Raman spectrum-based pesticide residue quantitative analysis system

Abstract

The invention discloses a pesticide residue quantitative analysis system based on Raman spectrum, and relates to the technical field of analytical chemistry and detection.The pesticide residue quantitative analysis system comprises an acquisition module, a processing module, an analysis module and a regulation and control module, the acquisition module is used for acquiring samples and Raman scattering signals, converting the Raman scattering signals into original spectrum data and transmitting the original spectrum data to the processing module; calibrating the original spectrum data, acquiring calibrated spectrum data, transmitting the calibrated spectrum data to an analysis module, comparing the calibrated spectrum data with a pesticide standard spectrum library, outputting a quantitative analysis result of pesticide residues, transmitting the quantitative analysis result to a regulation and control module, storing the pesticide standard spectrum library and the calibrated spectrum data, and correcting sample detection deviation by using the calibrated spectrum data. And through the regulation and control unit, full-process automatic operation from sample processing to result output is realized, the detection reliability is improved through spectrum compensation and characteristic peak comparison, efficient automation is realized through task scheduling and time sequence synchronization, and hardware fault tolerance, authority management and multi-scene adaptation capabilities are realized.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of analytical chemistry and detection technology, and particularly relates to a pesticide residue quantitative analysis system based on Raman spectrum. BACKGROUND

[0002] In the field of food safety detection, Raman spectrum technology has become a core means for pesticide residue quantitative analysis due to its advantages of rapid nondestructive, strong specificity, high sensitivity and simultaneous analysis of multiple components. This technology is based on Raman scattering effect, and generates characteristic vibration scattering signals of pesticide molecules in the sample through laser excitation. The molecular structure spectrum is obtained through processing. The detection does not require complex pretreatment and can be completed in seconds to minutes. The sample can be analyzed in situ. Its specificity is derived from the uniqueness of the Raman characteristic peak of pesticide molecules. The type and concentration of the residual substance can be accurately identified by comparison with the standard spectrum library. In practical application, this technology can penetrate the matrix of fruits and vegetables and grains, and is suitable for batch sample screening and quantification, which improves the detection accuracy in complex matrix. With the development of portable devices and intelligent technology, Raman spectrum technology is widely used in on-site detection of agricultural products, and provides strong support for food safety supervision.

[0003] However, the existing detection system has two key problems. Firstly, the resource scheduling efficiency is low during multi-task cooperative processing. The real-time performance of spectrum acquisition and data transmission tasks is poor due to resource preemption. Traditional thread allocation and priority strategy cannot dynamically adapt to load changes, resulting in data processing delay or resource waste. Secondly, the original spectrum data is disturbed by optical devices and environmental factors, and wavelength drift and noise problems are prone to occur. Inaccurate calibration will cause cumulative errors in quantitative analysis. Moreover, the traditional quantitative method based on a single characteristic peak cannot correct the matrix effect, and cannot meet the high-precision requirements of low-concentration residue detection. Therefore, a system capable of realizing hierarchical resource isolation, dynamic spectrum calibration and accurate quantitative analysis is urgently needed to improve the reliability and efficiency of pesticide residue detection. SUMMARY

[0004] The technical problem solved by the present application is that the existing detection system has two key problems. Firstly, the resource scheduling efficiency is low during multi-task cooperative processing. The real-time performance of spectrum acquisition and data transmission tasks is poor due to resource preemption. Traditional thread allocation and priority strategy cannot dynamically adapt to load changes, resulting in data processing delay or resource waste. Secondly, the original spectrum data is disturbed by optical devices and environmental factors, and wavelength drift and noise problems are prone to occur. Inaccurate calibration will cause cumulative errors in quantitative analysis. Moreover, the traditional quantitative method based on a single characteristic peak cannot correct the matrix effect, and cannot meet the high-precision requirements of low-concentration residue detection. Therefore, a system capable of realizing hierarchical resource isolation, dynamic spectrum calibration and accurate quantitative analysis is urgently needed to improve the reliability and efficiency of pesticide residue detection.

