Key Technology Of Optimization And Integration On Regulation Patterns For IoT Of River Crab Efficient Breeding

Over the past three decades,China has made remarkable progress in crab breeding industry.However,the long-term extensive cultivating methods as well as the backward cultivating techniques made the economic benefits of the crab breeding unobvious and caused a weak international competitiveness in China.The rapid expansion of WSN which as a core technology in Internet of Things was widely utilized in various industries of the national economy,and this achieved significant economic and social effectiveness.Combined with smart sensor technique and embedded computing technology,this subject aimed to put the WSN technology into crab breeding industry area.On the basis of the field investigation of the needs of the crab breeders,different wireless sensor network system architecture was designed for three typical the crab breeding modes.Key techniques of sensing control layer,key technology of self-organizing network transmission and the optimization problem of the detection parameters were studied and designed.To sum up,the principal work of this subject mainly included as following:Based on the analysis of the situation of the crab breeding industry,three patterns of WSN systems suitable for different breeding scale were designed,the main performance and operation flow were introduced in detail to illustrate how the three control mode systems were constructed.In order to ensure the accuracy of sensing layer,parameters Parameter calibration and temperature compensation methods were designed through analysis of measurement principle.Variable monitoring period was designed for balance of network energy consumption and ability of responding to dissolved oxygen ultralimit.The handling method of solution oxygen oversaturation was designed based on analysis of the mechanism of oversaturation of dissolved oxygen.In consideration of the control of multi-pump and problem of difference of distribution of dissolved oxygen,combining the control of multi-pump and multi-point measurement,the solution including pond division and node deployment was designed.The collection and control can't be separated from the reliable network transmission technology support.An improved channel assignment and transmission conflict based on competitive B-MAC protocol which supports low power consumption,LPL and CCA backoff mechanism.The multi-hop routing protocol CTP protocol was selected as the self-organizing network routing protocol.Its support for the two-way transmission was designed to satisfy the need of IoT of river crab breeding.The protocol optimizes the transmission path,improves the reliability of transmission,and gives consideration to energy expenditure.There are many parameters to be monitored for the crab breeding environment,and some parameter detection is expensive and poor.On the basis of analysis and research,total nitrogen total phosphorus was determined as the optimal target of detection parameters.The measured sample set was divided into training set and validation set.The typical random forest and GBDT algorithm in integrated learning were used to model the training set data,and the accuracy and generalization ability of total nitrogen total phosphorus was verified in the validation set.All of them have achieved good results and simplify the input features,GBDT showed relatively better performance from the experimental results.Finally Jintan district fishery science and technology demonstration base in Changzhou and three farmers were chosen for the demonstration test,self-organizing,packet loss,the control effect of network transmission and energy saving was tested right on the spot.The feasibility of the system architecture and the effectiveness of the key technologies were verified.This design of key technology of optimization and integration on regulation patterns for IoT of river crab efficient breeding,effectively balances the water quality parameters of data acquisition,processing,economic control,wireless self-organizing network transmission and parameter optimization problem,can be widely used in river crab breeding field.It has broad application prospect and good social and benefits.