.png){kind=icon}
.png){kind=icon}
Return
A model predictive controller for precision irrigation using discrete lagurre networks
.png){kind=icon}
PRE
AI
DOI：10.1016/j.compag.2020.105953.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
The increasing demand for water is depleting the natural water resources worldwide. One of the major consumers of water is the agricultural industry. Traditional means of irrigation results in a lot of water wastage due to which the notion of precision irrigation was introduced using embedded devices. Many of the existing techniques for precision irrigation produce either sub-optimal results or have high computational complexity not suitable for low cost and simple embedded devices such as Raspberry-Pi (RPi). To solve this issue, this paper presents a precision irrigation technique based on discrete time model predictive control (MPC). To reduce the computational complexity of the proposed techniques, the use of Laguerre functions to approximate the control horizon is proposed. Using simulations in MATLAB, the results shows that the proposed technique can approximate the behavior of a discrete linear quadratic regulator with sufficient accuracy. The experimental results in a greenhouse environment, show that the proposed that the proposed technique can reduce water consumption by 30% over a period of 21 days. Moreover, a performance analysis of the proposed technique is presented to show that the proposed technique has significantly lower the computational complexity as compared to other conventional MPC techniques, i.e., it results in a computational complexity at least three time lower than existing techniques. Thus, the proposed precision irrigation control technique can be an attractive choice for real time agricultural application.
Keywords:
Model predictive control
Linear quadratic regulator
Laguerre networks
Precision irrigation
Soil moisture content. Water
AI Summary
Key information extracted from the uploaded paper, including a brief overview, abstract, background, key highlights, visual analysis, and future outlook.
Journal
IF:
8.9
Papers:
10.0K
Citations:
4.8W