[0005] To solve the above technical problems, the present application provides the following technical solutions: a pesticide residue quantitative analysis system based on Raman spectrum, comprising a collection module, a processing module, an analysis module and a control module; The collection module is used for collecting samples and obtaining Raman scattering signals, converting the Raman scattering signals into original spectrum data and transmitting the original spectrum data to the processing module. The processing module is used for calibrating the original spectrum data, obtaining calibrated spectrum data and transmitting the calibrated spectrum data to the analysis module. The analysis module is used for comparing the calibrated spectrum data with a pesticide standard spectrum library, outputting a quantitative analysis result of pesticide residue and transmitting the quantitative analysis result to the control module. The control module is used for storing the pesticide standard spectrum library and the calibrated spectrum data, the calibrated spectrum data being used for correcting sample detection deviation, and realizing full-process automatic operation from sample processing to result output through a control unit.

[0006] As a preferred scheme of the pesticide residue quantitative analysis system based on Raman spectrum, the collection module comprises an excitation light source and an optical probe. The Raman scattering signals are collected through the optical probe. The Raman scattering signals are converted into original spectrum data through a photodetector. The original spectrum data comprises wave number information, intensity core information and pixel information.

[0007] As a preferred scheme of the pesticide residue quantitative analysis system based on Raman spectrum, calibrating the original spectrum data comprises a wavelength drift dynamic compensation mechanism. The wavelength drift dynamic compensation mechanism specifically comprises: The pixel information and the wave number information have an original mapping relationship. The calibrated spectrum data is generated through a wave number calibration laser built in the collection module. A cubic spline interpolation method is used to correct the mapping relationship between the pixel information and the wave number information of the original spectrum data in real time.

[0008] As a preferred scheme of the pesticide residue quantitative analysis system based on Raman spectrum, the control unit comprises a task priority division unit, a resource occupation monitoring unit, a dynamic thread allocation unit, a time constraint feedback unit and a task conflict resolution unit. The task priority division unit is used for presetting a priority level according to the real-time requirement of a detection task, and specifically comprises: The original spectrum data collection task performed by the collection module is configured as the highest priority. The data transmission task is configured as a medium priority, and the data transmission task includes the process of transmitting original spectrum data from the acquisition module to the processing module and transmitting calibrated spectrum data from the processing module to the analysis module; The result calculation task is configured as a low priority; The resource occupation monitoring unit is used for collecting CPU core load, memory occupation rate, data bus bandwidth parameters and total CPU core number in real time.

[0009] As a preferred scheme of the pesticide residue quantitative analysis system based on Raman spectrum, the dynamic thread allocation unit specifically comprises: The original spectrum data acquisition task fixes o independent CPU cores as exclusive processing threads; The data transmission task dynamically allocates threads according to the size of single data volume; When the single transmission data volume is greater than or equal to the first threshold value, p transmission threads are automatically activated for parallel processing; When the single transmission data volume is less than the second threshold value, q transmission threads are enabled; The result calculation task is configured with an elastic thread pool, the maximum number of threads does not exceed a third threshold value of the total number of CPU cores, and each thread is bound to an independent L2 cache partition; The time constraint feedback unit is used for accumulating the whole-process processing time of a single sample in real time; When the processing time of any task reaches the preset whole-process time, a cross-task thread borrowing mechanism is triggered; The cross-task thread borrowing mechanism comprises: Preferably, n threads are recovered from a low-load result calculation task thread pool and allocated to a high-priority task that has timed out; After borrowing the threads, the remaining processing steps of the result calculation task are automatically combined and executed by a single thread; An internal data transmission blocking monitoring module is provided, which automatically suspends the recording task of non-urgent system logs when the continuous memory bus occupation rate reaches i.

[0010] As a preferred scheme of the pesticide residue quantitative analysis system based on Raman spectrum, the task conflict resolution unit adopts a hierarchical resource isolation strategy, specifically comprising: The memory space exclusive mechanism is implemented for the spectrum acquisition task, a dedicated area is divided in the physical memory, and other tasks are prohibited from accessing; The data transmission task enables a double-buffer queue, specifically comprising: The original spectrum data in the high-priority queue adopts a first-in-first-out (FIFO) strategy; The historical spectrum data of the low-priority queue adopts a flow control strategy, and when the CPU core load, memory occupancy rate and real-time bandwidth occupancy rate of the spectrum data transmission bus are greater than a fourth threshold value, the transmission is automatically delayed; When multiple task requests for resources conflict, the task trigger sequence is marked by a hardware timestamp, and resources are allocated in time sequence according to the priority of the tasks, and the resources are preferentially allocated when the task is in the time sequence first. In the process of quantitative calculation of the intermediate results generated in the CPU register and memory buffer synchronization storage analysis module, when the intermediate results are found to be inconsistent, the task is automatically restarted.

