Plant Growth Status Analyser based on WOA-LSTM Prediction Model

Plant electrical signal is a kind of non-smooth and non-linear weak signal, and it is a low-frequency random signal. The plant growth state analyzer of the plant electrical signal is to receive the data from the plant electrical signal acquisition board first, and then preprocess the electrical signal data collected from the acquisition board to finally obtain the plant electrical signal with high signal-to-noise ratio, extract the maximum value, minimum value, peak value, average value, area, rising speed, falling speed as the feature value, and then get the deep learning model through the training of WOA-LSTM neural network to obtain the collected plant electrical signal and a relationship with the influences on the plant growth state. model to get the collected plant electrical signals and a relationship with the influence of plant growth state. The experimentally trained WOA-LSTM neural network model has a mean square error MSE and correlation coefficient R of 0.0232 and 0.9408, respectively. which is a good judgment, proving that the model has a high judgment accuracy. Finally, the model was deployed to the development board and the plant growth status was displayed on the feedback interface using QT. The plant growth state analyzer speculates the current growth state of the plant through the change of the electrical signal of the plant in order to adjust the environment of the plant in time. Through the electrical signal to determine whether the plant grows in a good direction or a bad direction to control it, it helps to promote the development of agriculture in a more accurate, efficient and sustainable direction.