Scholarlink

The presence of dissolved oxygen (DO) in water is an ecological factor that plays an important role in fostering the growth of aquatic goods. The amount of DO in recirculating aquaculture systems is a critical indicator of control. Its composition and the dynamic fluctuations have a substantial impact on the rohu fish capacity to grow in a healthy manner. The amount of dissolved oxygen in water is determined by several biological, physical and chemical features, including PH, BOD; EC, Temperature; Ca; Mg, Nitrate; Turbidity; PO4, Na and NO3 respectively. Because of this, it is essential to conduct dissolved oxygen tests on water. In former times, individuals predicted the water quality using conventional methods such as laboratory examination. Traditional approaches, on the other hand, are unable to collect non-linear and non-stationary data regarding water quality. In addition, laboratory methods demand a significant amount of both time and physical labour. In the fields of ML, DL and artificial neural networks (ANN), long short-term memory (LSTM), and gated recurrent units (GRU) are utilised quite frequently. ANN requires a significant quantity of data and has a high propensity for slipping into local minima. Although it is uncertain which of them is more suited for predicting DO in Aquaculture. To improve the accuracy and efficacy of DO prediction, an integrated forecasting model based on an LSTM-GRU hybrid model was developed. This model was used to estimate the concentration of DO in water for rohu fish development and survival. Compared to the current models, the proposed hybrid model produces more precise values.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Forecasting Liquefied Oxygen in Rohu Fish Ponds Using Hybrid Recurrent Neural Network Model

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • A. Peda Gopi,
  • B. Indira Sai,
  • P. Sravani,
  • C. H. Bhavana Sri,
  • Y. Mounika

摘要

The presence of dissolved oxygen (DO) in water is an ecological factor that plays an important role in fostering the growth of aquatic goods. The amount of DO in recirculating aquaculture systems is a critical indicator of control. Its composition and the dynamic fluctuations have a substantial impact on the rohu fish capacity to grow in a healthy manner. The amount of dissolved oxygen in water is determined by several biological, physical and chemical features, including PH, BOD; EC, Temperature; Ca; Mg, Nitrate; Turbidity; PO4, Na and NO3 respectively. Because of this, it is essential to conduct dissolved oxygen tests on water. In former times, individuals predicted the water quality using conventional methods such as laboratory examination. Traditional approaches, on the other hand, are unable to collect non-linear and non-stationary data regarding water quality. In addition, laboratory methods demand a significant amount of both time and physical labour. In the fields of ML, DL and artificial neural networks (ANN), long short-term memory (LSTM), and gated recurrent units (GRU) are utilised quite frequently. ANN requires a significant quantity of data and has a high propensity for slipping into local minima. Although it is uncertain which of them is more suited for predicting DO in Aquaculture. To improve the accuracy and efficacy of DO prediction, an integrated forecasting model based on an LSTM-GRU hybrid model was developed. This model was used to estimate the concentration of DO in water for rohu fish development and survival. Compared to the current models, the proposed hybrid model produces more precise values.