Search

Search Results

• SELECTING FEATURES OF THE MODEL TRANSFORMATION CHARACTERISTICS FOR AN INTELLIGENT PHYSICAL QUANTITY SENSOR

  S. I. Klevtsov

  2021-11-14

  Abstract ▼

  The paper discusses the issues of choosing the type and parameters of the model of the transformation
  characteristic of an intelligent sensor of physical quantities using the example of a pressure
  sensor. The transformation characteristic of an intelligent sensor is a mathematical, algorithmic
  and software for calculating a physical quantity based on electrical signals that come from the measuring
  channels of the sensor. The model of the conversion characteristic should be adapted to the
  configuration of the conversion function of the sensor's sensitive element and the behavior of this
  function under the influence of external destabilizing factors. The paper considers various models of
  the conversion characteristics, identifies the features of their application, advantages and disadvantages,
  attainable levels of approximation error of the real characteristic, which affect the final
  measurement accuracy of the smart sensor. Smart sensors are used for measuring physical quantities
  in various technical systems and the requirements for measurement accuracy in real-life tasks are
  different. The measurement accuracy is largely determined by the degree of approximation of the real
  characteristics of the sensor by its mathematical model. The more complex the model, the more difficult
  it is to implement in the sensor, and the higher the measurement cost. Therefore, it is important
  to control the conversion characteristic approximation error in order to use the sensor efficiently. To
  control the approximation error of the transformation characteristic of an intelligent pressure sensor,
  it is proposed to use the method of multi-segment spatial approximation, and use models of linear or
  nonlinear spatial elements as segments. The basic mathematical expressions, the error control
  scheme are determined. The results of modeling are presented, which show the possibility and advantages
  of using the method for the formation of spatial models of the transformation characteristics,
  which are adaptive to changes in the real transformation function of the sensor, take into account
  the influence of external factors on the measurement results. In addition, the method allows you
  to modify the current spatial approximation model by changing the types of local spatial elements
  and, thus, to control the measurement error