Rectangular forming method of digital nuclear pulse

Abstract

The invention discloses a nuclear pulse rectangle forming method of a digital nuclear energy spectrum measurement system. Aiming at a digital nuclear energy spectrum measurement system using a scintillator detector, the method does not adopt nuclear pulse amplitude to represent incident particle energy, but solves the incident particle energy according to an area formula of discrete time signals, then takes an energy value as rectangular amplitude in a time domain, and can form a rectangular pulse with adjustable bottom width through simple operation. And the original nuclear pulse data are converted into rectangular waves formed by rectangles. Compared with a classical nuclear pulse forming method, the rectangle forming method does not need complex operation such as a z transformation method, a convolution method or a recursion method, is small in design difficulty, small in calculation amount, low in cost and easy to implement in a digital chip, and can exert the performance of the scintillator detector to the maximum extent.

Description

Technical Field

[0001] This invention relates to the field of nuclear technology and its applications, specifically to a key technology of digital nuclear energy spectrum measurement systems: nuclear pulse shaping method. Background Technology

[0002] The digital nuclear energy spectrum measurement system mainly consists of a nuclear signal detector, a preamplifier, a main amplifier, an analog-to-digital converter (ADC), a digital signal processing module, and host computer software for energy spectrum acquisition and processing. The digital signal processing module mainly performs functions such as baseline estimation and subtraction, noise filtering, accumulation rejection, digital shaping, and pulse amplitude analysis of nuclear pulse data.

[0003] Digital shaping of nuclear pulses is one of the key technologies in digital nuclear energy spectrum measurement systems. To improve the system's energy resolution or pulse throughput, the fully digitized nuclear pulses must be properly shaped to maximize the signal-to-noise ratio, optimize the pulse shape to accurately measure X-ray energy based on signal amplitude, reduce signal buildup at high count rates, and ensure measurement accuracy and reliability.

[0004] Based on the pulse waveform shaped by the digitized kernel pulse, there are generally several digitized pulse shaping methods (hereinafter referred to as "classical kernel pulse shaping methods"): Gaussian shaping, trapezoidal (triangle) shaping, linear shaping, rectangular shaping, etc. Classical kernel pulse shaping methods generally employ digital signal processing methods such as time-domain discrete convolution, z-transform domain system function method, and time-domain recursion method.

[0005] Although classical nuclear pulse shaping methods are an effective way to improve the energy resolution and pulse throughput of digital nuclear energy spectrum measurement systems, and improving these methods is a research direction for high-precision, high-performance digital nuclear energy spectrum measurement systems, these classical nuclear pulse shaping methods are difficult to design, computationally intensive, and costly, and are generally difficult to apply to commercial nuclear energy spectrum measurement systems.

[0006] Especially for nuclear energy spectrum measurement systems using scintillator detectors, these classical nuclear pulse shaping methods offer limited performance improvement. Compared to semiconductor detectors, scintillator detectors have poorer energy resolution. Semiconductor detectors, such as high-purity germanium detectors, can accurately distinguish between rays of different energies, achieving an energy resolution of 0.1%-0.2%. In contrast, even the high-performance lanthanum bromide detectors among scintillator detectors only achieve an energy resolution of around 3% for gamma rays, while the commonly used NaI(Tl) detector achieves an energy resolution of 5%-10% for gamma rays. Although these classical nuclear pulse shaping methods, which are complex to design, computationally intensive, and costly, can theoretically improve the energy resolution of digital nuclear energy spectrum measurement systems, the inherently poor energy resolution of scintillator detectors limits the improvement in system energy resolution even when these classical nuclear pulse shaping methods are used in digital nuclear energy spectrum measurement systems based on scintillator detectors. In fact, they may reduce the system's real-time response speed, or even prevent the implementation of these classical nuclear pulse shaping methods due to limitations in embedded hardware and software performance.

[0007] In digital nuclear energy spectrum measurement systems based on nuclear signal detectors with good energy resolution (such as semiconductor detectors), the rectangular shaping method of this invention cannot maximize detector performance. Classical nuclear pulse shaping methods typically employ computational methods such as time-domain discrete convolution, z-transform domain system function methods, and time-domain recursive methods. If the nuclear pulse energy is represented by the nuclear pulse area as in this invention, the design complexity, computational load, and cost would be significantly increased. This invention does not employ computational methods such as time-domain discrete convolution, z-transform domain system function methods, or time-domain recursive methods. Therefore, the rectangular shaping method provided by this invention, which represents the nuclear pulse energy by the nuclear pulse area, has lower design complexity, lower computational load, lower cost, and is easy to implement in digital chips such as FPGAs, while also maximizing the performance of scintillator detectors.

