A Forecasting Method Based on ARIMA Model for Best-Fitted Nutrition Water Supplement on Fruits
Saravanakumar Venkatesan1, Sathishkumar V E2, Changsun Shin3, Yubin Kim4, Yongyun Cho5
1Saravana Kumar Venkatesan, Department of Information and Communication Engineering, Sunchon National University, Suncheon, South Korea.
2Sathish Kumar V. E., Department of Information and Communication Engineering, Sunchon National University, Suncheon, South Korea.
3Changsun Shin, Department of Information and Communication Engineering, Sunchon National University, Suncheon, South Korea.
4Yubin Kim, ELSYS Co. Ltd. Suncheon, Republic of Korea.
5Yongyun Cho*, Department of Information and Communication Engineering, Sunchon National University, Suncheon, South Korea.
Manuscript received on November 21, 2019. | Revised Manuscript received on December 15, 2019. | Manuscript published on December 30, 2019. | PP: 3167-3173 | Volume-9 Issue-2, December, 2019. | Retrieval Number: B4226129219/2019©BEIESP | DOI: 10.35940/ijeat.B4226.129219
Open Access | Ethics and Policies | Cite | Mendeley
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: The main focus of this research is to promote a forecasting method in the greenhouse of cultivation for the nutrition water level of strawberry fruits. In the greenhouse of cultivation, this study selects strawberry fruits as the focus on research. With adequate nutrition water supply conditions, the autoregressive integrated moving average and seasonal autoregressive integrated moving average (ARIMA-SARIMA) were utilized to create forecasting for the nutrition water level of strawberry leaves in the fruit greenhouse of cultivation, thus forecasting strawberry’s nutrition water rate through greenhouse environmental parameters. Next, the multi-scale feature vectors of greenhouse temperature and nutrition water parameters in the greenhouse have been extracted by using the data pre-processing method to eliminate the testing and training value of variables, thus improving the forecasting and generalization ability of the model. The extracted feature vectors have been used to train and optimize the SARIMA model, finally obtaining the forecasting model of nutrition water rate of strawberry fruits leaves in the greenhouse of cultivation, which has been compared in experiments with the autoregressive integrated moving average and seasonal autoregressive integrated moving average (ARIMA – SARIMA) model. The results indicate that when training samples become a certain amount, the forecasting accuracy and regression fitting degree of ARIMA – SARIMA can be higher than that of the two traditional models. We forecasted that the strawberry greenhouse included 233 samples collected from a strawberry greenhouse in South Korea, and the 6 variables involved are greenhouse maximum temperature, greenhouse minimum temperature, greenhouse average temperature, quality of nutrient water, humanity, and CO2 , which would influence the strawberry growth in production concentration directly or indirectly with the variation of nutrition water every day.
Keywords: Nutrition Water, Greenhouse Average Temperature, Humanity, Co2, ARIMA and SARIMA Model.