鱼塘溶氧量监测分析数据

定时采集鱼塘的溶氧、温度数据，对鱼塘环境进行监测，如有异常及时报警，保障各个养殖户养殖环境的安全，降低养殖风险。通过溶氧量分析，帮助养殖户及时地采取调水措施，平衡水体环境，保障鱼的产量；本数据结合鱼苗生长情况，可为水产养殖科学研究提供数据参考，同时可以为水产养殖业的数字化转型提供数据支撑，将经验用于外部环境，促进水产养殖产业的可移植化发展。1.使用双探头自清洁荧光法溶氧传感器，实时采集鱼塘的溶氧、温度数据，通过无线网络将数据传输给数字渔村运营服务管理平台，实现平台对鱼塘水质的实时监测。一旦发现溶氧偏低，及时通知养殖户开启增氧设施，保障养殖环境。
2.利用人工智能和深度学习在数据挖掘及信息特征提取上的优势，从海量实验样本数据中学习地理位置、养殖品种、天气等多源环境参数间的复杂高维关联，结合时间序列分析方法，得到各个时间点溶氧量的三道轨线：上限、下限和平均线。将设备采集的每个鱼塘每个时间点溶氧数据与三道轨线同一时刻的数值进行对比，若在上线与下线范围内，则正常，若在范围外，计算其偏离值，即高于上限=|（上线-溶氧）|/上线*100%、低于下限=|（下限-溶氧）|/下限*100%，若偏离值高于20%为异常。最终再按天统计总异常频率，异常频率=异常次数/监测总次数*100%，若超过30%，则水质需要人工介入进行测水、调水等工作。

This dataset conducts real-time monitoring of fishery pond environments by periodically collecting dissolved oxygen (DO) and temperature data, and triggers timely alarms in case of abnormalities, so as to ensure the safety of the breeding environment for all farmers and reduce breeding risks. Through DO concentration analysis, farmers can be assisted to take water regulation measures in a timely manner to balance the aquatic environment and ensure fish yield. Combined with the growth status of fish fries, this dataset can provide data references for scientific research in aquaculture, as well as data support for the digital transformation of the aquaculture industry, and help transplant proven experiences to external scenarios, promoting the transplantable development of the aquaculture industry.
1. A dual-probe self-cleaning fluorescence-based dissolved oxygen sensor is used to collect DO and temperature data of fish ponds in real time. The data is transmitted to the Digital Fishing Village Operation Service Management Platform via wireless networks, enabling the platform to conduct real-time monitoring of pond water quality. Once low DO concentration is detected, farmers will be notified promptly to turn on aeration facilities to ensure the breeding environment.
2. Leveraging the advantages of artificial intelligence (AI) and deep learning in data mining and information feature extraction, complex high-dimensional correlations among multi-source environmental parameters (such as geographical location, breeding species, and weather) are learned from massive experimental sample data. Combined with time series analysis methods, three trajectory lines of DO concentration at each time point are obtained: upper limit, lower limit and average line. The DO data of each pond at each time point collected by the device is compared with the values of the three trajectory lines at the same moment. If the data falls within the range between the upper and lower limits, it is considered normal; if it is outside the range, the deviation value is calculated as follows: for values higher than the upper limit, deviation = |(upper limit - DO)| / upper limit * 100%; for values lower than the lower limit, deviation = |(lower limit - DO)| / lower limit * 100%. If the deviation value exceeds 20%, the data is judged as abnormal. Finally, the total abnormal frequency is counted daily, where abnormal frequency = (number of abnormal times / total number of monitoring times) * 100%. If the abnormal frequency exceeds 30%, manual intervention such as water quality testing and water regulation is required for the water quality.

搜集汇总

数据集介绍

特点

鱼塘溶氧量监测分析数据包含鱼塘的溶氧量、水温等关键指标，用于实时监测鱼塘水质，异常时及时报警。数据通过先进传感器和AI技术采集分析，应用于水产养殖环境监测和科学研究，促进养殖业数字化转型。