Gamma energy spectrometry spectral drift and integral nonlinear correction method

Abstract

The invention discloses a gamma energy spectrometry spectral drift and integral nonlinear correction method, which involves a gamma-ray detector, an amplifying and shaping circuit, an ADC circuit anda digital pulse amplitude analyzer. The spectral drift and integral nonlinear correction method is implemented in a SOPC (system-on-a-programmable-chip) constructed on an FPGA (field programmable gatearray) and includes that the detector outputs voltage pulse, and pulse signals are amplified and shaped, then digitalized through the ADC circuit, analytically processed through the digital pulse amplitude analyzer and subjected to network input data conversion processing, spectral drift and integral nonlinear correction neural network and network output data conversion processing to obtain spectral data of gamma energy spectrometry. The method has advantages that the hardware circuit design is simplified, uncertainty of measurement results due to changes of a measurement environment and hardware system performances is reduced, and errors in spectral drift and integral nonlinear measurement are avoided. By adoption of the same neutral network for spectral drift and integral nonlinear correction, correction result conflicts caused by step-by-step correction can be avoided, and comprehensive spectral drift and integral nonlinear correction is realized.

Description

technical field [0001] The invention relates to measuring equipment in geological exploration, in particular to a method for correcting spectrum drift and integral nonlinearity by using gamma energy spectrum. Background technique [0002] Gamma energy spectrum measurement usually adopts the detector structure of NaI(Tl) scintillation crystal + photomultiplier tube + preamplifier circuit. Secondary electrons are generated to ionize or excite the scintillator molecules. When the NaI(Tl) scintillation crystal emits a large number of photons during de-excitation, after the photons gather to the photocathode of the photomultiplier tube, photoelectrons are shot on the photocathode through the photoelectric effect, and the photoelectrons are in the photoelectric The multiplier is gradually multiplied in the multiplier, and the electrons multiplied by each multiplier are collected at the anode, and the detector output voltage pulse is formed through the pre-amplification circuit. M...

AI Technical Summary

Problems solved by technology

[0014] Gamma energy spectrum measurement mainly adopts the correction method of additional reference source spectrum drift, LED spectrum drift correction method and the characteristic peak spectrum drift correction method of the nuclide to be measured, all of which assume that the spectral line drift is a linear drift, which is based on a reference peak in the instrument spectrum. Baseline, the gain of the pulse amplification is feedback-controlled through the drift of the reference peak-to-peak position, and the amplitude of all gamma-ray corresponding pulse signals in the entire measurement energy range is adjusted with the same gain value, so as to correct the drift of the spectral line in the entire energy range ; and the actually measured γ-ray energy has a nonlinear relationship with the peak position of the characteristic peak, so the spectral line drift is not a linear drift, especially when the spectral drift correction reference peak is far away from the position of the characteristic peak of the nuclide to be measured and the spectral drift There are nonlinear effects, and the effect of spectral drift correction is not ideal; and the above method also increases the complexity of the hardware circuit part

[0015] The software segmented calibration spectral drift correction method is to find the characteristic peaks of multiple target nuclides on the instrument spectrum, and use the characteristic peaks of multiple target nuclides in the instrument spectrum to re-scale the energy-track location, using every two adjacent features The peak-to-peak position is linearly scaled, that is, the entire measured energy segment is divided into multiple segments to be linearly scaled, and the spectral lines are rewritten according to the scale results; Fully considering the influence of non-linear factors between energy and track site in each narrow energy segment after segmentation, there are still errors and large differences in the determination of energy window counting and qualitative analysis of nuclide identification in the narrow energy segment of the assumed linear relationship. Certainty; and rewriting the instrument spectrum based on the recalibration results is a non-real-time correction, which is likely to cause spectral line distortion and expand statistical fluctuation errors, without using full-spectrum data analysis and research

[0016] Due to the particularity of gamma energy spectrum measurement, non-linear corrections such as look-up table method, inverse function method and artificial neural network used in other measurement fields (such as pressure and temperature) are not suitable for gamma energy spectrum measurement.

The gamma energy spectrum measurement input is the detected gamma rays with different energies, and the output is multi-channel gamma energy spectral data. The detected gamma-ray energy is determined by analyzing the characteristic peak positions of the multi-channel spectral data, so it is impossible to establish a look-up table A large number of known measurement databases required by the method; the network structure, training sample selection and network learning training methods used in artificial neural network nonlinear correction methods in other measurement fields are not suitable for gamma energy spectrum measurement; at the same time, gamma energy spectrum measurement system The non-linearity comes from many factors, the non-linear mapping is more complex, and the non-linear characteristics cannot be described by one or several functions, so the inverse function method of the commonly used polynomial fitting function construction is not suitable for the gamma energy spectrum Integral Nonlinearity Correction in Measurements

Images