Design Of Multi-sensor Water Environment Monitoring System Based On STM32

With the development of society and economy,the activities of human beings to obtain economic benefits have caused different degrees of water pollution in various places.The water environment is damaged,the suspended matter in the water body increases,and the transparency decreases.Seriously polluted water bodies may even turn black.Water pollution has affected people’s daily water use,crop growth and other aspects.Based on these issues,this paper focuses on water environment management and protection.A new water environment monitoring system is proposed by combining sensor technology and wireless sensor network technology.This paper focuses on the design and development of the water environment monitoring system and the research on the data fusion algorithm.The main work is as follows:Based on these issues,this paper focuses on water environment management and protection.Combined with sensor technology and wireless sensor network,a new water environment monitoring system is proposed.Emphasis is placed on the design and development of water environment monitoring systems and the research on data fusion algorithms.The main work is as follows:（1）The overall design of the water environment monitoring system is briefly introduced.A multi-sensor water environment monitoring system based on STM32 is designed and completed.It selects five sensors such as temperature,PH,TDS（total dissolved solids）,turbidity,conductivity and wireless communication modules to form a wireless sensor network.Use STM32 microprocessor to control the sensor to collect data.The collected data is transmitted to the sink node through the wireless local area network composed of Wi Fi wireless communication technology.After the STM32 microprocessor processes the data received by the sink node,it uploads the data to the cloud server through 4G communication technology.Use C# language to write the application management software on the PC side,display the data and save it.（2）In order to improve the accuracy of the data collected by a single sensor,a data fusion method of the same type of sensor using BP neural network is proposed.Correct and optimize the weights of the input layer and output layer to construct the network.After classifying the data collected by the collection node.The fusion value of each sensor is calculated separately,and the fusion error is obtained.Comparing the standard value with the fusion value,the fusion results of similar sensors are obtained.It shows that the algorithm has high accuracy,feasibility and reliability.（3）In order to improve the accuracy of data collected by multiple sensors,Mahalanobis distance is used to modify the combination rules of evidence theory for data fusion of multiple different types of sensors.The BPA（Basic Probability Assignment）is obtained by first fusing the data collected by the two sensors.Then,the BPA obtained from each group is fused according to the revised evidence theory combination rules to obtain the BPA of the water environment.Finally,the reliability of the data collected by each sensor is obtained.It provides a theoretical basis for the reliability of the data collected by the water environment monitoring system,and improves the accuracy and reliability of the data collected by the sensor.The feasibility and robustness of the water environment monitoring system are increased.