一种基于多组分营养液控制的马铃薯无土栽培系统及方法

By using a multi-component nutrient solution control system based on CNN-GRU and DRL models, the problems of poor nutrient solution formulation and low control precision were solved, realizing the precision, intelligence and efficiency of soilless potato cultivation, and improving potato yield and quality.

CN120770321BActive Publication Date: 2026-07-17SHANXI AGRI UNIV
View PDF 1 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
SHANXI AGRI UNIV
Filing Date
2025-07-07
Publication Date
2026-07-17

AI Technical Summary

Technical Problem

Existing soilless potato cultivation technologies suffer from insufficient nutrient solution formulation, low control precision, and inadequate intelligent technology, failing to meet the dynamic needs of different growth stages and thus limiting the development and application of soilless cultivation.

Method used

A multi-component nutrient solution control system based on CNN-GRU and DRL models is adopted. The system collects growth, environmental and nutrient solution data in real time through the data monitoring module, uses the CNN-GRU model to predict nutrient solution demand, combines the DRL model to generate dynamic regulation strategies, and optimizes the nutrient solution formula through multi-storage tank collaborative control, environmental auxiliary regulation and feedback correction unit.

Benefits of technology

This has enabled precise, intelligent, and efficient management of potato growth, improved the accuracy and sensitivity of nutrient solution control, reduced labor costs, and promoted healthy potato growth and increased yield.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN120770321B_ABST
    Figure CN120770321B_ABST
Patent Text Reader

Abstract

本申请实施例公开了一种基于多组分营养液控制的马铃薯无土栽培系统及方法，该系统包括数据监测模块、方案调控模块和方案执行优化模块，其中：数据监测模块用于实时监测并采集马铃薯栽培数据，包括长势监测数据、环境监测数据和营养液监测数据；方案调控模块用于根据卷积神经网络‑门控循环单元（Convolutional Neural Network‑Gated Recurrent Unit，CNN‑GRU）模型，结合马铃薯栽培数据，得到营养液需求数据，根据深度强化学习(Deep Reinforcement Learning，DRL)模型，结合营养液需求数据和营养液监测数据生成动态营养液调控策略；方案执行优化模块用于执行动态营养液调控策略，调控环境监测数据和营养液配方，并根据营养液参数和马铃薯生长参数优化营养液配方。本申请有助于提升马铃薯栽培的精准化、智能化和高效化。
Need to check novelty before this filing date? Find Prior Art