[0008] In summary, the rectangular forming method of the present invention has significant features and is compatible with the digital nuclear energy spectrum measurement system based on scintillator detectors. Summary of the Invention

[0009] From the circuit principle of a typical scintillator detector + RC feedback charge-sensitive preamplifier, although the nuclear pulse amplitude output by this circuit can reflect the energy of the incident particle, it is more accurate to characterize the energy by using the nuclear pulse area, which can avoid the problem of determining the peak value of the nuclear pulse. Moreover, compared with calculating the energy of the incident particle by the energy formula (although the energy formula is more accurate), the nuclear pulse area method is simpler and requires less calculation.

[0010] The energy of the incident particle is characterized by the nuclear pulse area, and the calculation formula is as follows: ns represents the start time of the original nuclear pulse, and ne represents the end time of the original nuclear pulse. This represents the value at each sampling point of the original nuclear pulse. This shaping method does not require transformation to the z-domain, complex convolution operations, or recursive solutions, making it simpler than the classical nuclear pulse shaping method. Obtaining the incident particle energy in this way can reduce the influence of high-frequency noise.

[0011] If the classical kernel pulse shaping methods based on convolution, recursion, or z-transform (Gaussian shaping, trapezoidal shaping, linear shaping, rectangular shaping, etc.) do not use the kernel pulse amplitude to characterize the energy of the incident particle, but instead calculate the energy of the incident particle according to the energy formula of the discrete-time signal, or use the kernel pulse area to characterize the energy of the incident particle, the design will be more complex, the computational load will be greater, and the implementation will be more difficult.

[0012] Whether using MATLAB simulation or Verilog HDL hardware description language programming to implement it in hardware systems such as FPGAs, the rectangular forming method of this invention can be implemented by following these steps.

[0013] Step 1: For digital nuclear energy spectrum measurement systems, the rectangular forming method of the present invention can be implemented after baseline estimation and subtraction, noise filtering and smoothing are performed on all raw nuclear pulse data.

[0014] The second step involves setting a nuclear pulse start threshold, Value_threshold_start, based on the typical nuclear pulse output from the scintillator detector and the RC feedback charge-sensitive preamplifier. If the sampled value of the nuclear signal exceeds this threshold at a certain moment, a nuclear pulse begins. Because the rising edge of the nuclear pulse is relatively steep, it is better to set the threshold at the start time slightly higher.

[0015] The third step involves setting a nuclear pulse termination threshold, Value_threshold_end, based on the typical nuclear pulse output from the scintillator detector and the RC feedback charge-sensitive preamplifier. If the sampled value of the nuclear signal falls below this threshold at a certain moment, the nuclear pulse terminates. Because the falling edge of the nuclear pulse is relatively slow, it is better to set the termination threshold slightly lower.

[0016] The fourth step is to iterate through all the raw nuclear pulse data. Add two arrays to record information about each nuclear pulse: still using arrays Record the value of each sampling point of the nuclear signal and add an array. Record the start time of each nuclear pulse and add an array. Record the termination time of each nuclear pulse. and They are of equal length, and their length corresponds to the number of nuclear pulses. Using the... The termination time minus the first At the starting moment, the first... The width of a nuclear pulse.

[0017] Step 5: Based on the typical core pulse output from the scintillator detector and the RC feedback charge-sensitive preamplifier, set the maximum width threshold `Width_max_threshold` and the typical pulse width value `Width_typical`. The maximum width threshold is greater than the typical pulse width value. Pulses with widths exceeding the maximum width threshold are usually tail-stacking pulses. If the pulse width exceeds the maximum width threshold but is less than twice the typical pulse width value, it is considered to have 2 staging pulses. If the pulse width exceeds twice the typical pulse width value, it is considered to have 3 staging pulses (the actual number of staging pulses may be more than 3, but it is only considered to have 3 due to the difficulty in accurate judgment). All staging pulses are removed, and the number of staging pulses is accumulated in the variable `elimination_counter`.

