Soft sensing of dissolved oxygen in fishpond via extreme learning machine

Dissolved oxygen is an important water quality index in aquaculture. It represents the environment of fish growth. At present it can´t be measured online precisely, so the control and optimal operation is hardly to be achieved. To deal with this problem, a soft sensing method for dissolved oxygen based on extreme learning machine (ELM) is proposed and a measuring device is also developed to achieve the proposed method. Industry experiments are conducted in the aquaculture production process and the results show the effectiveness of this method.