Network Fault Diagnosis Tool Using AI and Historical Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current network fault diagnosis methods in utility supply networks, such as mobile communications networks, are inefficient due to the overwhelming number of alarms and disparate systems, making it difficult for operators to prioritize repairs and determine actual faults, leading to delayed service restoration and customer dissatisfaction.
Innovation Solution
A network fault diagnosis tool that compares current performance data with historical data to identify faults, using both subjective user feedback and objective measurements, and employs an artificial intelligence algorithm to determine the likelihood of a fault, allowing for prioritization and efficient maintenance planning.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If network operators monitor all equipment alarms to identify faults, then fault detection capability is improved, but the complexity of managing thousands of alarms simultaneously becomes unmanageable
Solution Approach 1:
The patent segments the overwhelming set of all alarms into meaningful groups based on fault types, severity levels, and equipment categories. This segmentation allows operators to focus on specific alarm categories relevant to their current diagnostic needs rather than being overwhelmed by the complete alarm universe, thus improving fault detection while reducing management complexity.
Solution Approach 2:
The patent introduces an intermediary intelligence layer (AI/ML system) that sits between the raw alarm data and the operators. This intermediary automatically processes, filters, correlates, and prioritizes alarms, transforming thousands of individual alarm signals into a manageable set of diagnosed fault conditions, thereby resolving the contradiction between comprehensive monitoring and manageable complexity.
2Speed
If network operators respond to all alarms immediately, then service restoration speed is improved, but resource efficiency deteriorates due to inability to prioritize repairs
Solution Approach 1:
The patent performs preliminary actions by automatically analyzing alarm patterns and predicting potential faults before they fully manifest or before operators can manually assess them. The system pre-prioritizes alarms based on their likely impact on service, enabling operators to immediately address the most critical issues while maintaining resource efficiency through intelligent pre-screening.
Solution Approach 2:
The patent dynamically changes the priority parameter of alarms based on multiple factors including current network load, historical fault data, affected user count, and service criticality. This parameter transformation converts static alarm priorities into dynamic, context-aware priority levels, allowing operators to respond to the most impactful faults first while maintaining efficient resource utilization.
3Ease of operation
If network operators use crude measures like user count or revenue to prioritize repairs, then repair prioritization is simplified, but diagnostic precision deteriorates
Solution Approach 1:
The patent creates a universal prioritization framework that simultaneously incorporates multiple diagnostic dimensions including user count, revenue impact, service criticality, historical fault patterns, and real-time network conditions. This multi-functional assessment system maintains ease of operation through automated computation while achieving high diagnostic precision by considering diverse factors that crude single-metric approaches cannot capture.
Solution Approach 2:
The patent implements feedback loops where diagnostic outcomes and repair results are continuously fed back into the system to refine prioritization accuracy. The system learns from historical data and operator corrections, progressively improving its ability to accurately diagnose faults while maintaining simple automated prioritization, thus resolving the contradiction between operational simplicity and diagnostic precision.
4Adaptability or versatility
If network operators use disparate systems for managing faults and reporting issues, then system functionality is comprehensive, but customer service quality deteriorates due to inability to provide unified status information
Solution Approach 1:
The patent merges data from multiple disparate systems including fault management, performance monitoring, and customer service platforms into a unified diagnostic view. By integrating these previously separate information sources, the system eliminates information silos and provides consistent, accurate status information across all customer touchpoints while maintaining the comprehensive functionality of each individual system.
Data Source
AI summary
A method of diagnosing faults in a utility supply network involves receiving performance data indicative of performance of the utility supply network, and receiving historical performance data indicative of a historical performance of the utility supply network and a fault associated with the historical performance data. A fault in the utility supply network is determined based on a comparison of the performance data with the historical performance data.