Rectifier Site Metadata Models for Resistance Trend Prediction

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Current methods for predicting soil resistance trends in pipeline cathodic protection systems are limited by manual data analysis and rely heavily on resistance readings, lacking depth of insight and requiring frequent maintenance, especially for newly installed or prospective sites.

Innovation Solution

Implementing machine learning models trained on rectifier site metadata and historical resistance trends to predict resistance trends independently of voltage and current measurements, using features like soil characteristics, pipeline data, and anode configurations.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If manual data analysis methods are used for predicting resistance trends, then operational simplicity is maintained, but prediction accuracy and depth of insight are insufficient

Engineering Contradiction:

Improveprediction accuracyVSAvoidsystem complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The patent replaces manual mechanical data analysis with automated machine learning models. The system uses algorithms trained on historical resistance trends and site metadata to automatically predict future resistance trends, substituting human analytical efforts with computational intelligence that provides deeper insights and higher accuracy.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The patent introduces machine learning models as an intermediary between raw data collection and prediction output. These models process historical resistance trends, site metadata, and environmental factors to generate accurate predictions, acting as a mediator that transforms complex data into actionable insights without requiring direct manual analysis.

Inventive Principle:

Principle #24Intermediary (Mediator)

2Reliability

If resistance readings are required for trend prediction, then prediction reliability is improved, but the system cannot predict trends for prospective sites without installed rectifiers

Engineering Contradiction:

Improveprediction reliabilityVSAvoidapplicability to prospective sites

Core Design Contradiction:

ReliabilityVSAdaptability or versatility

Solution Approach 1:

The patent performs preliminary training of machine learning models using historical data from installed rectifiers before deployment. The models are pre-trained on resistance trends and site metadata, enabling them to make predictions for prospective sites based on metadata alone, without requiring actual resistance readings from uninstalled rectifiers.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent uses site metadata as a digital copy or representation of the physical site characteristics. By training models on metadata from installed sites, the system creates a knowledge base that can be applied to prospective sites through metadata comparison, effectively copying the predictive patterns from established sites to new locations without physical presence.

Inventive Principle:

Principle #26Copying

3Measurement precision

If frequent data collection is implemented, then prediction accuracy is improved, but operational costs and time commitment increase

Engineering Contradiction:

Improveprediction accuracyVSAvoidtime commitment

Core Design Contradiction:

Measurement precisionVSLoss of time

Solution Approach 1:

The patent implements continuous automated data collection and model training that operates in the background without requiring periodic manual intervention. The system continuously ingests new resistance readings and site metadata, automatically retrains models, and updates predictions, maintaining high accuracy while eliminating the time commitment associated with frequent manual data collection cycles.

Inventive Principle:

Principle #20Continuity of useful action

Data Source

PatentUS12579445B2Models for predicting resistance trends

Publication Date: 2026.03.17 MOBILTEX DATA LTD

AI summary

A computer-implemented method of training a classification model includes the steps of obtaining, by at least one computer, a plurality of historical resistance trends from a plurality of installed rectifier sites and rectifier site metadata for each installed rectifier site of the plurality of installed rectifier sites; labelling each historical resistance trend of the plurality of historical resistance trends as one of a plurality of historic resistance trend classifications; and, inputting into a machine learning algorithm the historical resistance trends and the rectifier site metadata of the plurality of installed rectifier sites to train the classification model to output a predicted resistance trend classification in response to rectifier site metadata input into the model.