but the convergence rate is low and real-time requirements cannot be met
Solution Approach 1:
The patent transforms the scheduling problem into an optimization problem by changing parameters: it uses binary encoding to represent scheduling schemes, defines fitness functions based on real-time requirements, and applies evolutionary algorithms with adjustable parameters (mutation rate, crossover rate, population size) to achieve high convergence rate while meeting real-time constraints
Solution Approach 2:
The patent replaces traditional deterministic scheduling algorithms with a probabilistic evolutionary computation system. Instead of using fixed mechanical rules (RM, EDF), it employs genetic algorithms that simulate natural selection processes to iteratively optimize scheduling tables, achieving both high convergence and real-time performance
2Reliability
If multi-objective evolutionary algorithm is used to optimize scheduling, then real-time requirements are met, but the computational complexity increases
Solution Approach 1:
The patent segments the complex multi-objective optimization into manageable components: it divides the scheduling table construction into iterative generations, separates objective functions into fitness evaluation criteria, and breaks down the search space into encoded individual chromosomes representing different scheduling schemes
Solution Approach 2:
The patent introduces dynamic elements to balance complexity and performance: the population adapts through generations, parameters like mutation rate can be dynamically adjusted, and the algorithm terminates when real-time requirements are satisfied, making the computational complexity adaptive rather than static
Data Source
AI summary
The present invention provides a method and a system for train periodic message scheduling based on a multi-objective evolutionary algorithm, relating to the field of information and communications technology, mainly including: acquiring an MVB periodic message table; binary encoding the MVB periodic message table and initializing it randomly, to generate an iterative population; performing crossover and mutation operations on the individuals of the iterative population using a genetic algorithm, to update the iterative population; constructing an MVB periodic scheduling table that meets scheduling needs and minimizes the macro cycle according to a multi-objective algorithm and the updated iterative population; scheduling train periodic messages according to the MVB periodic scheduling table, thereby meeting the real-time requirement of periodic data transmission in actual scheduling scenarios.