Amplitude analysis method and system suitable for X fluorescence waste battery classification system

Abstract

The invention relates to the field of waste battery classification methods, in particular to an amplitude analysis method and system suitable for an X fluorescence waste battery classification system.The system comprises a signal collecting module and a signal processing module, and the collecting module collects signals and converts the signals into digital signals; the signal processing module is used for sequentially carrying out trapezoid forming operation, peak value extraction operation and energy spectrum drawing operation on the digital signals; the method comprises the following steps: carrying out trapezoidal forming operation on a digital signal obtained by acquisition and conversion; carrying out peak value extraction operation on formed pulses obtained by trapezoidal forming operation to extract features, capturing the peak value amplitude of each pulse, and carrying out hardware acceleration through a finite-state machine architecture by a single-threshold peak value extraction algorithm; and carrying out classified statistics and accumulation on the captured peak amplitudes to form an energy spectrum distribution diagram. According to the method, the anti-interference capability and the detection reliability are remarkably improved, and the method is particularly suitable for a digital multi-channel pulse amplitude analysis scene requiring high-precision and high-stability peak value extraction in a strong noise environment.

Description

Technical Field

[0001] This invention relates to the field of waste battery classification methods, and specifically to an amplitude analysis method and system applicable to X-ray fluorescence waste battery classification systems. Background Technology

[0002] In the field of waste battery resource utilization, rapid and accurate identification of battery types and components is crucial for achieving efficient recycling and reuse. Currently, component analysis methods based on X-ray fluorescence detection technology are widely used in the non-destructive sorting process of waste batteries. X-ray fluorescence detection technology uses X-ray fluorescence energy dispersive spectroscopy analysis to achieve qualitative and quantitative identification of valuable metal elements in batteries, thereby completing automated classification.

[0003] In existing X-ray fluorescence detection devices, energy spectrum acquisition largely relies on analog multichannel pulse amplitude analyzers (analog MCAs), which typically require a dedicated amplification and shaping circuit to process the pulse signals output by the detector. Because X-ray photon events occur randomly, signals may overlap in time, leading to pulse accumulation, which severely affects the shape of the energy spectrum and the accuracy of element identification. Traditional systems often employ pulse accumulation rejection circuits to block the signal when accumulation is detected, preventing it from entering the MCA analysis stage. While this method can suppress the interference of accumulation on the energy spectrum to some extent, it also causes a significant loss in the effective count rate, reducing detection efficiency.

[0004] Especially when dealing with waste batteries with complex composition and large differences in element content, if the detection sensitivity of low-content elements is to be improved, the intensity of the radiation source is usually increased to increase the counting rate. However, this will exacerbate pulse accumulation, leading to an increase in system dead time. The actual increase in counting effect is limited, and it also brings radiation protection and management burden. Summary of the Invention

[0005] The present invention aims to provide an amplitude analysis method and system suitable for X-ray fluorescence waste battery classification systems, in order to solve the problems of insufficient detection reliability, weak anti-interference ability, limited energy spectrum counting rate, and low stability under complex environments.

[0006] According to one aspect of the present invention, an amplitude analysis system suitable for X-ray fluorescence waste battery classification system is provided, including a signal acquisition module and a signal processing module; The signal acquisition module is used to acquire the output signal of the X-ray detector after preprocessing by the differentiating circuit and the preamplifier circuit, and to convert the output signal, which is an analog signal, into a digital signal. The signal processing module is used to sequentially perform trapezoidal shaping, peak extraction, and energy spectrum plotting operations on the digital signal output by the signal acquisition module. The signal processing module performs peak extraction using a single-threshold peak extraction algorithm, which is hardware-accelerated using a finite state machine architecture. The finite state machine architecture includes a minimum search state, a maximum search state, and a peak output state.

[0007] The beneficial effects of this plan are: By employing a finite state machine state transition mechanism during data processing, it ensures that only one peak result is output in each complete signal processing cycle. This effectively prevents false detections and multiple peak outputs caused by signal jitter or noise interference, significantly improving the system's anti-interference capability and detection reliability. It is particularly suitable for digital multichannel pulse amplitude analysis scenarios that require high-precision and high-stability peak extraction in noisy environments.

[0008] Furthermore, it also includes a clock module and a communication interface module. The clock module is used to provide a synchronization clock signal for each module, and the communication interface module is used to transmit the processed energy spectrum data to the host computer. The communication interface module adopts USB or Ethernet protocol.

[0009] Furthermore, the signal processing module is configured with a programmable threshold parameter register, which is used to dynamically adjust the preset threshold voltage and counting threshold in the single threshold peak extraction algorithm.