[0018] Step 6: Using a loop, apply the area formula for discrete-time signals. Calculate the energy of each nuclear pulse and store it in an array. Based on the typical nuclear pulse output from the scintillator detector and the RC feedback charge-sensitive preamplifier, a maximum energy threshold `energy_max_threshold` and a typical energy value `energy_typical` are set for the nuclear pulse. Pulses exceeding the maximum energy threshold are typically peak-stacking pulses. If the pulse energy exceeds the maximum energy threshold but is less than twice the typical energy value, it is considered to have two peak-stacking pulses. If the pulse energy exceeds twice the typical energy value, it is considered to have three peak-stacking pulses (the actual number of peak-stacking pulses may be more than three, but it is only considered to have three due to the difficulty in accurate judgment). All peak-stacking pulses are removed, and the number of peak-stacking pulses is accumulated in the variable `elimination_counter`.

[0019] Steps five and six above also eliminated those pulses that were both peak-stacking and tail-stacking.

[0020] Based on the statistical characteristics of nuclear pulse signals, it is known that as long as the measurement time is long enough and the number of detected nuclear pulses is sufficient, these eliminated accumulated pulses do not affect the final shape of the nuclear energy spectrum. The number of eliminated accumulated pulses is accumulated in the variable `elimination_counter`. If the nuclear energy spectrum measurement system only performs qualitative analysis, then the number of eliminated accumulated pulses is not needed. If the nuclear energy spectrum measurement system requires quantitative analysis, the energy array of the nuclear pulses can be calculated after calculating the energy of all nuclear pulses. Each element value is multiplied by this scaling factor. Proportionality coefficient = (array) (length + number of accumulated pulses elimination_counter) ÷ (array) (length).

[0021] Step 7: Based on the energy array of the nuclear pulse , Start time array Shaped rectangular wave.

[0022] To save storage space, in-situ computation is used here. First, all the original nuclear pulse data is... Each element is set to zero, and then the shaped rectangular wave data is stored in an array. middle.

[0023] Let the base width of the rectangle be... (suggestion It should be one-third of the typical pulse width (Width_typical), and preferably not exceed two-thirds. If it's too small, it's inconvenient to observe. (If it's too big, it will accumulate), the first The subroutine flow for shaping a nuclear pulse into a rectangle is as follows: No. The time coordinate of the first discrete point of the rectangle is: ; According to the time sequence, give the first A rectangle Assign values ​​to discrete points: ; ;……; .

[0024] At the base width of the rectangle With a fixed radius, the rectangular amplitude (i.e., the area of ​​the original nuclear pulse) characterizes the nuclear pulse energy. The base width can be adjusted as needed without causing accumulation. .

[0025] Based on the energy array of the nuclear pulse Once the rectangular wave is formed, nuclear energy spectrum analysis can be performed, and the corresponding energy spectrum diagram can be drawn. Attached Figure Description

[0026] Figure 1 The basic components of the digital nuclear energy spectrum measurement system of Embodiment 1 of the present invention.

[0027] Figure 2 A schematic diagram of the pulse waveforms before and after forming in Embodiment 1 of the present invention.

[0028] Figure 3 Schematic diagram of threshold settings for the start and end times of the nuclear pulse in Embodiment 1 of the present invention; Figure 3 In the middle: 1. Starting point coordinates (32, 1.0837), 2. Ending point coordinates (120, 0.2007).

[0029] Figure 4 Schematic diagram of the stacked pulse in Embodiment 1 of the present invention; Figure 4 In the middle: 3. Typical pulse, 4. Tail-stacking pulse, 5. Peak-stacking pulse, 6. Pulse with both peak-stacking and tail-stacking. Detailed Implementation

[0030] To better understand the present invention, the present invention will be further described below with reference to Embodiment 1 and the accompanying drawings. However, Embodiment 1 is only used to explain the present invention and does not constitute a limitation on the scope of protection of the present invention.

[0031] Example 1: The digital nuclear energy spectrum measurement system mainly consists of a nuclear signal detector (scintillator detector), a preamplifier, a main amplifier, an analog-to-digital converter (ADC), a digital signal processing module, and energy spectrum acquisition and processing software on a host computer. The basic components are as follows: Figure 1 The digital signal processing module primarily performs functions such as baseline estimation and subtraction, noise filtering, digital shaping, stacking rejection, and pulse amplitude analysis of nuclear pulse data. Digital shaping of nuclear pulses is one of the key technologies of a digital nuclear energy spectrum measurement system.