[0011] As a preferred scheme of the pesticide residue quantitative analysis system based on Raman spectrum, wherein: the photodetector array is configured with a temperature compensation circuit, and a platinum resistance temperature sensor is integrated to monitor the working temperature of the detector in real time. When the temperature fluctuation exceeds the set threshold, the transimpedance amplifier gain compensation unit is automatically triggered.

[0012] As a preferred scheme of the pesticide residue quantitative analysis system based on Raman spectrum, wherein: the pesticide standard spectrum library pre-stores the standard spectrum offset threshold of more than s pesticides. When the peak position offset of the measured calibrated spectrum data exceeds the corresponding threshold, the abnormality is automatically marked and the signal resampling procedure of the dual-wavelength reference channel is triggered. The calibrated spectrum data is compared with the pesticide standard spectrum library using the characteristic peak intensity ratio method. The characteristic peak intensity ratio method performs quantitative calibration by calculating the intensity ratio of the characteristic peaks of the Raman spectrum of the pesticide residue to be detected to the internal standard peak, and outputs the quantitative analysis result of the pesticide residue.

[0013] As a preferred scheme of the pesticide residue quantitative analysis system based on Raman spectrum, wherein: the data interaction interface of the regulation module supports the IEEE 1588 precision clock synchronization protocol. The acquisition module, processing module, analysis module and regulation module realize μs-level precision detection timing synchronization through hardware timestamps, and realize spectrum acquisition and sample processing action cooperation. The excitation light source is integrated with a wavelength locker, which monitors the laser wavelength in real time through a fiber Bragg grating. When the laser wavelength drift exceeds the set threshold, the laser temperature control current is automatically adjusted.

[0014] As a preferred scheme of the pesticide residue quantitative analysis system based on Raman spectrum, wherein: the human-computer interface of the regulation module is provided with multi-level permission management, specifically including: The detection personnel log in and view the real-time detection curve through fingerprint identification; The administrator-level account supports standard spectrum library updating function and calibration parameter threshold setting function.

[0015] The present application has the following advantages: through the excitation light source wavelength locking, detector temperature compensation and wavelength drift dynamic compensation mechanism, the original spectrum data is ensured to be accurate and stable, the feature peak intensity ratio method and the standard spectrum library comparison and double-wavelength reference channel resampling mechanism are combined to improve the reliability of quantitative analysis, the processing and control module realizes efficient task scheduling and conflict resolution through task priority grading, dynamic thread allocation, cross-task resource cooperation and hierarchical resource isolation strategy, the full-process automatic operation reduces human error, the precise clock synchronization ensures the timing accuracy, the hardware fault tolerance and data transmission blockage monitoring capability are also provided, the multi-level permission management is set in the man-machine interface to ensure safety, and different hardware configurations are compatible, which can be flexibly adapted to multiple scenes, and the expansibility and stability are considered. BRIEF DESCRIPTION OF DRAWINGS

[0016] Figure 1 A basic flowchart of a pesticide residue quantitative analysis system based on Raman spectrum is provided for an embodiment of the present application. DETAILED DESCRIPTION

[0017] In order to make the above-mentioned objects, features and advantages of the present application more obvious and easy to understand, the specific embodiments of the present application will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present application, rather than all the embodiments.

[0018] Embodiment 1, refer to Figure 1 For an embodiment of the present application, a pesticide residue quantitative analysis system based on Raman spectrum is provided, which includes a collection module, a processing module, an analysis module and a control module. The collection module is used for collecting samples and obtaining Raman scattering signals, converting the Raman scattering signals into original spectrum data, and transmitting the original spectrum data to the processing module. The processing module is used for calibrating the original spectrum data, obtaining calibrated spectrum data, and transmitting the calibrated spectrum data to the analysis module. The analysis module is used for comparing the calibrated spectrum data with a pesticide standard spectrum library, outputting a quantitative analysis result of pesticide residue, and transmitting the quantitative analysis result to the control module. The control module is used for storing the pesticide standard spectrum library and the calibrated spectrum data, the calibrated spectrum data is used for correcting sample detection deviation, and the control module realizes full-process automatic operation from sample processing to result output through a control unit.