[0010] The beneficial effect is that by setting the threshold parameter register, it is possible to adapt to input pulse signals of different intensities and shapes.

[0011] Furthermore, the signal processing module employs configurable forming parameters during the trapezoidal forming operation, including forming time and attenuation constant.

[0012] The beneficial effects are: by performing trapezoidal forming operations with configurable forming parameters, signal noise suppression performance and pulse throughput can be optimized.

[0013] According to another aspect of the present invention, an amplitude analysis method suitable for X-ray fluorescence waste battery classification systems is provided, comprising the following steps: S100 acquires the output signal of the X-ray detector after preprocessing by the differentiating circuit and the preamplifier circuit, and converts it into a digital signal. S200, first perform a trapezoidal forming operation on the digital signal of S100, and perform the trapezoidal forming operation with a configurable forming time and attenuation constant; S300, the peak extraction operation is performed on the shaped pulse obtained from S200 to extract features. Peak extraction is performed by a single threshold peak extraction algorithm to capture the peak amplitude of each pulse. The single threshold peak extraction algorithm is hardware accelerated by a finite state machine architecture, which includes a minimum value search state, a maximum value search state, and a peak output state. S400 classifies, statistically analyzes, and accumulates the peak amplitudes captured by S300 to form an energy spectrum distribution map.

[0014] The beneficial effects of this plan are: By applying a single-threshold finite state machine peak extraction algorithm to a waste battery X-ray fluorescence classification system, the system achieves rapid analysis and accurate identification of complex multi-element energy spectra such as nickel, cobalt, and manganese through a state transition mechanism. This solves the industry problem of inaccurate feature peak extraction in mixed energy spectra of battery materials using traditional methods. Strict state transition conditions ensure that only one peak is output per signal cycle, effectively avoiding false detections and repeated triggering, achieving true real-time energy spectrum acquisition and processing, and improving noise suppression and anti-interference performance.

[0015] Furthermore, in S300, when the finite state machine architecture is in the minimum value search state, it continuously compares the current sampled value with a preset threshold voltage. If the current sampled value is lower than the preset threshold voltage, the minimum value register is dynamically updated to record the minimum level of the signal. If the current sampled value exceeds the preset threshold voltage, it immediately switches from the minimum value search state to the maximum value search state. After entering the maximum value search state, an internal counter is started to count, and the maximum value register is compared and updated every clock cycle to capture the highest level of the signal. In this state, if the current sampled value is detected to be lower than the preset threshold voltage again, it is immediately reset and returns to the minimum value search state. The transition from the maximum value search state to the peak output state is only allowed when the following two conditions are met simultaneously: Condition 1, the value of the internal counter exceeds the preset time count value, indicating that the pulse width has reached the requirement; Condition 2, the current sampled value has fallen back below the preset threshold voltage, indicating that the pulse signal has entered the falling phase. In the peak output state, the difference between the values ​​stored in the maximum value register and the minimum value register is calculated, and this difference is taken as the peak result of the current acquisition pulse. At the same time, a write enable signal is generated to trigger the data write operation. After the peak output is completed, the internal counter is automatically cleared and the maximum value register and the minimum value register are reset, thereby ending the current peak finding cycle and preparing to start the processing of the next signal cycle.

[0016] The beneficial effects are: through the transformation of various states, it is possible to quickly analyze and accurately identify the complex multi-element energy spectrum of waste batteries, thereby improving the anti-interference ability.

[0017] Furthermore, in step S400, the peak amplitude and the energy spectrum address are linearly fitted using a preset formula, which is expressed as follows: y = 400.0476x - 44.7143; Where x is the peak amplitude and y is the energy spectrum address. Attached Figure Description

[0018] Figure 1 This is a schematic diagram of the amplitude analysis system of the X-ray fluorescence waste battery classification system according to Embodiment 1 of the present invention; Figure 2 This is a linear relationship diagram between pulse amplitude and energy spectrum address in the amplitude analysis method of the X-ray fluorescence waste battery classification system according to Embodiment 2 of the present invention; Figure 3 This is a waveform diagram of the signal generator in the amplitude analysis method of the X-ray fluorescence waste battery classification system according to Embodiment 2 of the present invention. Detailed Implementation

[0019] The following detailed description provides further details on specific implementation methods.

[0020] Example 1 An amplitude analysis system suitable for X-ray fluorescence waste battery sorting systems employs a fully digital multi-channel pulse amplitude analysis architecture, realizing a complete processing flow from signal acquisition to energy spectrum generation on a single board circuit. Figure 1 As shown: It includes a detector, a signal acquisition module, a signal processing module, a clock module, and a communication interface module.

