A signal processing method of giant magneto-impedance sensor for distance measurement

Abstract

The invention discloses a method for processing the signal of a giant magneto-impedance sensor for distance measurement. The specific steps include: collecting the output signal of the giant magneto-impedance sensor when a magnetic target is in motion; The data processing fusion model of giant magneto-impedance sensor signal is established on the basis of the above; the wavelet energy characteristics of the output signal of giant magneto-impedance sensor collected in the direction of target motion are calculated, and it is used as the input sample training of the neural network to obtain the wavelet neural network model, and its relationship with The relationship between the target position; for the target at any position in the moving direction, first judge the position interval of the target through the wavelet neural network model, and then select the model corresponding to the position interval in the data processing fusion model for inversion calculation to obtain the final target Location. The processing method provided by the invention has universality, solves the multi-valued problem of the measurement result, improves the measurement accuracy and widens the measurement range.

Description

technical field

[0001] The invention relates to a processing method of giant magneto-impedance sensor signals, in particular to a processing method of distance measurement giant magneto-impedance sensor signals. Background technique

[0002] The giant magnetoimpedance sensor using the longitudinal excitation mode has an inherent "double peak" characteristic curve of the giant magnetoimpedance effect. In order to expand the measurement range of the giant magnetoimpedance sensor, it is necessary to solve the multi-valued problem of the measurement results. At present, regardless of the giant magneto-impedance sensor that has been produced or in the relevant literature, the hardware signal processing circuit of the giant magneto-impedance sensor can be divided into two categories: one is using a detection circuit, and the other is using a coherent demodulation circuit , but they only use the amplitude information of the output signal of the giant magneto-impedance sensor; in ad...

Examples

Embodiment 1

[0052] Embodiment one: for magnetic target A, the specific steps are as follows:

[0053] One, the giant magneto-impedance sensor obtains the distance signal when the magnetic target A moves according to a certain distance interval, and collects the output signal of the giant magneto-impedance sensor;

[0054] (1) Amplitude change rate

[0055] (1) Calculate the amplitude change rate of the signal to obtain the amplitude change rate curve;

[0056] Determine the amplitude change rate U of the magnetic target A at position i r (i):

[0057]

[0058] U(i) represents the amplitude of the output signal of the giant magnetoimpedance sensor when the corresponding target A is at position i, U(0) is the amplitude of the output signal of the giant magnetoimpedance sensor when there is only the ambient magnetic field, U r (i) is the amplitude change rate of the position i signal; with the target position as the abscissa, with the amplitude change rate as the ordinate, the amplitud...

Embodiment 2

[0134] Embodiment 2: Change target A to target B. Referring to the above steps, a brief description is as follows:

[0135] Compared with target A, the main difference between target B and target A is that the inherent magnetism has changed, and the influence on the giant magneto-impedance sensor is different at the same distance, which will lead to the change of the demarcation point and coefficient in the data processing model.

[0136] Collect the output signal of the giant magneto-impedance sensor when the target B is at different positions in a certain direction of motion, extract the amplitude and energy characteristics of the output signal, and obtain its amplitude change rate curve (see Figure 5 ), the energy change rate curve; find the position d corresponding to the peak value of the amplitude change rate curve 1 =10cm, the position d corresponding to the intersection point of the error curve of the energy change rate model and the amplitude change rate model 2 =2...

Embodiment 3

[0143] Embodiment 3: For the target A in Embodiment 1, replace the giant magneto-impedance sensor. Referring to the above steps, a brief description is as follows:

[0144] Different giant magneto-impedance sensors, even the same type of giant magneto-impedance sensors, have different characteristics of sensitive materials, and the degree of natural giant magneto-impedance effect will also be different. The amplitude and The energy change rate curve will also change, but it is only a change in quantity, and the basic change law remains unchanged.

[0145] Collect the output signal of the giant magneto-impedance sensor when the target A is at different positions in a certain direction of motion, refer to the above steps to obtain its amplitude and energy change rate curve, and calculate the dividing point d of the distance between the two targets 1 = 7cm and d 2 = 30cm (it can be seen that for a specific target, even if the giant magneto-impedance sensor is replaced, its dema...

technical field

[0001] The invention relates to a processing method of giant magneto-impedance sensor signals, in particular to a processing method of distance measurement giant magneto-impedance sensor signals. Background technique

[0002] The giant magnetoimpedance sensor using the longitudinal excitation mode has an inherent "double peak" characteristic curve of the giant magnetoimpedance effect. In order to expand the measurement range of the giant magnetoimpedance sensor, it is necessary to solve the multi-valued problem of the measurement results. At present, regardless of the giant magneto-impedance sensor that has been produced or in the relevant literature, the hardware signal processing circuit of the giant magneto-impedance sensor can be divided into two categories: one is using a detection circuit, and the other is using a coherent demodulation circuit , but they only use the amplitude information of the output signal of the giant magneto-impedance sensor; in ad...

Examples

Embodiment 1

[0052] Embodiment one: for magnetic target A, the specific steps are as follows:

[0053] One, the giant magneto-impedance sensor obtains the distance signal when the magnetic target A moves according to a certain distance interval, and collects the output signal of the giant magneto-impedance sensor;

[0054] (1) Amplitude change rate

[0055] (1) Calculate the amplitude change rate of the signal to obtain the amplitude change rate curve;

[0056] Determine the amplitude change rate U of the magnetic target A at position i r (i):

[0057]

[0058] U(i) represents the amplitude of the output signal of the giant magnetoimpedance sensor when the corresponding target A is at position i, U(0) is the amplitude of the output signal of the giant magnetoimpedance sensor when there is only the ambient magnetic field, U r (i) is the amplitude change rate of the position i signal; with the target position as the abscissa, with the amplitude change rate as the ordinate, the amplitud...

Embodiment 2

[0134] Embodiment 2: Change target A to target B. Referring to the above steps, a brief description is as follows:

[0135] Compared with target A, the main difference between target B and target A is that the inherent magnetism has changed, and the influence on the giant magneto-impedance sensor is different at the same distance, which will lead to the change of the demarcation point and coefficient in the data processing model.

[0136] Collect the output signal of the giant magneto-impedance sensor when the target B is at different positions in a certain direction of motion, extract the amplitude and energy characteristics of the output signal, and obtain its amplitude change rate curve (see Figure 5 ), the energy change rate curve; find the position d corresponding to the peak value of the amplitude change rate curve 1 =10cm, the position d corresponding to the intersection point of the error curve of the energy change rate model and the amplitude change rate model 2 =2...

Embodiment 3

[0143] Embodiment 3: For the target A in Embodiment 1, replace the giant magneto-impedance sensor. Referring to the above steps, a brief description is as follows:

[0144] Different giant magneto-impedance sensors, even the same type of giant magneto-impedance sensors, have different characteristics of sensitive materials, and the degree of natural giant magneto-impedance effect will also be different. The amplitude and The energy change rate curve will also change, but it is only a change in quantity, and the basic change law remains unchanged.

[0145] Collect the output signal of the giant magneto-impedance sensor when the target A is at different positions in a certain direction of motion, refer to the above steps to obtain its amplitude and energy change rate curve, and calculate the dividing point d of the distance between the two targets 1 = 7cm and d 2 = 30cm (it can be seen that for a specific target, even if the giant magneto-impedance sensor is replaced, its dema...

Images