[0019] In one of the embodiments, four core components including a collection module, a processing module, an analysis module and a control module are included, the collection module acquires Raman scattering signals of a sample, generates original spectral data including wave number, intensity and pixel information through photoelectric conversion, and transmits the original spectral data to the processing module; the processing module calibrates the original spectral data by using a wavelength drift dynamic compensation calibration mechanism (such as correcting the mapping relationship between pixels and wave numbers by using a calibration laser combined with a cubic spline interpolation method), forms calibrated spectral data, and transmits the calibrated spectral data to the analysis module; the analysis module compares and analyzes the calibrated data with a database of more than or equal to 500 pesticide standard spectra, and outputs pesticide residue quantitative results to the control module; the control module integrates an automatic control unit of task priority division and dynamic thread allocation, realizes automatic control of the whole process from sample collection to result output, stores a standard spectrum library and calibration data for detection deviation correction, and ensures efficient and accurate operation of the system.

[0020] The collection module includes an excitation light source and an optical probe; The Raman scattering signals are collected by the optical probe; The Raman scattering signals are converted into original spectral data by a photoelectric detector; The original spectral data include wave number information, intensity core information and pixel information.

[0021] Calibrating the original spectral data includes a wavelength drift dynamic compensation mechanism; The wavelength drift dynamic compensation mechanism specifically includes: The pixel information and the wave number information have an original mapping relationship; The calibrated spectral data are generated by using a wave number calibration laser built in the collection module; The mapping relationship between the pixel information and the wave number information of the original spectral data is corrected in real time by using a cubic spline interpolation method.

[0022] In one of the embodiments, the collection module is equipped with an excitation light source and an optical probe, the Raman scattering signals of a sample are collected by the optical probe, original spectral data including wave number information, intensity core information and pixel information are generated by a photoelectric detector, the processing module calibrates the original spectral data by using a wavelength drift dynamic compensation mechanism, calibration data are generated by using a wave number calibration laser built in the collection module based on the original mapping relationship between the pixel information and the wave number information, the mapping relationship between the pixel information and the wave number information is corrected in real time by using a cubic spline interpolation method, the wavelength drift error is eliminated, and the accuracy of the calibrated spectral data is ensured.

[0023] The control unit includes a task priority division unit, a resource occupation monitoring unit, a dynamic thread allocation unit, a time constraint feedback unit and a task conflict resolution unit; The task priority division unit is configured to preset a priority level according to the real-time requirement of the detection task, and specifically includes: The raw spectrum data acquisition task performed by the acquisition module is configured as the highest priority; The data transmission task is configured as a medium priority, and the data transmission task includes the process of transmitting the raw spectrum data from the acquisition module to the processing module and transmitting the calibrated spectrum data from the processing module to the analysis module; The result calculation task is configured as a low priority; The resource occupation monitoring unit is configured to collect the CPU core load, memory occupation rate, data bus bandwidth parameters and total CPU core number in real time.

[0024] In one of the embodiments, the regulation unit includes a task priority division unit, a resource occupation monitoring unit, a dynamic thread allocation unit, a time constraint feedback unit and a task conflict resolution unit. The task priority division unit presets three priority levels according to the real-time requirement of the detection task: the raw spectrum data acquisition task performed by the acquisition module is set as the highest priority to ensure the real-time performance of data acquisition; the data transmission task (including the transmission process of raw spectrum data from the acquisition module to the processing module and the transmission process of calibrated spectrum data from the processing module to the analysis module) is configured as a medium priority; the result calculation task is configured as a low priority. The resource occupation monitoring unit collects the CPU core load, memory occupation rate, data bus bandwidth and total CPU core number in real time to provide real-time data support for dynamic resource scheduling.

