ADC differential nonlinear software correction method of multichannel analyzer

Abstract

The invention relates to an ADC differential nonlinear software correction method for a multichannel analyzer, and the method comprises the steps: S1, selecting a radioactive source for measurement, and adjusting a high voltage to enable a measured spectral line to cover all channel addresses of the multichannel analyzer; a relatively smooth spectral line is obtained by measuring for a certain time length; s2, performing smooth filtering processing on the original spectrum data by using spectrum data analysis software, wherein the phase shift of the selected smooth filtering algorithm is zero; s3, importing the original spectrum data and the filtered spectrum data into Excel software, obtaining a 2 sigma value of each channel of the filtered spectrum data, and screening out channel addresses exceeding the 2 sigma value in the original spectrum data; and S4, obtaining the ratio of the channel value of the filtered spectrum data to the channel value corresponding to the original spectrum data exceeding the 2 sigma value, the ratio being a correction coefficient, and writing the correction coefficient into a multichannel analyzer to realize correction. According to the invention, the correction can be realized without the assistance of additional instruments, and particularly, the assembled energy spectrum probe can be corrected without disassembly and assembly.

Description

Technical Field

[0001] The present invention relates to the field of radioactive measurement, and particularly to an ADC differential nonlinearity software correction method for a multi-channel analyzer. Background Art

[0002] The multi-channel pulse amplitude analyzer is an important instrument in nuclear radiation analysis. Its main function is to convert the pulse amplitude into a digital quantity, and this crucial step is completed by the ADC (analog-to-digital conversion circuit). The ADC divides the maximum pulse amplitude that can be analyzed into several voltage intervals. The amplitude of the input pulse falling into which interval is recorded in that channel. However, these intervals are not uniform, which will cause data points to deviate along the counting axis during energy spectrum measurement. We use differential nonlinearity to measure this index of channel width inconsistency. Generally, the method to improve differential nonlinearity is to average adjacent several channels and combine them into one channel. However, this approach requires an ADC with a higher number of bits. For example, for a 1024-channel multi-channel analyzer, to improve differential nonlinearity by averaging adjacent 4 channels and combining them into one channel, a 12-bit ADC is required to achieve, which increases the cost. Summary of the Invention

[0003] The purpose of the present invention is to provide an ADC differential nonlinearity software correction method for a multi-channel analyzer, which can achieve correction without the assistance of additional instruments, especially for an assembled energy spectrum probe, which can be corrected without disassembly.

[0004] To achieve the above purpose, the present invention provides the following technical solution: An ADC differential nonlinearity software correction method for a multi-channel analyzer, the method comprising the following steps:

[0005] Step S1: Select a radiation source for measurement, adjust the high voltage so that the measured spectrum line can cover all channel addresses of the multi-channel analyzer; after measuring for a certain period of time, obtain a relatively smooth spectrum line;

[0006] Step S2: Use spectrum data analysis software to perform smoothing filtering on the original spectrum data, and the phase shift of the selected smoothing filtering algorithm should be zero;

[0007] Step S3: Import the original spectrum data and the filtered spectrum data into Excel software, obtain the 2σ value of each channel of the filtered spectrum data, and screen out the channel addresses in the original spectrum data that exceed the 2σ value;

[0008] Step S4: Obtain the ratio of the channel value of the filtered spectrum data to the corresponding channel value of the original spectrum data that exceeds the 2σ value. This ratio is the correction coefficient, and write the correction coefficient into the multi-channel analyzer to achieve correction.

[0009] Further, the smoothing filtering algorithm in step S2 can be selected as an FIR filter, and the order is 20.

[0010] Further, the calculation formula for obtaining the 2σ value of each channel of the filtered spectral data in step S3 is further: 2σ = 2 * SQRT(spectral data).