[0032] For digital nuclear energy spectrum measurement systems using scintillator detectors, this invention adopts the following shaping method: instead of characterizing the energy of the incident particle by the nuclear pulse amplitude, it uses a formula based on the discrete-time signal. The nuclear pulse area is used to characterize the energy of the incident particle. This represents the start time of the original nuclear pulse. This is the termination time of the original nuclear pulse. The value of each sampling point of the original nuclear pulse is used as the basis, and then the area value is used as the amplitude of the rectangle to form a rectangular pulse with an adjustable bottom width. In this way, all the original nuclear pulse data is shaped into a rectangular wave, such as... Figure 2 .

[0033] From the circuit principle of a typical scintillator detector + RC feedback charge-sensitive preamplifier, the nuclear pulse amplitude output by this circuit does reflect the energy of the incident particle. However, using the nuclear pulse area to characterize the energy is more accurate and can avoid the difficulty of determining the peak value of the nuclear pulse. Moreover, compared with calculating the energy of the incident particle through the energy formula (although the energy formula is more accurate), the nuclear pulse area method is simpler and requires less computation.

[0034] The rectangular shaping method based on the area of ​​the nuclear pulse provided by this invention does not require the z-transform method, complex convolution operations, or recursive solutions, making it simpler than the classical nuclear pulse shaping method. This method of obtaining the incident particle energy can reduce the influence of high-frequency noise.

[0035] Whether using MATLAB simulation or Verilog HDL hardware description language programming to implement it in hardware systems such as FPGAs, the rectangular forming method of this invention can be implemented by following these steps.

[0036] Step 1: For digital nuclear energy spectrum measurement systems, the rectangular forming method of the present invention can be implemented after baseline estimation and subtraction, noise filtering and smoothing are performed on all raw nuclear pulse data.

[0037] The second step involves setting a nuclear pulse start threshold, Value_threshold_start, based on the typical nuclear pulse output from the scintillator detector and the RC feedback charge-sensitive preamplifier. If the sampled value of the nuclear signal exceeds this threshold at a certain moment, a nuclear pulse begins. Figure 3 As can be seen from marker 1 "starting point coordinates (32, 1.0837)", because the rising edge of the nuclear pulse is relatively steep, it is better to set the threshold at the start time slightly higher. Figure 3 The example in the example can set the starting threshold. Between. For the sake of convenience and intuitiveness in describing this invention, Figure 3 The starting point coordinates are for reference only.

[0038] The third step involves setting a nuclear pulse termination threshold, Value_threshold_end, based on the typical nuclear pulse output from the scintillator detector and the RC feedback type charge-sensitive preamplifier. If the sampled value of the nuclear signal falls below this threshold at a certain moment, the nuclear pulse terminates. Figure 3 As can be seen from marker 2, "Termination point coordinates (120, 0.2007)," because the falling edge of the nuclear pulse is relatively slow, it is better to have a slightly smaller threshold at the termination time. Figure 3 The example in the example can set the termination threshold. Between. For the sake of convenience and intuitiveness in describing this invention, Figure 3 The coordinates of the endpoint are for reference only.

[0039] The fourth step is to iterate through all the raw nuclear pulse data. Add two arrays to record information about each nuclear pulse: still using arrays Record the value of each sampling point of the nuclear signal and add an array. Record the start time of each nuclear pulse and add an array. Record the termination time of each nuclear pulse. and They are of equal length, and their length corresponds to the number of nuclear pulses. Using the... The termination time minus the first At the starting moment, the first... The width of a nuclear pulse.

[0040] Fifth, based on the typical core pulse output from the scintillator detector and the RC feedback charge-sensitive preamplifier, set the maximum width threshold `Width_max_threshold` and the typical pulse width value `Width_typical`. The maximum width threshold is greater than the typical pulse width value. Pulses with widths exceeding the maximum width threshold are usually tail-stacking pulses. Figure 4 Mark 3 is "typical pulse". Figure 4 Mark 4 as "tail-stacking pulse". If the pulse width exceeds the maximum width threshold but is less than twice the typical pulse width, it is considered to have 2 staging pulses. If the pulse width exceeds twice the typical pulse width, it is considered to have 3 staging pulses (there may actually be more than 3 staging pulses, but due to the difficulty in accurate judgment, it is only considered to have 3). All staging pulses are removed, and the number of staging pulses is accumulated in the variable elimination_counter.