[0025] The dynamic thread allocation unit specifically includes: The raw spectrum data acquisition task is fixedly reserved o independent CPU cores as exclusive processing threads; The data transmission task is dynamically allocated threads according to the size of single data volume; When the single transmission data volume is greater than or equal to the first threshold value, p transmission threads are automatically activated for parallel processing; When the single transmission data volume is less than the second threshold value, q transmission threads are enabled; The result calculation task is configured with an elastic thread pool, the maximum number of threads does not exceed the third threshold value of the total number of CPU cores, and each thread is bound to an independent L2 cache partition; The time constraint feedback unit is configured to accumulate the whole-process processing time of a single sample in real time; When the processing time of any task reaches the preset whole-process time, a cross-task thread borrowing mechanism is triggered; The cross-task thread borrowing mechanism includes: n threads are preferentially recovered from the low-load result calculation task thread pool and allocated to the high-priority task that has timed out; After borrowing the threads, the remaining processing steps of the result calculation task are automatically combined for single-thread execution; Built-in data transmission congestion monitoring module, when the data bus occupancy rate reaches i for m consecutive times, automatically suspends the recording task of non-urgent system log.

[0026] In one of the embodiments, the dynamic thread allocation unit implements differentiated thread scheduling strategies according to task characteristics: o (usually 2-4, set according to the total number of CPU cores and the real-time demand of the acquisition task) independent CPU cores are fixedly reserved as exclusive processing threads for the original spectral data acquisition task; the data transmission task dynamically adjusts the number of threads according to the single data volume, activates p (usually 4-8, to ensure efficient parallel transmission) threads for parallel processing when the data volume is greater than or equal to 1MB (first threshold value), and enables q (usually 1-2) threads when the data volume is less than 1MB (second threshold value); the result calculation task is configured with a flexible thread pool, the maximum number of threads does not exceed 50% (third threshold value) of the total number of CPU cores, and each thread is bound to an independent L2 cache partition to improve calculation efficiency, the time constraint feedback unit monitors the single sample full-process processing time in real time, and when the processing time of any task reaches 80% (typical value is 400ms, this threshold value is set according to the system hardware configuration such as 4-8 total CPU cores, 10GB / s data bus bandwidth and the real-time requirement of pesticide residue detection, and the average processing time of single sample full process is 500ms) of the average processing time of single sample full process, triggers the cross-task thread borrowing mechanism, and preferentially recycles n (usually 1-2) threads from the result calculation task thread pool to support high-priority tasks, while merging the remaining steps of the calculation task into single-thread execution; the built-in data transmission congestion monitoring module automatically suspends the recording of non-urgent system log when the data bus occupancy rate reaches 100% (i value) for 50ms (m value) in a row, to ensure core data transmission.

[0027] The task conflict resolution unit adopts a hierarchical resource isolation strategy, which specifically includes: Implementing a memory space exclusive mechanism for spectral acquisition tasks, dividing a dedicated area in physical memory and prohibiting access by other tasks; The data transmission task enables a double-buffer queue, which specifically includes: The original spectral data in the high-priority queue adopts a first-in-first-out (FIFO) strategy; The historical spectral data in the low-priority queue adopts a flow control strategy, which automatically delays transmission when the CPU core load, memory occupancy rate, and real-time bandwidth occupancy rate of the spectral data transmission bus are greater than the fourth threshold value; When multiple tasks request resources conflict, the hardware timestamp is used to mark the task trigger sequence, and the resources are allocated according to the time sequence of tasks with the same priority, and the resources are preferentially allocated when the task time sequence is earlier; The intermediate results generated in the quantitative calculation process of the CPU register and the memory buffer synchronization storage analysis module are found, and the task is automatically restarted when the intermediate results are inconsistent.

[0028] In one of the embodiments, the task conflict resolution unit adopts a hierarchical resource isolation strategy to ensure the stable operation of the acquisition module, the processing module, the analysis module and the control module: a memory space exclusive mechanism is implemented for the spectrum acquisition task, a dedicated area is divided in the physical memory and other tasks are prohibited from accessing, to ensure that the core data acquisition is not disturbed; a double-buffer queue is enabled for the data transmission task, the original spectrum data in the high-priority queue adopts a first-in-first-out (FIFO) strategy to ensure real-time performance, and the historical spectrum data in the low-priority queue adopts a flow control strategy, which automatically delays transmission when any of the CPU core load, memory occupancy rate and spectrum data transmission bus bandwidth occupancy rate exceeds 80% (the fourth threshold value is set according to the critical safety value of the resource to avoid overload affecting the core task); when there is a conflict in the resource request of multiple tasks, the task triggering sequence is marked by a hardware timestamp, and the resources are allocated to the tasks of the same priority according to the time sequence to ensure that the task triggered first acquires the resources first; the intermediate results of the quantitative calculation of the CPU register and the memory buffer synchronization storage analysis module are stored, and if it is detected that the two are inconsistent, the task is automatically restarted to ensure the accuracy and reliability of the calculation process.

