根据水产品无水低温保活运输监测需求,设计了可植入式血糖传感器及信号处理电路,以实时获取血糖传感信号,信号经过软件滤波进行去噪和平滑处理后,构建了血糖变化的时序预测模型;以鲟鱼作为实验对象,对所提出的方法进行了实验验证.结果表明:所设计的可植入式血糖传感器能在鲟鱼体内对血糖信号进行稳定的采集,传感器信号经过信号调理电路后输出拟合程度达到0.960 8,电路灵敏度为27.047 mV/nA,零点漂移量为722.83 mV;所构建的血糖信号自回归移动平均模型(ARIMR)预测值与真实值之间的平均绝对误差为-0.014 mmol/L,平均相对误差为-0.117％,预测精度和平滑度较其他模型具有优势.该信号监测方法对于提高水产品无水低温保活全程透明度和追溯性,并推断水产品生命体及营养物质变化具有重要参考价值.%Waterless and low temperature keeping alive transportation is considered as a kind of promising technology to improve aquatic animals' survival and save the transport costs,and monitoring biological signal is useful to improve the transparency and traceability of transportation process.A monitoring method of glucose was presented,which consisted of three components:a signal conditioning circuit,a software digital filter and a prediction method.The signal conditioning circuit was responsible to sense glucose signal by the implantable glucose sensor and signal pretreatment by a hardware filters;the software digital filter can filter and smooth the amplified signal to provide reliable data for the dynamic character analysis.The time series method served to predict change of glucose.Sturgeons were selected as experimental objectives to validate the reliability of the method.The experiment result showed that the implanted glucose sensor can sense data accurately with sensitivity of 27.047 mV/nA,the zero drift of 722.83 mV and signal output fitting of 0.960 8 after signal conditioning circuit;the average absolute error between the predicted value by autoregressive integrated moving average model(ARIMR) and the real sensed value was-0.014 mmol/L,and the average relative error was-0.117％,which indicated that the ARIMR model had advantages in prediction accuracy and smoothness compared with other prediction models;the proposed monitoring method can be adopted as a reference value to monitor the change of glucose and obtain the change of biological nutrients during waterless keeping alive transportation of the aquatic product.
展开▼