[0021] The detector is used to detect X-ray fluorescence photons generated by the stimulation of waste battery samples and converts them into corresponding analog electrical signals through a photomultiplier tube. The analog electrical signals are first pre-conditioned by existing differentiating circuits and preamplifier circuits to eliminate baseline drift and enhance signal amplitude. The detector uses an existing sodium iodide detector.

[0022] The signal acquisition module is used to acquire the output signal of the X-ray detector after preprocessing by the differentiating and preamplifying circuits, and converts the analog output signal into a digital signal. Based on actual requirements, the signal acquisition module uses an existing high-speed ADC module. This high-speed ADC module employs a 50MHz sampling frequency analog-to-digital converter, responsible for acquiring the X-ray detector output signal after preprocessing by the differentiating and preamplifying circuits and converting it into a high-precision digital signal. A high-speed ADC module, such as the existing AD9226 analog-to-digital converter, supports a maximum sampling frequency of 65MHz, but is actually configured with a 50MHz sampling frequency to balance signal fidelity and system resource consumption.

[0023] The signal processing module sequentially performs trapezoidal shaping, peak extraction, and energy spectrum plotting operations on the digital signal output from the signal acquisition module. During the trapezoidal shaping operation, the signal processing module employs configurable shaping parameters, including shaping time and attenuation constant. Peak extraction is performed using a single-threshold peak extraction algorithm, which is hardware-accelerated using a finite state machine architecture. This finite state machine architecture includes a minimum search state, a maximum search state, and a peak output state.

[0024] The signal processing module uses an existing FPGA module, such as the EP4CE10 model, including a laddering unit for laddering operations, a peak extraction unit for peak extraction, an energy spectrum plotting unit for energy spectrum plotting, and a data transmission unit for transmitting data to the host computer. The laddering unit performs digital filtering and waveform shaping on the raw digital signal output from the ADC to suppress noise and improve pulse shape, providing a standardized signal waveform for subsequent peak extraction. The peak extraction unit employs a single-threshold peak extraction algorithm based on a finite state machine, accurately capturing the peak amplitude of each pulse through a preset threshold voltage judgment and state transition mechanism. The energy spectrum plotting unit classifies, statistically analyzes, and accumulates the amplitude data captured by the peak extraction unit to form an energy spectrum distribution map. The data transmission unit can transmit the data in real time via asynchronous serial communication. The host computer is a standard computer host and monitor assembly, which will not be described in detail here.

[0025] The signal processing module is equipped with a programmable threshold parameter register, which is used to dynamically adjust the preset threshold voltage and counting threshold in the single threshold peak extraction algorithm.

[0026] The preset threshold voltage is dynamically adjusted by calculating using a threshold formula, expressed as: ; in, This is the base threshold, which is the factory calibration value; The background noise level is estimated in real time; This is the detector drift compensation term, obtained through real-time measurement; This is a material adaptive adjustment item; , , These are the weighting coefficients.

[0027] The background noise level can be estimated by measuring the energy spectrum in the battery-free sample, expressed as follows: It is automatically measured and updated in real time every N minutes, as shown in the following diagram: ; Where M is the number of measurements, and n is the spatial variable of the measurements.

[0028] The formula for calculating the material adaptive adjustment term is: ; Where K is the number of material types identified by the system, such as LFP, NMC, LCO, NCA, Al, etc. The weighting coefficient for the i-th material is calculated based on its recent frequency of occurrence. The preferred response mode for the i-th material is defined as the energy dependence function, where n is the number of channels, and each... It is a vector of length 1024 (corresponding to 1024 energy-blocking spectral channels), and the value represents the sensitivity that needs to be adjusted in that channel, for example... The range of its value is [-100, +100].

[0029] Weighting coefficient The calculation, for example, takes the classification results of the most recent 20 batteries: LFP: 8, weighted =0.4; NMC: 7, weights =0.35; NCA: 3, weighted =0.15; Al: 2, weights =0.1.

[0030] The calculation of the material adaptive adjustment term can then be expressed as: .

[0031] For example, at channel 400, corresponding to ~6.0 keV, the specific numerical values ​​in the Fe / Mn region are as follows: ; ; This indicates that you are not particularly concerned about it. This indicates that they are not particularly concerned about it, and ultimately... .

[0032] The clock module provides a synchronous clock signal for each module, and the communication interface module transmits the processed energy spectrum data to the host computer. The communication interface module uses USB or Ethernet protocol.

