| At present,with the rapid development of information technology,the various industries to obtain information on the accuracy,speed and other requirements are increasingly stringent,but access to information,sensor technology cannot meet this requirement.The reason is that in the working environment,the measurement results of sensors are affected by environmental factors,making the accuracy of the measurement results is not high.In recent years,the intelligent sensor system represents the development direction of sensor technology,and it will be the target sensor and non-target sensor measured data sent to the microprocessor for data fusion processing.Because of the different information in the system needs different sensors to detect,which requires us to find a reasonable and effective processing methods to deal with data,and thus make the data fusion algorithm is widely used in multi-sensor data processing system.In a multi-sensor system,data fusion is the combination of data from different sensors.Such processing results not only are more reliable and accurate than the estimation and reasoning obtained from a single data source,but also reduce the impact of environmental factors on the sensor.On the basis of the rapid development of current sensor and data fusion technology,this paper mainly solved the problems of the influence of environmental temperature on the accuracy of sensor measurement.The main contents are as follows:?1?By analyzing the working principles of hall sensor and temperature characteristics,this paper introduced the multiple regression analysis and support vector machine algorithm,and used the genetic algorithm for the selection of least squares support vector machine?SVM?parameters optimization.?2?The two-dimensional calibration experiment was carried out.Based on the measured data,the mathematical model of the regression equation and the least square support vector machine?LS-SVM?algorithm optimized by genetic algorithm were deduced.Additionally,using the multivariate regression method and the improved least squares support vector machine algorithm in hall displacement sensor for temperature compensation.For instance,after the temperature compensation,zero temperature coefficient fell from7.08×10-3°C-1 to 6.75×10-4°C-1 and 6.80×10-4°C-1,the temperature sensitivity coefficient fell from 2.50×10-3°C-1 to 1.78×10-3°C-1 and 2.69×10-3°C-1,the relative error fell from 6.30% to 4.49% and 0.68% respectively.As can be seen from the above data,the two algorithms played a significant role in temperature compensation for the hall displacement sensor.Specifically,the improved LS-SVM algorithm had more obvious effects on the temperature compensation of the sensor,the zero temperature coefficient and temperature sensitivity coefficient were increased by an order of magnitude as well,the relative error had also been greatly improved,so as to achieve the purpose of temperature compensation of the hall displacement sensor.?3?In order to design a real-time displacement measurement system with temperature compensation function,we combined hall sensors,temperature sensors and STM microprocessors to build a hall displacement sensor measurement system.The least squares support vector machine?LS-SVM?algorithm was embedded in the microprocessor.The output of the hall sensor and the temperature sensor were sent to the microprocessor to compensate the temperature of the hall displacement sensor.Thus,it aims to achieve the impact of hall displacement sensor system measurement accuracy of temperature and other factors compensation.In addition,the system is also equipped with LCD liquid crystal display circuit that can display the real-time displacement value and environmental temperature value. |
PDF Full Text Request