| In recent years,in order to improve the transportation efficiency,many enterprises have been overloaded.Therefore,the transportation department and related researchers pay more and more attention to the research of vehicle weighing system.At present,the vehicle weighing system mainly includes the separate type and the vehicle type.The research on the separate type vehicle weighing system started early and has high measurement accuracy,but its detection cost is high,the efficiency is low,and it can not be detected in real time;while the research time of the vehicle weighing system is short,the technology is not mature,and the measurement accuracy is lower than the separate weighing system,but it has the advantages of real-time detection.Therefore,the research of vehicle weighing system is of great significance.In this paper,the mature TPMS technology is used to establish the mathematical model between tire pressure and vehicle load through multi-sensor data fusion algorithm,so that the driver can understand the vehicle load information at the vehicle end,and reduce the traffic safety accidents caused by vehicle overload.In addition,the system also has the function of real-time monitoring the tire pressure and temperature in the tire,which greatly improves the driving safety.The main contents of this paper are as follows:(1)Through mechanical analysis and mathematical theory derivation,the function model of vehicle load and tire pressure is analyzed under two conditions: temperature invariant and temperature change when the vehicle is at rest.Secondly,the various interference factors that affect the measurement accuracy of the vehicle under moving conditions are analyzed.(2)The components selection and hardware circuit design of data acquisition,wireless transmission and vehicle data receiving and processing are completed.The schematic diagram and PCB board are drawn by Altium designer,and then processed and welded into real objects.(3)The software design of data collection module,wireless transceiver module,data receiving and processing module,serial communication module and display module has been completed.Among them,the acquisition module completes the tasks of data collection,data preprocessing,etc.Secondly,the low-power software design is completed.In addition,the format of wireless communication data instruction pack is configured,and the software design of a master-slave wireless communication protocol is completed.(4)Taking the six wheel truck as the carrier,the tire pressure values corresponding to six tires are obtained by changing the load as the sample data,and 85 groups of sample data are trained by BP neural network to establish the mathematical model between the tire pressure information of each tire and the truck load.The remaining 15 groups of data are used as the prediction samples to test the advantages and disadvantages of the training model.(5)In order to improve the global search speed without destroying the diversity of the population,a new strategy of genetic operation probability phasing is presented.The classical GEP algorithm is improved by designing the gene mutation rate,gene recombination rate and gene transposal rate stages of GEP algorithm.The vehicle load model generated by the classical GEP algorithm is fused,the vehicle load model generated by the improved GEP algorithm is compared with the vehicle load model generated by the BP network.The improved GEP calculation is proved by combining various indexes.Superiority of the law model. |