[0033] Compared with existing technologies, the amplitude analysis system in this embodiment adopts a fully digital multi-channel pulse amplitude analysis architecture, which breaks through the signal conditioning limitations of traditional analog front-ends and realizes single-chip integration of signal acquisition, shaping, extraction and energy spectrum construction. This significantly improves the system's integration, reliability and long-term operational stability, reduces system maintenance costs, and is suitable for long-term continuous operation in industrial fields. It realizes real-time acquisition, processing and energy spectrum construction of X-ray fluorescence detector signals, and has the characteristics of fast response speed, strong anti-interference ability and high integration. It is suitable for the rapid identification and sorting of various metal elements in waste battery classification systems.

[0034] Example 2 An amplitude analysis method applicable to X-ray fluorescence waste battery sorting systems, based on the amplitude analysis system of Example 1, includes the following steps: S100 acquires the output signal of the X-ray detector after preprocessing by the differentiating circuit and the preamplifier circuit, and converts it into a digital signal.

[0035] S200 first performs a trapezoidal shaping operation on the digital signal of S100, that is, performs digital filtering and waveform shaping on the original digital signal output by the ADC to suppress noise and improve pulse shape, providing a standardized signal waveform for subsequent peak extraction. The trapezoidal shaping operation is performed with configurable shaping time and attenuation constant to optimize signal noise suppression performance and pulse throughput.

[0036] S300, the peak extraction operation is performed on the shaped pulse obtained from S200 to extract features. Peak extraction is performed by a single threshold peak extraction algorithm to capture the peak amplitude of each pulse. The single threshold peak extraction algorithm is hardware accelerated by a finite state machine architecture, which includes a minimum value search state, a maximum value search state, and a peak output state.

[0037] When the finite state machine architecture is in the minimum value search state, it continuously compares the current sampled value with a preset threshold voltage. The threshold voltage is set according to actual needs. If the current sampled value is lower than the preset threshold voltage, the minimum value register is dynamically updated to record the minimum level of the signal. If the current sampled value exceeds the preset threshold voltage, it immediately switches from the minimum value search state to the maximum value search state. After entering the maximum value search state, an internal counter is started to count, and the maximum value register is compared and updated in each clock cycle to capture the highest level of the signal; in this state, if the current sampled value is detected to be lower than the preset threshold voltage again, it is immediately reset and returns to the minimum value search state. The transition from the maximum value search state to the peak output state is allowed only when both of the following conditions are met simultaneously: Condition 1, the value of the internal counter exceeds the preset time count value, indicating that the pulse width has reached the requirement; Condition 2, the current sampled value has fallen back below the preset threshold voltage, confirming that the pulse signal has passed completely, indicating that the pulse signal has entered the falling phase. In the peak output state, the difference between the values ​​stored in the maximum value register and the minimum value register is calculated, and this difference is taken as the peak result of the current acquisition pulse. At the same time, a write enable signal is generated to trigger the data write operation. After the peak output is completed, the internal counter is automatically cleared and the maximum value register and the minimum value register are reset, thereby ending the current peak finding cycle and preparing to start the processing of the next signal cycle.

[0038] S400 classifies, statistically analyzes, and accumulates the peak amplitudes captured by S300 to form an energy spectrum distribution map.

[0039] Adopting such Figure 3 The signal generator shown produces an exponential pulse wave with a frequency of 400 kHz and an amplitude of 1V-8V as the test signal. The pulse amplitude and the energy spectrum channel address are linearly fitted using Excel software.

[0040] like Figure 2 As shown, a linear fit is performed between the peak amplitude and the energy spectrum address using a preset formula, which is expressed as follows: y = 400.0476x - 44.7143; Where x is the peak amplitude and y is the energy spectrum address. Goodness of fit The value >0.99 indicates that the linearity between the pulse amplitude and the corresponding channel address is relatively good, suggesting that the measured energy spectrum channel address is basically consistent with the actual pulse amplitude, and the energy spectrum result is relatively accurate.

[0041] The method in this second embodiment applies a single-threshold finite state machine peak extraction algorithm to a waste battery X-ray fluorescence classification system. Through a state transition mechanism, it achieves rapid analysis and accurate identification of complex multi-element energy spectra such as nickel, cobalt, and manganese, solving the industry problem of inaccurate characteristic peak extraction in mixed energy spectra of battery materials using traditional methods. It possesses strong noise suppression and anti-interference capabilities, ensuring that only one peak is output per signal cycle through strict state transition conditions, effectively avoiding false detections and repeated triggering, and achieving true real-time energy spectrum acquisition and processing. Embedding a programmable parameter register in the FPGA supports dynamic configuration of core parameters such as extraction threshold and forming time, giving the system excellent adaptability and allowing flexible matching with different battery systems (such as ternary lithium and lithium iron phosphate) and production line conditions, thus broadening the equipment's application scope and lifespan.