[0041] Step 6: Using a loop, apply the area formula for discrete-time signals. Calculate the energy of each nuclear pulse and store it in an array. Based on the typical nuclear pulse output by the scintillator detector and the RC feedback type charge-sensitive preamplifier, the maximum energy threshold energy_max_threshold and the typical energy value energy_typical of the nuclear pulse are set. Pulses exceeding the maximum energy threshold are usually peak-stacking pulses. Figure 4 Mark 5 as "peak-stacking pulse". If the pulse energy exceeds the maximum energy threshold but is less than twice the typical energy value, then there are considered to be 2 peak-stacking pulses. If the pulse energy exceeds twice the typical energy value, then there are considered to be 3 peak-stacking pulses (there may actually be more than 3 peak-stacking pulses, but since it is difficult to judge accurately, we only consider it to be 3). All peak-stacking pulses are removed, and the number of peak-stacking pulses is accumulated in the variable elimination_counter.

[0042] Steps five and six above also eliminated those pulses that were both peak-stacking and tail-stacking. Figure 4 The label 6 is "a pulse with both peak accumulation and tail accumulation".

[0043] Based on the statistical characteristics of nuclear pulse signals, it is known that as long as the measurement time is long enough and the number of detected nuclear pulses is sufficient, these eliminated accumulated pulses do not affect the final shape of the nuclear energy spectrum. The number of eliminated accumulated pulses is accumulated in the variable `elimination_counter`. If the nuclear energy spectrum measurement system only performs qualitative analysis, then the number of eliminated accumulated pulses is not needed. If the nuclear energy spectrum measurement system requires quantitative analysis, the energy array of the nuclear pulses can be calculated after calculating the energy of all nuclear pulses. Each element value is multiplied by a scaling factor. Proportionality coefficient (array) (length + number of accumulated pulses elimination_counter) ÷ (array) (length).

[0044] Step 7: Based on the energy array of the nuclear pulse , Start time array Shaped rectangular wave.

[0045] To save storage space, in-situ computation is used here. First, all the original nuclear pulse data is... Each element is set to zero, and then the shaped rectangular wave data is stored in an array. middle.

[0046] Let the base width of the rectangle be... (suggestion It should be one-third of the typical pulse width (Width_typical), and preferably not exceed two-thirds. If it's too small, it's inconvenient to observe. (If it's too big, it will accumulate), the first The subroutine flow for shaping a nuclear pulse into a rectangle is as follows: No. The time coordinate of the first discrete point of the rectangle is: ; According to the time sequence, give the first A rectangle Assign values ​​to discrete points: ; ;……; .

[0047] At the base width of the rectangle With a fixed radius, the rectangular amplitude (i.e., the area of ​​the original nuclear pulse) characterizes the nuclear pulse energy. The base width can be adjusted as needed without causing accumulation. .

[0048] Based on the energy array of the nuclear pulse Once the rectangular wave is formed, nuclear energy spectrum analysis can be performed, and the corresponding energy spectrum diagram can be drawn.

Claims

1. A rectangular shaping method for digital nuclear pulses, the principle of which is: during the nuclear pulse shaping process of a digital nuclear energy spectrum measurement system using a scintillator detector, the energy of the incident particle is not characterized by the nuclear pulse amplitude, but rather by the area formula of the discrete-time signal. The incident particle energy is obtained, where ns is the start time of the original nuclear pulse and ne is the end time of the original nuclear pulse. The value of each sampling point of the original nuclear pulse is used as the area value, and then in the time domain, this area value is used as the amplitude of the rectangle to form a rectangular pulse with an adjustable bottom width. In this way, all the original nuclear pulse data is formed into a rectangular wave composed of rectangles.

2. The rectangular shaping method for a digital nuclear pulse according to claim 1, characterized in that: This method is limited to digital nuclear energy spectrum measurement systems that use scintillator detectors.

3. The rectangular shaping method for a digital nuclear pulse according to claim 1, characterized in that: Before formation, the energy of the incident particle is not characterized by the amplitude of the nuclear pulse, but rather by the area formula of the discrete-time signal. The energy of the incident particle is obtained.

4. The rectangular shaping method for a digital nuclear pulse according to claim 1, characterized in that: Instead of using convolution, recursion, or z-transform methods during the shaping process, the area value of the original kernel pulse is used as the amplitude of the rectangle in the time domain to shape it into a rectangular pulse with an adjustable bottom width.

Images