[0011] Further, in step S4, writing the correction coefficient into the multi-channel analyzer to achieve correction is further to connect the computer and the multi-channel analyzer using a USB to serial cable, copy the correction data of each channel to the serial port software and send it to the multi-channel analyzer, so that each channel output data output by the multi-channel analyzer is multiplied by the correction coefficient to achieve correction.

[0012] Advantages of the present invention: Compared with the existing technology, for a 1024-channel multi-channel analyzer, to improve differential non-linearity by averaging adjacent 4 channels into one channel requires a 12-bit ADC, which increases the cost. However, using the method of the present invention only requires a 10-bit ADC to achieve the same effect. In the traditional software correction method for ADC differential non-linearity, a slip pulse generator is used to count for each channel to determine the correction coefficient of each channel. The software correction method of the present invention does not require the assistance of additional instruments. In particular, for an assembled energy spectrum probe, correction can be achieved without disassembly. The method of the present invention has obvious effects in actual product use. Especially when using the ADC built in some MCUs, its differential non-linearity is very poor, and there will be sudden changes in the channel values of many channels in the measured spectral line, making the spectral line look discontinuous. Even averaging adjacent several channels into one channel is not ideal. After using the method of the present invention for correction, this situation can be well improved. Description of the Drawings

[0013] Figure 1 is a schematic diagram of the measurement environment;

[0014] Figure 2 is a schematic diagram of the original spectral line collected;

[0015] Figure 3 is a schematic diagram of the spectral line after filtering the original spectral line;

[0016] Figure 4 is the spectral line re-collected by the multi-channel analyzer after writing the correction coefficient. Detailed Embodiments

[0017] The present invention will be further described below with reference to the drawings.

[0018] Please refer to Figures 1 to 4 , the present invention provides an embodiment: An ADC differential non-linearity software correction method for a multi-channel analyzer, the method includes the following steps:

[0019] Step S1: Select a radiation source for measurement, adjust the high voltage so that the measured spectrum can cover all the channel addresses of the multi-channel analyzer; after measuring for a certain period of time, obtain a relatively smooth spectrum;

[0020] Step S2: Use spectrum analysis software to perform smoothing filtering on the original spectrum data, and the phase shift of the selected smoothing filtering algorithm should be zero;

[0021] Step S3: Import the original spectrum data and the filtered spectrum data into Excel software, obtain the 2σ value of each channel of the filtered spectrum data, and screen out the channel addresses in the original spectrum data that exceed the 2σ value;

[0022] Step S4: Obtain the ratio of the channel value of the filtered spectrum data to the channel value of the corresponding channel of the original spectrum data that exceeds the 2σ value. This ratio is the correction coefficient, and write the correction coefficient into the multi-channel analyzer to achieve correction.

[0023] The following further illustrates the present invention with a specific embodiment:

[0024] A software correction method for ADC differential nonlinearity of a multi-channel analyzer. When measuring, a radiation source with as large a ray energy as possible should be selected, so that the high voltage does not need to be adjusted too high to ensure that each channel has a relatively large number of count values. Generally, it is more appropriate to select 60Co as the measurement source. There is no requirement for the energy resolution of the selected scintillator, and generally NaI can be used. The computer needs to install software that can export spectrum data and spectrum analysis software. The construction of the measurement environment is as Figure 1 shown. Specifically:

[0025] 1. Select a radiation source for measurement, adjust the high voltage so that the measured spectrum can cover all the channel addresses of the multi-channel analyzer; after measuring for a certain period of time, obtain a relatively smooth spectrum;

[0026] 2. Use spectrum analysis software to perform smoothing filtering on the original spectrum data, and the phase shift of the selected smoothing filtering algorithm should be zero;

[0027] The smoothing filtering algorithm can be selected as an FIR filter with an order of 20.

[0028] 3. Import the original spectrum data and the filtered spectrum data into Excel software, obtain the 2σ value of each channel of the filtered spectrum data, and screen out the channel addresses in the original spectrum data that exceed the 2σ value; 2σ is the square root multiplied by 2.

