Field Soil Classification Data Fusion Without Lab Equipment
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The lack of standardized and cost-effective methods for determining soil classification in the field, particularly for compressed earth block (CEB) construction, which relies on laboratory equipment and specialized training, limits the widespread adoption of CEBs due to variable soil properties and inconsistent international standards.
Innovation Solution
A method using field tests such as wash, feel, and jar tests, combined with a neural network trained on a validation dataset, to determine soil classification without laboratory equipment, achieving accuracy within 10% of laboratory results, and enabling the production of CEBs with standardized properties.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If laboratory testing is used to determine soil classification, then measurement precision is improved, but device complexity and cost increase
Solution Approach 1:
The patent creates a computational model (neural network) that copies the classification logic of laboratory testing, allowing field-based determination of soil classification that mirrors laboratory accuracy without requiring complex lab equipment. The model translates simple field observations into classification decisions that replicate professional laboratory outcomes.
Solution Approach 2:
The patent replaces the mechanical/physical laboratory testing system with an information-processing system. Instead of using physical equipment to measure soil properties directly, the system uses computational algorithms that process data from simple field tests to determine classification, substituting mechanical measurement with intelligent data analysis.
2Measurement precision
If laboratory equipment is used for soil classification, then measurement precision is improved, but ease of operation deteriorates
Solution Approach 1:
The system enables practitioners to perform soil classification themselves using simple field tests and the computational model, without requiring specialized laboratory training or equipment. The model handles the complex analysis automatically, allowing users to obtain professional-grade classification results from basic field observations.
Solution Approach 2:
The computational model acts as an intermediary that bridges simple field tests and professional classification outcomes. It translates easily-collected field data into accurate classification results, mediating between the simplicity of field testing and the precision of laboratory standards.
3Manufacturing precision
If standardized soil classification methods are implemented, then manufacturing precision is improved, but device complexity increases
Solution Approach 1:
The patent transforms the classification approach by changing from direct physical measurement parameters to computational parameters. The system uses data from simple field tests as inputs to a neural network model, which outputs standardized classification results. This parameter transformation enables consistent CEB production standards without requiring complex classification equipment.
Solution Approach 2:
The computational model provides a universal classification system that can be applied across different locations and soil types using the same algorithm. This multi-functional approach standardizes CEB production by providing consistent classification outcomes regardless of local conditions, without requiring location-specific complex equipment.
4Ease of operation
If field tests are used instead of laboratory testing, then ease of operation is improved, but measurement precision deteriorates
Solution Approach 1:
The system performs preliminary classification through simple field tests, then uses the computational model to refine and finalize the classification accuracy. The field tests provide initial data, and the neural network processes this data to achieve laboratory-grade precision, combining the accessibility of field testing with the accuracy of professional standards.
Solution Approach 2:
The computational model provides feedback processing of field test data, continuously refining the classification accuracy based on the input observations. The system uses the field test results as feedback to the model, which adjusts its classification output to maintain high precision despite the simplicity of the input data collection method.
Data Source
AI summary
Provided herein is a method for determining a soil classification comprising: obtaining a soil sample; conducting two or more of the following tests in the field, wherein the tests are selected from wash test, a feel test, a test tube particle graduation, or a jar test, to obtain raw data for each test; and calculating the soil classification from the raw data obtained in the field by applying a previously obtained validation dataset obtained from a training and validation soil classification calculation using known samples. The invention also includes compressed earth blocks made using the present invention.