[0029] The photodetector array is configured with a temperature compensation circuit, which monitors the working temperature of the detector in real time through an integrated platinum resistance temperature sensor; When the temperature fluctuation exceeds the set threshold, the gain compensation unit of the transimpedance amplifier is automatically triggered.

[0030] The photodetector array is configured with a temperature compensation circuit, which monitors the working temperature of the detector in real time through an integrated platinum resistance temperature sensor; When the temperature fluctuation exceeds the set threshold, the gain compensation unit of the transimpedance amplifier is automatically triggered.

[0031] The pesticide standard spectrum library pre-stores the standard spectrum offset threshold of more than s pesticides; When the peak position offset of the measured and calibrated spectrum data exceeds the corresponding threshold, the abnormality is automatically marked and the signal re-sampling procedure of the dual-wavelength reference channel is triggered; The calibrated spectrum data is compared with the pesticide standard spectrum library using the characteristic peak intensity ratio method; The characteristic peak intensity ratio method quantitatively calibrates the intensity ratio of the Raman spectrum characteristic peak corresponding to the pesticide residue to be detected to the internal standard peak, and outputs the quantitative analysis result of the pesticide residue.

[0032] In one embodiment, the photodetector array integrates a temperature compensation mechanism, which monitors the operating temperature in real time through a built-in platinum resistance temperature sensor. When the temperature fluctuation exceeds ±0.5℃ (set threshold, set according to the temperature characteristic curve of the detector sensitivity to ensure that the temperature drift has less than 0.1% impact on signal acquisition), the automatic trigger cross resistance amplifier gain compensation unit maintains signal acquisition stability. The pesticide standard spectrum library pre-stores the standard spectrum offset threshold of more than 500 kinds of pesticides (the single peak position offset threshold is typically 0.5 cm -1 , set according to the resolution of the spectrometer and the characteristic peak broadening characteristics of the pesticide), when the measured calibration spectrum peak position offset exceeds the corresponding threshold, it automatically marks the abnormal and starts the double-wavelength reference channel re-sampling program; the characteristic peak intensity ratio method is used for quantitative analysis, the intensity ratio of the characteristic peak of the pesticide to be detected and the internal standard peak (selected from the stable chemical bond vibration peak in the sample matrix) is calculated, and the intensity ratio range of the corresponding pesticide in the standard spectrum library is directly compared to output the quantitative analysis result of pesticide residue, ensuring the detection accuracy and repeatability.

[0033] The data interaction interface of the regulation module supports IEEE 1588 precision clock synchronization protocol; The acquisition module, processing module, analysis module and regulation module realize μs level precision detection timing synchronization through hardware time stamp, realizing the cooperation of spectrum acquisition and sample processing action; The excitation light source integrates a wavelength locker, which monitors the laser wavelength in real time through a fiber Bragg grating; When the laser wavelength drift exceeds the set threshold, the laser temperature control current is automatically adjusted.

[0034] The human-computer interface of the regulation module sets multi-level permission management, including: The detection personnel log in and view the real-time detection curve through fingerprint recognition; The administrator account supports standard spectrum library update function and calibration parameter threshold setting function.

[0035] In one of the embodiments, the regulation module configures the data interaction interface of IEEE 1588 precision clock synchronization protocol, and realizes the detection timing synchronization with the μs level precision through the hardware timestamp with the acquisition module, the processing module and the analysis module, ensures the accurate cooperation of the spectrum acquisition action and the sample processing flow, excites the light source integrated wavelength locker, monitors the laser wavelength in real time through the fiber Bragg grating, when the wavelength drift exceeds the preset threshold ±0.05nm (the threshold is set, according to the laser temperature coefficient 0.08~0.1nm / ℃ and the spectrometer resolution 0.06nm, and the influence of the wavelength fluctuation on the spectrum analysis is less than 0.1%), the laser temperature control current is automatically adjusted to maintain the wavelength stability, the human-computer interface of the regulation module adopts the multi-level permission management mechanism: the detection personnel can view the real-time detection curve after logging in through the fingerprint recognition; the administrator level account supports the advanced functions of the standard spectrum library update and the calibration parameter threshold setting, and guarantees the safety of the operation and the standardization of the data management.

