Optimization of the technology for applying transglutaminase to produce surimi gel from tilapia by neural network

The effect of main processing condition(TGase addition,reaction temperature,reaction time) on the gel strength of frittered meat of tilapia surimi gel were investigated. The processing parameters were optimized by neural network and the neural network model was established. The results showed that the established neural network model was more reliable which could accurately predict the relationship of the gel strength and processing condition. The relative error of predicted values and experimental values was relatively small(R2=0.9936). When the condition of TGase addition of 37U/100 g surimi,reaction temperature of 37.5℃,reaction time of 43 min,the gel strength reached(9103±725)g·mm. Under these processing conditions,the gel strength was appropriate.