LED Plant Intelligent Lighting System

The growth of plants is inseparable from a suitable light environment.Sunlight is the most important source of light for plants in nature.Appropriate light has a very important impact on the good growth and development of plants.Plant factories are the development direction of modern agriculture.At present,plant factories are developing rapidly.Artificial light source is the key link and core technology of plant factories.Aiming at the high-quality lighting requirements of plants in plant factories,it is an inevitable direction and approach for the development of modern agricultural plant factories to accurately identify the current growth state of plants and provide matching intelligent lighting,so it has practical research significance and value.Aiming at the technical requirements of plant status recognition in the plant intelligent lighting system,according to the characteristics of plant lighting,an LED plant intelligent lighting scheme and system is designed and developed.Monitor the growth status of plants in real time,provide relevant information such as plant diseases,and analyze plant light requirements.The lighting parameters of the system are intelligently adjusted through PWM to match the lighting requirements for the current plant species and conditions for growth,and the lighting application efficiency is improved.First,using the deep learning neural network image processing algorithm,on the basis of VGG-16,VGG-19,Res Net-50 and other network models,an improved model based on the attention mechanism is proposed,and then a more efficient PFR-Net is proposed.A series of common crops such as tomatoes,cucumbers,soybeans,etc.,can efficiently identify whether they have suffered plant diseases,and combined with modern agriculture,it provides a parameter basis for intelligent adjustment of lighting systems.And achieved a classification accuracy of 91.97% on the Plant Village common disease dataset.Then,according to the plant growth state identified by the PFR-Net model,the lighting requirements corresponding to the current state of the plant are obtained,which accurately matches the lighting parameters required for plant growth,and the plant state results can be monitored according to the deep learning neural network algorithm model.A scheme for intelligently regulating plant lighting in the growth cycle.Finally,a LED intelligent plant lighting system based on single-chip PWM linear control is designed and developed.The system light source module is composed of a multi-primary single-color LED chip array.Raspberry Pi is used to collect image information of plant growth status and upload it to the cloud,and the PFR-Net model is used to classify and provide the lighting parameters required by plants,and then intelligently control the plant lighting system to meet the needs of plants.The lighting requirements for efficient growth at different stages provide a system solution for LED plant intelligent lighting for plant factory technology upgrades.