Biochemical Oxygen Demand Detection Based On Image Technology And Device Empolder

BOD testing is one of the important aspects of water quality testing.The sewage treatment plant uses it as the main parameter to the process design and processing capacity,and it is also one of the main measures to measure the efficiency and operation status of the sewage treatment plant.The traditional detection methods are time consuming,labor-intensive,and the chemical properties cannot be reflected in time,and the disadvantages of the new BOD test instruments are high cost and complexity of the system.Therefore,it is significant to design and develop an accurate and fast BOD on-line detection device.In this paper,based on the iodometric method and computer vision,according to the basic principles of spectrophotometry,using machine vision technology to carry out related BOD detection research,developed a new type of BOD on-line detection device.The hardware part of the device includes a computer,a shadowless regulated light source,a relay reaction tank,and a valve.The software part includes an image acquisition program,Image processing program and dosing control program.The hardware provides stable external conditions for chemical reactions,and software allows precise control of dosing.Through comparison experiment of natural light source,fluorescent lamp and shadowless regulated light source,the best experimental light source was a shadowless regulated light source.The test determined the best color characteristic value as R value,and the optimal titration amounts of sodium thiosulfate were 5 mL and 15 mL,respectively.Experimental data analysis,when the BOD value is less than 2mg/L,we use the third set of linear models,the two models measured at intervals of 5 days areY=-0.05048x+187.1,R~2=0.9764;Y=-0.05439x+187.2,R~2=0.989.When the BOD value is greater than 2 mg/L,the first set of linear models is used.The two linear models are Y=-0.03858x+190.8.R~2=0.9747;Y=-0.0458x+192.7,R~2=0.988.According to the results of multiplemeasurements,a comparative analysis was performed.The selected range of solutions for[0,620],[0,558],and[0,496]were determined to determine the linear interval.After 5 days,[0,558]?[0,527]?[0,434]linear intervals.The first measured linear interval[0,620],[0,558],and the second measured linear interval[0,496],[0,434].The relational model was written into the automatic detection module,and a repetitive inspection experiment was performed to obtain the absolute error.The experimentalresults show that the measured BOD values are within the specified standard intervals and the two experimental models obtained are reliable.For thedetection of BOD by the image method of this paper,computer vision and image processing techniques are used to determine the titration end point,instead of the human eye,the titration is more accurate.Although it is the same time as the traditional method of iodometry,but filled the blank image detection BOD.During the experiment,the entire process of dosing was controlled by a computer,which was more accurate than manual dosing.