Study On Multi-sensor Information Fusion And Its Application On The Measurement System Of Soot Particles Concentration

Soot particles is an important factor in the formation of atmosphericpollution,and the measurement of soot particles concentration is necessary. In thisthesis, the automatic tracking of flue gas flow rate based on pitot tube parallel to theprinciple of isokinetic isokinetic sampling method soot collection system, anintegrated microprocessor, pressure sensors, temperature sensors, moisture contentsensors and other sensors within the host computer. As the dust concentrationmeasurement is a complex topic of theoretical and experimental techniques,because the measurement process needs to collect a variety of parameters, but alsosusceptible to interference from the scene environment. Paper build a multi-sensordata fusion system for this uncertainty and nonlinear problems of dust concentrationmeasurements, which effectively increase the dust concentration measurementaccuracy.Papers on decision-making D-S evidence theory depth study, summarizes themain issues of evidence theory exists, and proposed the corresponding solution. Aparticular focus on D-S evidence theory of conflict in the evidence, we propose anew synthetic rules. The credibility of the rules of evidence by the conflictinginformation source allocation, focusing on the consistency of the evidence reflects the extent of the use and operation. Theoretical analysis and numerical experimentsshow that the rules proposed for the synthesis of highly conflicting evidence andconsistency of evidence are very effective.Finally, the paper selected20groups of typical soot concentration measurementsamples, improved integration rules based on D-S evidence theory, the feasibilityand effectiveness of the multi-sensor information fusion system constructed in thispaper can provide accurate and reliable data for field engineers and technicians.