Plant growth prediction method, device and equipment based on machine learning model

Abstract

The invention provides a plant growth prediction method and device based on a machine learning model, equipment and a computer readable storage medium. The machine learning model comprises a first sub-model, a second sub-model and a third sub-model, and the plant growth prediction method comprises the steps of obtaining greenhouse control parameters, weather data and plant state data; utilizing the first sub-model to generate a first feature vector according to the greenhouse control parameters; generating a second feature vector according to the weather data by using the second sub-model; generating a third feature vector according to the plant state data by using the third sub-model; and generating a plant growth prediction result after the predetermined time period based on the first feature vector, the second feature vector and the third feature vector.

Description

technical field [0001] The present disclosure relates to the field of plant growth prediction, and in particular, relates to a plant growth prediction method based on a machine learning model, a plant growth prediction device and equipment, and a computer-readable storage medium. Background technique [0002] When predicting plant growth, a physical simulator based on physical formulas is usually used to predict the daily growth rate of plants based on the light intensity, temperature, humidity, and carbon dioxide concentration of the greenhouse at each moment, combined with the metabolism of plants in different growth stages. growing situation. This physical simulator based on physical formulas uses physical formulas to deduce the daily growth of plants in different greenhouse environmental conditions. However, due to the fact that there are many factors that affect the state of plants in a real greenhouse, and there are interactions between different influencing factors ,...

AI Technical Summary

Problems solved by technology

This physical simulator based on physical formulas uses physical formulas to deduce the daily growth of plants in different greenhouse environmental conditions. However, due to the fact that there are many factors that affect the state of plants in a real greenhouse, and there are interactions between different influencing factors , the existing physical simulators cannot accurately simulate the effects of different weather conditions on plants in real greenhouses. Therefore, the prediction results of physical simulators on plant growth states usually have a large gap with real greenhouses

In addition, the simulation speed of the physical simulator based on the physical formula is slow, and it usually takes 3-5 seconds to complete the simulation of the plant for one day, so the time cost required for simulating the long planting cycle and finding the optimal planting strategy is relatively low. Big

Images