IMPLEMENTASI ALGORITMA LONG-SHORT TERM MEMORY (LSTM) DALAM PENENTUAN WAKTU PENYIRAMAN DENGAN PARAMETER KELEMBAPAN TANAH UNTUK TANAMAN CABAI MERAH
Total View This Week0
Total View This Week0
Institusion
Institut Teknologi Perusahaan Listrik Negara
Author
PUTRI M, DITAVIA AINNAYA
Haris, Abdul
Sikumbang, Hengki
Haris, Abdul
Sikumbang, Hengki
Subject
Teknik Informatika
Datestamp
2023-06-05 07:01:56
Abstract :
This study aims to produce a monitoring system for water sprinkling by implementing the Long Short Term Memory (LSTM) algorithm on soil moisture data to form an intelligent computing system. Watering the red chili plants is still done manually so it takes a long time and the lack of knowledge about soil moisture is considered inefficient because one of the most important parameters for the growth of red chili plants is soil moisture. To help make decisions in giving chili plant water with soil moisture parameters made using the Long Short Term Memory (LSTM) Algorithm. The LSTM algorithm has three stages, namely data normalization, calculations using the LSTM method, and model evaluation using the Mean Absolute Error (MAE) and Root Mean Square Error (RMSE). The data used in the study were 100 soil moisture data. The results obtained are in the form of graphical modeling where the right time to water is at soil moisture values ranging from 0.6 ? 0.8 so the system will activate the selenoid valve to open the faucet so that it can flow water to moisten the soil while when the value is below 0.6 the system will close or not open tap because the soil is already at the right moisture point in the application of water. The level of accuracy is close to the measurement data with an error value (MAE) reaching 0.11 and an error value (RMSE) reaching 0.14.
This study aims to produce a monitoring system for water sprinkling by implementing the Long Short Term Memory (LSTM) algorithm on soil moisture data to form an intelligent computing system. Watering the red chili plants is still done manually so it takes a long time and the lack of knowledge about soil moisture is considered inefficient because one of the most important parameters for the growth of red chili plants is soil moisture. To help make decisions in giving chili plant water with soil moisture parameters made using the Long Short Term Memory (LSTM) Algorithm. The LSTM algorithm has three stages, namely data normalization, calculations using the LSTM method, and model evaluation using the Mean Absolute Error (MAE) and Root Mean Square Error (RMSE). The data used in the study were 100 soil moisture data. The results obtained are in the form of graphical modeling where the right time to water is at soil moisture values ranging from 0.6 ? 0.8 so the system will activate the selenoid valve to open the faucet so that it can flow water to moisten the soil while when the value is below 0.6 the system will close or not open tap because the soil is already at the right moisture point in the application of water. The level of accuracy is close to the measurement data with an error value (MAE) reaching 0.11 and an error value (RMSE) reaching 0.14.
Download
book
BibTex
Latex, Jabref
cloud_download
Original Resource
url resource Institution
