Study On The Temperature Characteristic Analysis And Compensation Of Capacitance Weighing Sensor

Capacitance sensors, because of their high sensitivity, lower power consumptionand simple structure, are extensively used in various applications to measure force,pressure, position. However, some drawbacks of capacitance sensor are that:(1)characteristics of the nonlinearity of the capacitance weighing sensor, i.e. thenonlinear relationship between the output the sensor and the loading (2) the responsecharacteristics of it are easily influenced by the environmental factors in the worksurrounds, e.g., humidity, vibration, electromagnetic interference and temperature,especially the temperature. Capacitance weighing sensor realizes mass measurementbased on capacitance sensor. The method of nonlinear calibration and temperaturecompensation should be established in order to reduce the interference of nonlinearcharacteristics and temperature factor for the purpose of to ensure the accuracy andstability of the results. Based on this background, the method to achieve nonlinearcalibration and temperature compensation has been studied, and the main contents areshown below.(1)The measurement principle, sensitivity, linearity and the effects of temperatureon capacitance weighing sensor have been analyzed. The expression of output ofcapacitance weighing sensor under the interference of the temperature has been given.The theoretical basis of temperature characteristic analysis has been established.(2)The mechanics analysis model of weighing sensor of weighing range200gand scale interval0.001g has been established through theory calculation andsimulation experiment. A theoretical basis and technical guidance for the design of theelastic element has been provided through the design and analysis of the elasticelement of the sensor based on the APDL parametric design language.(3)The influence of temperature on the output characteristic of capacitance weighingsensor has been analyzed based on the fact that the influence of temperature oncapacitance weighing sensor is mainly reflected by the elastic element, mainlyincluding the zero temperature characteristic, minimum static load temperaturecharacteristic, full range of temperature characteristic analysis and the influence onsensitivity and linearity of the weighing sensor of the temperature. The compensationmethod of temperature has been given. (4) Considering characteristics of the nonlinearity of the capacitance weighingsensor, i.e. the nonlinear relationship between the output of the sensor and the loading,an improved BP neural network was established to improve the nonlinear calibrationcapabilities. The effectiveness of the improved method has been confirmed throughthe comparison with the traditional gradient descent algorithm on rate of convergence,calculation accuracy and generalization capability.