| At present,agricultural production has gradually shifted from high-speed development to high-quality development.The rough field agricultural production model is no longer suitable for the development requirements of modern agriculture,and the microclimate model for fine planting has been widely used in agricultural production.Plant factories,a typical representative of facility agriculture,have developed particularly rapidly.In the process of plant factory construction,building an environmental information collection system based on wireless communication network is an important basis for realizing intelligent control of plant factory.However,a large number of data collection nodes in wireless sensor network lead to high redundancy of environmental data and affect the accuracy of data collection.Therefore,it is necessary to fuse the collected environmental data.Accordingly,this paper builds a plant factory monitoring system based on multi-sensor data fusion,and uses similar multi-sensor data fusion algorithms to reduce data redundancy and improve data accuracy,so as to achieve accurate collection and facility of environmental data in plant factories.intelligent control.The main research contents of this paper are as follows:（1）Through the analysis of the environmental impact factors of the plant factory,the relevant requirements of crop growth are determined,and the design and realization of the system functions of the lower computer and the upper computer of the plant factory are carried out according to the requirements of crop growth.The lower computer mainly realizes the collection of environmental parameters and the control of facilities.It is divided into two parts: the hardware design of the lower computer and the software design of the lower computer: the hardware design of the lower computer mainly carries out the selection of wireless communication modules and sensors,the hardware design of the main control unit and the terminal node.Hardware design;lower computer software design mainly carries out the transplantation of μC/OS-III real-time operating system and the development of sensor driver.The upper computer mainly realizes the function of human-computer interaction,and designs a C/S client based on the QT framework to perform real-time display of environmental data and automatic control of facilities.（2）Aiming at the unavoidable problems of wireless sensor network such as high data redundancy,high transmission delay and great influence by environmental noise,a two-level data fusion model is used to fuse the collected environmental data of the same type to obtain information about a certain environmental parameter.consistent interpretation.First,use median filtering to smooth the environmental data with gross errors,and perform first-level data fusion.The advantage of using median filtering is that it will not lose data details,so that the relevant data collection results are more in line with the actual situation;then use the adaptive weighted fusion algorithm The second-level fusion is performed on the smoothed data,and the fusion value represents the environmental information of the entire plant factory for subsequent intelligent control of the facility.（3）Jointly test the relevant functions of the upper computer and lower computer in the plant factory,and analyze the experimental results related to actual data fusion.First,use the multi-function serial port debugging assistant to test the environmental parameter acquisition and facility control functions of the lower computer,and test the real-time environmental data display and facility control functions of the upper computer client.It can dynamically refresh real-time parameters and realize automatic and manual control of control facilities,indicating that the plant The plant monitoring system functions normally.Secondly,a two-level data fusion model is used to process environmental parameters.The experimental results show that the redundancy of data is reduced by 83.3%,which greatly reduces the amount of data sent.At the same time,the accuracy of the data is improved by about 8%.In summary,the use of the twolevel data fusion model achieves the purpose of reducing data redundancy and improving data accuracy,and basically meets the requirements of plant factory data processing accuracy. |