[0036] The application guarantees the accuracy and stability of the original spectrum data through the wavelength locking of the excitation light source, the temperature compensation of the detector and the dynamic compensation mechanism of the wavelength drift, improves the reliability of the quantitative analysis by combining the characteristic peak intensity ratio method with the standard spectrum library comparison and the double-wavelength reference channel resampling mechanism, realizes the efficient task scheduling and conflict resolution through the task priority classification, the dynamic thread allocation, the cross-task resource cooperation and the hierarchical resource isolation strategy of the processing and regulation modules, reduces the human error through the full-process automatic operation, ensures the timing accuracy through the precise clock synchronization, has the hardware fault tolerance and the data transmission blockage monitoring capability, guarantees the safety through the multi-level permission management of the human-computer interface, and is compatible with different hardware configurations, can be flexibly adapted to multiple scenes, and takes into account the expansibility and the stability.

[0037] Those skilled in the art will appreciate that embodiments of the present application can be readily used as a method, a system or a computer program product. Accordingly, the present application can take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application can take the form of a computer program product on one or more computer-usable storage media (or computer- readable storage media) having computer-usable program code embodied in the medium. The medium may Figure 1 one or more functions specified in the flow or flows and / or blocks Figure 1 one or more functions specified in the flow or flows and / or blocks

[0038] It should be noted that the above-mentioned embodiments are only used to illustrate but not to limit the technical solutions of the present application. Although the present application is described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present application can be modified or equivalently replaced, without departing from the spirit and scope of the technical solutions of the present application, which should be covered in the scope of the claims of the present application.

Claims

1. A quantitative analysis system for pesticide residues based on Raman spectroscopy, characterized by, The system comprises a collecting module, a processing module, an analyzing module and a regulating module. The collecting module is configured to collect samples and obtain Raman scattering signals, convert the Raman scattering signals into raw spectrum data, and transmit the raw spectrum data to the processing module. The processing module is configured to calibrate the raw spectrum data, obtain calibrated spectrum data, and transmit the calibrated spectrum data to the analyzing module. The analyzing module is configured to compare the calibrated spectrum data with a pesticide standard spectrum library, output quantitative analysis results of pesticide residues, and transmit the quantitative analysis results to the regulating module. The regulating module is configured to store the pesticide standard spectrum library and the calibrated spectrum data, correct sample detection deviation by using the calibrated spectrum data, and realize full-process automatic operation from sample processing to result output by using a regulating unit.

2. The Raman spectroscopy-based quantitative analysis system for pesticide residues according to claim 1, characterized by: The collecting module comprises an excitation light source and an optical probe. The Raman scattering signals are collected by the optical probe. The Raman scattering signals are converted into raw spectrum data by a photoelectric detector. The raw spectrum data comprises wave number information, intensity core information and pixel information.

3. The Raman spectroscopy-based quantitative analysis system for pesticide residues according to claim 2, characterized by: The calibration of the raw spectrum data comprises a wavelength drift dynamic compensation mechanism. The wavelength drift dynamic compensation mechanism comprises: The pixel information and the wave number information have an original mapping relationship. The calibrated spectrum data is generated by a wave number calibration laser built in the collecting module. The mapping relationship between the pixel information and the wave number information of the raw spectrum data is corrected in real time by using a cubic spline interpolation method.

4. The Raman spectroscopy-based quantitative analysis system for pesticide residues according to claim 3, characterized by: The regulating unit comprises a task priority division unit, a resource occupation monitoring unit, a dynamic thread allocation unit, a time constraint feedback unit and a task conflict resolution unit. The task priority division unit is configured to preset a priority level according to the real-time requirement of a detection task, and specifically comprises: The raw spectrum data collection task performed by the collecting module is configured as the highest priority. The data transmission task is configured as a medium priority, and the data transmission task comprises the process of transmitting the raw spectrum data from the collecting module to the processing module and transmitting the calibrated spectrum data from the processing module to the analyzing module. The result calculation task is configured as a low priority. The resource occupation monitoring unit is configured to collect CPU core load, memory occupation rate, data bus bandwidth parameters and total CPU core number in real time.