[0029] The calculation formula for obtaining the 2σ value of each channel of the filtered spectrum data is further: 2σ = 2 * SQRT(spectrum data). 2 * SQRT(spectrum data) means multiplying the square root of the spectrum data by 2, and the spectrum data is based on the actually collected data.

[0030] 4. Obtain the ratio of the channel value of the filtered spectral data to the channel value of the original spectral data exceeding the 2σ value. This ratio is the correction coefficient, and write the correction coefficient into the multichannel analyzer to achieve correction.

[0031] Writing the correction coefficient into the multichannel analyzer to achieve correction is further to connect the computer and the multichannel analyzer using a USB-to-serial cable, copy the correction data of each channel to the serial port software and send it to the multichannel analyzer, so that the output data of each channel output by the multichannel analyzer is multiplied by the correction coefficient to achieve correction.

[0032] In summary, the method of the present invention has obvious effects in the actual product use. Especially when using the ADC built in some MCUs, its differential nonlinearity is very poor, and there will be sudden changes in the channel values of many channels in the measured spectral line, making the spectral line look discontinuous. Even if the adjacent several channels are averaged into one channel, it is not ideal. After using the method of the present invention for correction, this situation can be well improved. As Figure 1 shown, Figure 1 is the original spectral line collected. Due to the nonlinearity of the ADC, there are many "prominent points" in the lower spectral line. These "prominent points" are essentially different from the "spikes" caused by random fluctuations on the spectral line. The "spikes" caused by random fluctuations can be made smooth by increasing the acquisition time, while these "prominent points" will not decrease with the increase of the acquisition time; as Figure 2 shown, Figure 2 is the spectral line after filtering the original spectral line, and it can be seen that the spectral line becomes smooth; as Figure 3 shown, Figure 3 is the spectral line re-collected by the multichannel analyzer that has written the correction coefficient, and there are basically no "prominent points" on the spectral line.

[0033] The above are only the preferred embodiments of the present invention, and should not be construed as limitations to this application. All equivalent changes and modifications made according to the scope of the patent application of the present invention shall fall within the scope covered by the present invention.

Claims

1. A software correction method for ADC differential nonlinearity of a multi-channel analyzer, characterized in that: The method comprises the following steps: Step S1, select a radioactive source for measurement, adjust the high voltage so that the measured spectrum can cover all the channels of the multi-channel analyzer; after measuring for a certain period of time, obtain a relatively smooth spectrum; Step S2, using spectrum data analysis software to perform smoothing filtering on the original spectrum data, and the phase shift of the selected smoothing filtering algorithm must be zero; Step S3, importing the original spectrum data and the filtered spectrum data into Excel software, obtaining the 2σ value of each channel of the filtered spectrum data and screening out the channel addresses exceeding the 2σ value in the original spectrum data; Step S4, obtaining the ratio of the channel value of the filtered spectrum data to the channel value corresponding to the original spectrum data exceeding the 2σ value, the ratio is the correction coefficient, and the correction coefficient is written into the multi-channel analyzer to implement correction.

2. The method for correcting ADC differential nonlinearity of a multi-channel analyzer according to claim 1, characterized in that: The smoothing filter algorithm of step S2 can be performed using an FIR filter with an order of 20.

3. The method for correcting ADC differential nonlinearity of a multi-channel analyzer according to claim 1, characterized in that: The calculation formula for obtaining the 2σ value of each channel of the filtered spectrum data in step S3 is further: 2σ=2*SQRT(spectrum data).

4. The method for correcting ADC differential nonlinearity of a multi-channel analyzer by software according to claim 1, characterized in that: Writing the correction coefficient into the multi-channel analyzer to achieve correction in step S4 is further performed by connecting the computer and the multi-channel analyzer using a USB to serial port cable, copying the correction data of each channel to the serial port software and sending it to the multi-channel analyzer, so that each output data of the multi-channel analyzer is multiplied by the correction coefficient to achieve correction.

Images