Neural Network Growth Analysis System for Plant Component Correlation

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing plant cultivation analysis systems cannot derive a correlation between growth conditions and the components of plants grown under those conditions.

Innovation Solution

A growth analysis system utilizing a neural network that learns from growth condition and component data to output correlations, allowing for the derivation of optimal growth conditions and component information through deep learning.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If conventional plant cultivation analysis techniques are used, then plant component analysis is possible, but correlation between growth conditions and plant components cannot be derived

Engineering Contradiction:

Improveplant component analysis capabilityVSAvoidcorrelation information between growth conditions and plant components

Core Design Contradiction:

Measurement precisionVSLoss of information

Solution Approach 1:

The system performs preliminary data collection and storage of growth condition information and plant component information before analysis. Training data is prepared in advance by collecting paired data of growth conditions and corresponding plant components, which enables the neural network to learn correlations beforehand rather than requiring real-time derivation

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

A neural network is introduced as an intermediary between growth condition data and plant component data. The neural network learns the complex, non-linear relationships between growth conditions and plant components through training, acting as a mediator that derives correlations that would be difficult to obtain through conventional direct analysis methods

Inventive Principle:

Principle #24Intermediary (Mediator)

2Loss of information

If deep learning is applied to derive correlations, then comprehensive correlation analysis is enabled, but system complexity increases

Engineering Contradiction:

Improvecorrelation information completenessVSAvoidneural network system complexity

Core Design Contradiction:

Loss of informationVSDevice complexity

Solution Approach 1:

The neural network is designed with multi-functionality to handle various types of growth condition data (environmental conditions, cultivation management operations, etc.) and predict multiple plant component outcomes. This universal approach allows a single system to perform comprehensive correlation analysis across different data types and plant components, reducing the need for multiple separate analysis systems

Inventive Principle:

Principle #6Universality (Multi-functionality)

Solution Approach 2:

The neural network performs self-learning through automated training processes. The system automatically collects training data, adjusts network parameters, and improves correlation derivation capabilities without requiring manual intervention for each analysis task, thereby managing complexity through automation rather than manual configuration

Inventive Principle:

Principle #25Self-service

Data Source

PatentUS11526730B2Growth analysis system, growth analysis method, and growth analysis program

Publication Date: 2022.12.13 NEC CORP

AI summary

A growth analysis system (1) according to an example aspect of the present disclosure, comprising a correlation deriving unit (60) comprising a first neural network that uses growth condition information as input data to output contained component information as output data. The correlation deriving unit (60) causes the first neural network to learn by repeatedly performing deep learning using, as training data, the growth condition information and the contained component information indicating the component contained in the organisms that are actually grown in accordance with the growth condition.