Researches On Micro-environment Gas Detection Method Of Fresh Grape Cold Chain Based On Sensor Array

It has been proved that,keeping the cold chain micro-environment an appropriate temperature and a certain amount of gas preservation could delay the fruit ripening period,and ensure the quality of fresh life.Therefore,in the process of cold chain logistics transportation how to detect the gas signal real time in the cold chain logistics,and further to explore the change of the gas signal and the quality of fruits,and realize the dynamic monitoring of the cold chain process,has been an urgent key problem in the current cold chain logistics industry.Based on the above issues,the gas detection method based on sensor array for the preservation of fresh grape in cold chain logistics and carries out a series of studies in this thesis.Aiming at the gas components and their physical and chemical volume fraction of fresh-keeping gases,current gas sensor techniques are studied and compared.A variety of Metal Oxide sensors?11 TGS sensors?and electrochemical sensors?2 Honeywell4SO₂-2000?are selected to constitute the initial array.These sensors are tested and analyzed for their performance by a gas sensor testing platform.The tested gas components mainly include sulfur dioxide?SO₂?gas,carbon dioxide?CO₂?gas,and oxygen?O₂?in the cold chain micro-environment.In order to improve the adaptability,the response of the sensor to two kinds of interfering gases is also discussed,which are methane?CH₄?and carbon monoxide?CO?.In order to achieve target gas detection using less sensors to achieve the small volume and low power consumption of the equipment in the cold chain logistics,the methods of gas sensor array optimization are analyzed and researched,and a gas sensor array optimization selection method using forward search algorithm and Probabilistic Neural Network algorithm is proposed to optimize the preliminary structure of the sensor array and selected feature parameters.The optimal sensor and the best characteristic parameters are obtained and the best sensor array is obtained,which is composed of 2 TGS2603,1TGS2610,1 TGS2611,and 1 4SO₂-2000.In order to meet the needs of cold chain logistics micro-environment gas testing,the existing gas sensor test platform has been improved.New gas concentration distributions such as SO₂,CO₂,and O₂,and corresponding detection modules have been added,and existing Metal Oxide sensors arrays have been improved,the electrochemical sensor array test module is newly added.By controlling the gas distribution module a specific concentration of test gas is flowing through the electrochemical sensor array cavity and the Metal Oxide sensors sensor array cavity in order to meet the requirements of the hybrid array test.Finally,the response of three typical gases in the cold chain micro-environment within the required concentration range are tested systematically by the selected optimal array.The response signals of the sensor array are processed in conjunction with the Probabilistic Neural Network,and finally the quantification and recognition results of SO_2,CO₂ and O₂ are compared with those of commonly used BP neural networks.Aiming at the gas composition in the cold chain micro-environment and the actual test situation,a portable detection system scheme based on KL25 single chip microcomputer in the cold chain micro-environment is proposed.