論文的題目是《基于天氣預(yù)報(bào)的水稻灌溉決策強(qiáng)化學(xué)習(xí)方法》。

A reinforcement learning approach to irrigation decision-making for rice using weather forecasts

Mengting Chen, Yufeng Luo

文章介紹了在智能灌溉決策方面的最新進(jìn)展。歡迎下載引用詳見：https://www.sciencedirect.com/science/article/pii/S0378377421001037

文章發(fā)表在科學(xué)導(dǎo)報(bào)ScienceDirect 上：https://doi.org/10.1016/j.agwat.2021.106838

文章要點(diǎn)

●提出并驗(yàn)證了灌溉決策的一種強(qiáng)化學(xué)習(xí)方法。

●通過(guò)明智的學(xué)習(xí)方法解決利用灌溉經(jīng)驗(yàn)和天氣預(yù)報(bào)的不確定性的問(wèn)題。

●該方法能在不損失產(chǎn)量的前提下節(jié)約灌溉水量，縮短灌溉時(shí)間。

●所提出的灌溉強(qiáng)化學(xué)習(xí)方法對(duì)于智能灌溉實(shí)踐具有很好的應(yīng)用前景。

論文摘要

充分利用降雨提高農(nóng)業(yè)用水效率是農(nóng)業(yè)節(jié)水的有效途徑之一。當(dāng)前，天氣預(yù)報(bào)可以用于潛在地節(jié)約灌溉用水，但應(yīng)避免不必要灌溉的風(fēng)險(xiǎn)和由于天氣預(yù)報(bào)的不確定性造成的，可能存在的產(chǎn)量損失。為此，提出了一種基于短期天氣預(yù)報(bào)的深度Q學(xué)習(xí)灌溉決策策略。以南昌地區(qū)水稻為例，驗(yàn)證了該方法的實(shí)用性。收集了南昌附近臺(tái)站2012-2019年水稻生育期的短期天氣預(yù)報(bào)和觀測(cè)氣象資料。比較了常規(guī)灌溉和DQN灌溉兩種灌溉決策策略，并對(duì)其節(jié)水效果進(jìn)行了評(píng)價(jià)。結(jié)果表明，該模型的日降水預(yù)報(bào)性能良好，具有潛在的學(xué)習(xí)和開發(fā)空間。DQN灌溉策略訓(xùn)練后具有較強(qiáng)的泛化能力，可用于利用天氣預(yù)報(bào)進(jìn)行灌溉決策。在我們的案例中，模擬結(jié)果表明，與傳統(tǒng)灌溉決策相比，DQN灌溉產(chǎn)生必要的節(jié)水優(yōu)勢(shì)，灌溉節(jié)水23mm，排水量平均減少21mm，灌溉時(shí)間平均減少1.0倍，產(chǎn)量沒有明顯下降。DQN灌溉策略借鑒了過(guò)去的灌溉經(jīng)驗(yàn)和天氣預(yù)報(bào)的不確定性，避免了天氣預(yù)報(bào)不完善的風(fēng)險(xiǎn)。

Highlights

• A reinforcement learning approach for irrigation decision-making is proposed and tested.

• Past irrigation experiences and uncertainties of weather forecasts are intelligently learned.

• The proposed method can conserve irrigation water and reduce irrigation time without yield loss.

• The proposed reinforcement learning approach for irrigation is promising for smart irrigation practices.

Abstract

Improving efficiency with the use of rainfall is one of the effective ways to conserve water in agriculture. At present, weather forecasting can be used to potentially conserve irrigation water, but the risks of unnecessary irrigation and the yield loss due to the uncertainty of weather forecasts should be avoided. Thus, a deep Q-learning (DQN) irrigation decision-making strategy based on short-term weather forecasts was proposed to determine the optimal irrigation decision. The utility of the method is demonstrated for paddy rice grown in Nanchang, China. The short-term weather forecasts and observed meteorological data of the paddy rice growth period from 2012 to 2019 were collected from stations near Nanchang. Irrigation was decided for two irrigation decision-making strategies, namely, conventional irrigation (i.e., flooded irrigation commonly used by local farmers) and DQN irrigation, and their performance in water conservation was evaluated. The results showed that the daily rainfall forecasting performance was acceptable, with potential space for learning and exploitation. The DQN irrigation strategy had strong generalization ability after training and can be used to make irrigation decisions using weather forecasts. In our case, simulation results indicated that compared with conventional irrigation decisions, DQN irrigation took advantage of water conservation from unnecessary irrigation, resulting in irrigation water savings of 23 mm and reducing drainage by 21 mm and irrigation timing by 1.0 times on average, without significant yield reduction. The DQN irrigation strategy of learning from past irrigation experiences and the uncertainties in weather forecasts avoided the risks of imperfect weather forecasting.

文章來(lái)源：http://irripro.com.cn/