Impedance Spectroscopy Concrete Analysis with Machine Learning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current impedance spectroscopy methods for concrete lack effective equivalent circuit models and correlations between resistance, capacitance, and internal constituents, limiting their ability to predict the water-cement ratio and durability of cement-based materials.
Innovation Solution
An impedance spectroscopy analytical method using machine learning that normalizes equivalent circuits based on a theoretical model, incorporating a conductive path reflecting concrete microstructure, and employs machine learning models like Gaussian Process Regression, Support Vector Regression, and Decision Trees to estimate the water-cement ratio from resistance and capacitance parameters.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If multiple different equivalent circuit theories (10 different circuits) are applied to analyze impedance spectroscopy of concrete, then various research perspectives can be explored, but the results become inconsistent and difficult to interpret generally
Solution Approach 1:
The patent develops a unified equivalent circuit model that can universally represent different concrete microstructures (hardened and unhardened states, different compositions) through a single standardized circuit configuration. This universal model enables consistent interpretation of impedance spectroscopy results across various concrete types and conditions, resolving the inconsistency caused by using multiple different circuit theories.
Solution Approach 2:
The patent introduces normalized parameters (normalized resistance and normalized capacitance) that transform the impedance data into a consistent framework. By changing the parameter representation from raw impedance values to normalized forms that account for concrete composition and microstructure, the model achieves generalizability while maintaining accuracy across different concrete types.
2Adaptability or versatility
If the number of resistances and capacitances varies depending on the applied circuit theory, then different circuit models can be used, but it becomes difficult to derive correlations with constituent materials such as mixing ratio
Solution Approach 1:
The unified equivalent circuit model uses a fixed number and configuration of resistance and capacitance elements that consistently represent different concrete states. This standardization enables the derivation of meaningful correlations between the circuit parameters and constituent materials (such as water-cement ratio, aggregate content) because the same parameters can be compared across different concrete compositions without the confusion of varying circuit structures.
3Ease of manufacture
If traditional quality control tests (slump test, air volume test) are used for unhardened concrete, then basic properties can be measured, but performance and durability prediction capabilities are insufficient
Solution Approach 1:
The patent replaces traditional mechanical quality control tests (slump test for workability, air volume test for porosity) with electrical impedance spectroscopy measurements. The impedance measurements, when processed through the unified equivalent circuit model, provide information about concrete composition and microstructure that correlates with performance and durability, thereby substituting mechanical testing with an electrical measurement system that offers better predictive capability.
4Productivity
If impedance spectroscopy is applied to concrete with different equivalent circuit theories, then research on electrical properties can be conducted, but the development of test methods for estimating internal components is very limited
Solution Approach 1:
The patent transforms impedance spectroscopy data into meaningful estimates of internal concrete components (water-cement ratio, porosity, composition) by using normalized parameters derived from the unified equivalent circuit model. The normalization process and machine learning algorithms establish accurate relationships between the electrical measurements and physical concrete properties, enabling precise estimation of internal components that was previously unachievable with traditional circuit theories.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
Enables immediate field responses to changing conditions, allowing for accurate mixture information derivation and durability prediction, contributing to reliable quality control techniques for unhardened concrete.
Implementation Method 1
Electrochemical Impedance Spectroscopy (EIS) is a non-destructive technique that uses a node that measures electricity to identify the electrical flow with moisture and conductive ions present inside a target material.
Data Source
AI summary
Provided is an impedance spectroscopy analytical method for concrete using machine learning. The method comprises identifying electrical flow through moisture and a conductive ion present in concrete using a node that measures electricity based on electrochemical impedance spectroscopy (EIS); generating a theoretical equivalent circuit model comprising a conductive path reflecting the electrical flow: normalizing an equivalent circuit reflecting a concrete microstructure based on an impedance experiment using the theoretical equivalent circuit model; and generating a predictive model for estimating a water and cement ratio from a parameter value of the equivalent circuit through machine learning. Accordingly, the accuracy and reliability of estimating the microstructure and mixing ratio of the cement-based material can be increased.