Aquaculture dissolved oxygen concentration online forecasting method and system

Abstract

The invention provides an aquaculture dissolved oxygen concentration online forecasting method and an aquaculture dissolved oxygen concentration online forecasting system. The method comprises the following steps of: acquiring water quality index and related meteorological factor data in a scheduled time period to establish an original data set; performing data standardized pre-processing on the original data set by using a normalization method to acquire a training sample data set of a least square support vector regression (LSSVR) machine model; training the LSSVR machine model by using the training sample data set, and optimizing the parameters of the LSSVR machine model to acquire an optimal LSSVR machine model; and acquiring water quality index and related meteorological factor data of an aquaculture ecological environment on line in real time, and inputting the acquired water quality index and related meteorological factor data to the optimal LSSVR machine model to acquire a dissolved oxygen concentration forecasting value. By using the method and the system, accurate and efficient forecasting of the aquaculture dissolved oxygen concentration is realized.

Description

technical field [0001] The invention relates to the field of information technology and the interdisciplinary field of aquaculture technology, in particular to an online prediction method and system for dissolved oxygen concentration in aquaculture. Background technique [0002] Dissolved oxygen in aquaculture water is a necessary condition for the survival of aquatic animals and plants. Fast and accurate prediction of dissolved oxygen concentration will play an important role in preventing water quality deterioration, improving aquatic product quality and healthy aquatic product farming, and promoting fishery information modernization. [0003] At present, the hardware of the commonly used water quality automatic monitoring and prediction system is based on the online automatic analysis instrument. The prediction method adopts methods such as polynomial regression, mathematical statistics, gray system theory, neural network model, and water quality simulation model. Predic...

AI Technical Summary

Problems solved by technology

[0003] At present, the hardware of the commonly used water quality automatic monitoring and prediction system is based on the online automatic analysis instrument. The prediction method adopts methods such as polynomial regression, mathematical statistics, gray system theory, neural network model, and water quality simulation model. Prediction has achieved certain results, but there are still some problems in the application process: (1) lack of all-weather and long-term monitoring and forecasting of water quality changes; (2) dissolved oxygen concentration is easily affected by multiple parameters such as water quality and meteorology However, the existing methods do not fully consider the influence of dissolved oxygen and other parameters, and the monitoring parameters are few, and the prediction methods have defects such as nonlinearity, uncertainty, high dimensionality, and easy to fall into local minimum points, which seriously Reduced the effectiveness of its application

Images