[0042] The above descriptions are merely embodiments of the present invention, and common knowledge regarding specific structures and characteristics is not elaborated upon here. It should be noted that those skilled in the art can make various modifications and improvements without departing from the structure of the present invention, and these should also be considered within the scope of protection of the present invention. These modifications and improvements will not affect the effectiveness of the present invention or the practicality of the patent. The scope of protection claimed in this application should be determined by the content of its claims, and the specific embodiments described in the specification can be used to interpret the content of the claims.

Claims

1. An amplitude analysis system suitable for X-ray fluorescence waste battery sorting systems, including a signal acquisition module and a signal processing module; The signal acquisition module is used to acquire the output signal of the X-ray detector after preprocessing by the differentiating circuit and the preamplifier circuit, and to convert the output signal, which is an analog signal, into a digital signal. Its features are: The signal processing module is used to sequentially perform trapezoidal shaping, peak extraction, and energy spectrum plotting operations on the digital signal output by the signal acquisition module. The signal processing module performs peak extraction using a single-threshold peak extraction algorithm, which is hardware-accelerated using a finite state machine architecture. The finite state machine architecture includes a minimum search state, a maximum search state, and a peak output state.

2. The amplitude analysis system for X-ray fluorescence waste battery classification system according to claim 1, characterized in that: It also includes a clock module and a communication interface module. The clock module is used to provide a synchronous clock signal for each module, and the communication interface module is used to transmit the processed energy spectrum data to the host computer. The communication interface module adopts USB or Ethernet protocol.

3. The amplitude analysis system for X-ray fluorescence waste battery classification system according to claim 1, characterized in that: The signal processing module is configured with a programmable threshold parameter register, which is used to dynamically adjust the preset threshold voltage and counting threshold in the single threshold peak extraction algorithm.

4. The amplitude analysis system for X-ray fluorescence waste battery classification system according to claim 3, characterized in that: The signal processing module employs configurable forming parameters during the trapezoidal forming operation, including forming time and attenuation constant.

5. An amplitude analysis method applicable to X-ray fluorescence waste battery sorting systems, comprising the following steps: S100 acquires the output signal of the X-ray detector after preprocessing by the differentiating circuit and the preamplifier circuit, and converts it into a digital signal. Its characteristic is that it further includes: S200, first perform a trapezoidal forming operation on the digital signal of S100, and perform the trapezoidal forming operation with a configurable forming time and attenuation constant; S300, the peak extraction operation is performed on the shaped pulse obtained from S200 to extract features. Peak extraction is performed by a single threshold peak extraction algorithm to capture the peak amplitude of each pulse. The single threshold peak extraction algorithm is hardware accelerated by a finite state machine architecture, which includes a minimum value search state, a maximum value search state, and a peak output state. S400 classifies, statistically analyzes, and accumulates the peak amplitudes captured by S300 to form an energy spectrum distribution map.

6. The amplitude analysis method for X-ray fluorescence waste battery classification system according to claim 5, characterized in that: In S300, when the finite state machine architecture is in the minimum value search state, it continuously compares the current sampled value with the preset threshold voltage. If the current sampled value is lower than the preset threshold voltage, the minimum value register is dynamically updated to record the minimum level of the signal. If the current sampled value exceeds the preset threshold voltage, the search will immediately switch from the minimum value search state to the maximum value search state. After entering the maximum value search state, an internal counter is started to count, and the maximum value register is compared and updated in each clock cycle to capture the highest level of the signal; In this state, if the current sampled value is detected to be lower than the preset threshold voltage again, it will be immediately reset and return to the minimum value search state. It is only allowed to switch from the maximum value search state to the peak output state when the following two conditions are met at the same time: Condition 1, the value of the internal counter exceeds the preset time count value, indicating that the pulse width has reached the requirement; Condition 2, the current sampled value has fallen back to below the preset threshold voltage, indicating that the pulse signal has entered the falling phase. In the peak output state, the difference between the values ​​stored in the maximum value register and the minimum value register is calculated, and this difference is taken as the peak value of the current acquisition pulse. At the same time, a write enable signal is generated to trigger the data write operation. After the peak output is completed, the internal counter is automatically cleared and the maximum value register and minimum value register are reset, thereby ending the current peak finding cycle and preparing to start the processing of the next signal cycle.

7. The amplitude analysis method for X-ray fluorescence waste battery classification system according to claim 5, characterized in that: In step S400, the peak amplitude and the energy spectrum channel address are linearly fitted using a preset formula, which is expressed as follows: y = 400.0476x - 44.7143; Where x is the peak amplitude and y is the energy spectrum address.

Images