5. The Raman spectroscopy-based quantitative analysis system for pesticide residues according to claim 4, characterized by: The dynamic thread allocation unit specifically comprises: The raw spectrum data collection task is fixedly reserved o independent CPU cores as exclusive processing threads. The data transmission task is dynamically allocated threads according to the size of single data volume. When the single transmission data volume is greater than or equal to a first threshold value, p transmission threads are automatically activated for parallel processing. When the single transmission data volume is less than a second threshold value, q transmission threads are enabled. The result calculation task is configured with an elastic thread pool, the maximum number of threads does not exceed a third threshold value of the total CPU core number, and each thread is bound to an independent L2 cache partition. The time constraint feedback unit is configured to accumulate the full-process processing time of a single sample in real time. When the processing time of any task reaches the preset time of the full process, a cross-task thread borrowing mechanism is triggered. The cross-task thread borrowing mechanism comprises: Prioritize the result calculation task thread pool from low load to recover n threads, assigned to handle high priority tasks timeout; After borrowing the thread, the remaining processing steps of the result calculation task are automatically merged into single-thread execution; Built-in data transmission blocking monitoring module, when the continuous m occupancy rate of the data bus reaches i, automatically pause the recording task of non-emergency system log.

6. The Raman spectroscopy-based quantitative analysis system for pesticide residues according to claim 5, characterized by: The task conflict resolution unit adopts a hierarchical resource isolation strategy, specifically including: Implementing a memory space exclusive mechanism for spectral acquisition tasks, dividing a dedicated area in physical memory and prohibiting access by other tasks; The data transmission task enables double-buffered queues, specifically including: The original spectral data in the high-priority queue uses the first-in-first-out (FIFO) strategy; The historical spectral data in the low-priority queue uses a flow control strategy, which automatically delays transmission when the CPU core load, memory occupancy rate, and real-time bandwidth occupancy rate of the spectral data transmission bus parameters are greater than the fourth threshold value; When multiple task requests conflict for resources, the hardware timestamp is used to mark the task trigger order, and resources are allocated in time sequence for tasks of the same priority. When the task is in the time sequence first, it is given priority in resource allocation; In the CPU register and memory buffer synchronization storage analysis module, the intermediate results generated in the quantitative calculation process are calculated, and when the intermediate results are inconsistent, the task is automatically restarted.

7. The Raman spectroscopy-based quantitative analysis system for pesticide residues according to claim 6, characterized by: The photodetector array is configured with a temperature compensation circuit, which uses an integrated platinum resistance temperature sensor to monitor the operating temperature of the detector in real time; When the temperature fluctuation exceeds the set threshold, the transimpedance amplifier gain compensation unit is automatically triggered.

8. The Raman spectroscopy-based quantitative analysis system for pesticide residues according to claim 7, characterized by: The pesticide standard spectrum library pre-stores the standard spectrum offset threshold of s pesticides or more; When the peak position offset of the measured and calibrated spectral data exceeds the corresponding threshold, the abnormality is automatically marked and the signal resampling procedure of the dual-wavelength reference channel is triggered; The calibrated spectral data are compared with the pesticide standard spectrum library using the characteristic peak intensity ratio method; The characteristic peak intensity ratio method quantitatively calibrates the intensity ratio of the Raman spectral characteristic peak corresponding to the pesticide residue to be detected to the internal standard peak, and outputs the quantitative analysis result of the pesticide residue.

9. The Raman spectroscopy-based quantitative analysis system for pesticide residues according to claim 8, characterized by: The data interaction interface of the regulation module supports the IEEE 1588 precision clock synchronization protocol; The acquisition module, processing module, analysis module, and regulation module achieve μs-level precision detection timing synchronization through hardware timestamps, realizing the coordination of spectral acquisition and sample processing actions; The excitation light source is integrated with a wavelength locker, which uses a fiber Bragg grating to monitor the laser wavelength in real time; When the laser wavelength drift exceeds the set threshold, the laser temperature control current is automatically adjusted.

10. The Raman spectroscopy-based quantitative analysis system for pesticide residues according to claim 9, characterized by: The human-machine interface of the regulation module sets up multi-level permission management, specifically including: The detection personnel log in and view the real-time detection curve through fingerprint recognition; The administrator-level account supports standard spectrum library update function and calibration parameter threshold setting function.

Images