Regional Digital Weighing Network for Bridge Load Monitoring
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing vehicle load detection techniques face challenges such as high construction costs, complex installation and maintenance, limited regional coverage, and inaccuracies in weight measurement, particularly for trucks and bridges with structural damage.
Innovation Solution
A regional traffic heavy load digital weighing method that utilizes existing physical weighing systems and intelligent technology to create a digital scale network model. This method involves selecting reference physical weighing systems, constructing single digital bridge scale models, and forming branch scale network models through migration learning and correction, enabling accurate vehicle load monitoring across a wide region.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a dynamic weighing (WIM) system is used for vehicle load detection, then measurement accuracy can be guaranteed under normal driving conditions, but construction cost increases, installation becomes troublesome, operation and maintenance complexity increases, and traffic must be interrupted
Solution Approach 1:
The patent creates a digital copy (virtual weighing system) that replicates the functionality of the physical WIM system. The digital twin model learns from and mimics the physical system's weighing capabilities, allowing virtual weight detection without requiring physical infrastructure. This resolves the contradiction by providing measurement functionality through digital replication rather than physical deployment.
Solution Approach 2:
The patent replaces the mechanical/physical WIM system with a digital/computational system. Instead of using physical sensors and weighing infrastructure, the solution employs computer vision, deep learning models, and digital twin technology to infer vehicle weights. This substitution eliminates the need for complex physical installation while maintaining measurement capabilities.
2Area of stationary object
If pure computer vision is used to judge vehicle weight by model recognition, then the whole region of monitored vehicles can be identified, but large bias exists in weight judgment of trucks only through appearance
Solution Approach 1:
The patent creates a digital twin that copies not just the visual appearance but also the physical properties and weighing behavior of vehicles. This digital replica enables accurate weight inference by learning from the physical system's actual measurements, combining the broad coverage of computer vision with the accuracy of physical weighing data.
Solution Approach 2:
The patent implements a feedback mechanism where the digital twin model continuously learns from and is corrected by actual physical weighing system data. This feedback loop allows the model to improve its weight estimation accuracy over time by comparing its predictions with ground truth measurements, thereby reducing the bias inherent in pure computer vision approaches.
3Adaptability or versatility
If WIM system weighing data tracking using computer vision is used, then vehicle identification can be achieved, but accurate measurement of each vehicle load cannot be guaranteed due to large uncertainty of vehicle travelling and limited identification range
Solution Approach 1:
The patent creates a comprehensive digital twin that captures and replicates the complete weighing behavior of vehicles across the regional network. This digital copy maintains accurate weight information for each vehicle by learning from multiple physical weighing points, enabling precise load measurement even as vehicles move through different locations in the network.
4Measurement precision
If Bridge Weigh-in-motion (BWIM) vehicle weight inversion based on structural response is used, then weighing results can be obtained from structural response, but the identification range is limited due to reliance on the weighing system and the results cannot be directly used for enforcement or safety analysis
Solution Approach 1:
The patent creates a universal digital twin system that serves multiple functions: it provides accurate weight measurements for enforcement purposes, enables broad regional coverage across multiple bridges, and generates data suitable for both operational control and long-term safety analysis. The digital twin model is designed to be universally applicable across different bridge types and locations, eliminating the limitations of bridge-specific BWIM systems.
Data Source
AI summary
The present application discloses a regional traffic heavy load digital weighing method and synergy system, the weighting method includes: based on data of a physical weighing system available in a region and basic data of a bridge in the region, selecting a reference physical weighing system at a position and a number M and determining a reasonable number N of bridge group transfer layers; based on the reference physical weighing system, constructing a corresponding single digital bridge scale model; obtaining a branch scale network model based on the reference physical weighing system through migration learning and correction; composing a regional digital scale network model based on branch scale network models corresponding to different reference physical weighing systems.