Method and apparatus for detecting particulate matter in cleanliness extracts based on light scattering
By binding the filter membrane weighing and light scattering detection data to the same source and reconstructing the dual path, the error problem when the filter membrane weighing results and light scattering particle size results are used independently is solved. This achieves quality closure verification and error correction for the detection of particulate matter in the cleanliness extract, improving the accuracy and reliability of the detection.
Patent Information
- Application Number
- CN202611636010.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-10-22
- Publication Date
- 2026-11-17
AI Technical Summary
In the existing technology, when the filter membrane weighing result and the light scattering particle size result are used independently, the error of a single detection result cannot be identified. This leads to insufficient background scattering subtraction, particle size range assignment shift, sampling window not meeting the conditions, and detection errors caused by extract sedimentation and missed detection of large-diameter particles that cannot be located in a timely manner.
By acquiring batch detection data of the extract, a verification identifier is generated through homogeneous binding. The mass difference of filter membrane particles is calculated and a weighing participation identifier is generated. The sampling window characteristics are determined. The equivalent particle mass of light scattering is calculated based on the optical path flow sampling dataset. The mass closure difference is reconstructed using a dual-path method to generate the original closed output report.
It achieves quality closure verification between filter membrane weighing results and light scattering particle size results, and can identify and correct problems such as insufficient background scattering subtraction, particle size range assignment offset, sampling window not meeting requirements, and missed detection of large-diameter particles, thereby improving the accuracy and reliability of detection.
Smart Images

Figure CN123111902A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of particulate matter detection in cleanliness extracts, specifically to a method and apparatus for detecting particulate matter in cleanliness extracts based on light scattering. Background Technology
[0002] With the development of precision component cleanliness testing, water pollution particulate matter monitoring, and automated detection devices, the number, size distribution, and mass of particles in cleanliness extracts need to be cross-verified. In existing solutions, the filter membrane sampling and weighing process primarily determines the particle retention mass based on the difference between the final mass and the initial mass of the filter membrane, while the light scattering detection process mainly outputs particle detection results based on the scattering pulse peak value, particle size range count, and particle size distribution.
[0003] For example, patent CN112585447B discloses a particulate sensor, including a light source and a photodetector mounted on a substrate. A housing is attached to the substrate and includes a first portion and a second portion stacked and attached to each other on the substrate, such that the first portion is disposed between the substrate and the second portion. The first and second portions together define a light reflection chamber, a fluid flow conduit, a particle-light interaction chamber, and a light trap chamber. The first portion has a first aperture through which light emitted by the light source can reach a reflective surface within the light reflection chamber. The reflective surface is configured to reflect light toward the particle-light interaction chamber, where the light can interact with particles in a fluid flowing in the fluid flow conduit. The first portion has a second aperture through which light scattered in the particle-light interaction chamber due to interaction with one or more particles can pass for sensing by the detector. The fluid flow conduit includes a fluid inlet portion having an end directly coupled to the particle-light interaction chamber.
[0004] For example, patent application CN111247418B discloses a particle measurement system and method, comprising: at least one laser generating an illumination beam; a focusing lens that focuses the illumination beam onto a particle moving relative to the illumination beam through a focusing region of the focusing lens at a known or predetermined angle to the illumination beam; and at least two forward-looking detectors for detecting the interaction between the particle and the illumination beam in the focusing region of the focusing lens. The focusing lens is a cylindrical lens forming the focusing region, which is (i) narrow in the direction of relative motion between the particle and the illumination beam, and (ii) wide in a direction perpendicular to a plane defined by the optical axis of the system and the direction of relative motion between the particle and the illumination beam. Each of the two forward-looking detectors consists of two segmented linear detector arrays.
[0005] In the existing technology, the existing system outputs the filter membrane weighing result and the light scattering particle size result separately. It lacks mass closure verification and deviation direction back calculation under the same extraction batch. This results in insufficient background scattering deduction, particle size range assignment offset, sampling window not meeting the conditions, and detection errors caused by extract sedimentation and missed detection of large-diameter particles that cannot be located in a timely manner.
[0006] Therefore, in order to address the above problems, there is an urgent need for a method and device for detecting particulate matter in cleanliness extracts based on light scattering. Summary of the Invention
[0007] Technical problems to be solved
[0008] To address the shortcomings of existing technologies, this invention provides a method and apparatus for detecting particulate matter in cleanliness extracts based on light scattering, which solves the problem that errors in a single detection result cannot be identified when the filter membrane weighing result and the light scattering particle size result are used independently.
[0009] Technical solution
[0010] To achieve the above objectives, the present invention provides the following technical solution: a method for detecting particulate matter in cleanliness extracts based on light scattering, comprising: S1, acquiring batch-specific test data of the extract, performing homogeneous binding, generating a verification identifier, and obtaining a batch-specific test dataset of the extract; S2, calculating the mass difference of filter membrane particles based on the batch-specific test dataset of the extract and generating a weighing participation identifier; S3, generating sampling window features based on the batch-specific test dataset of the extract, determining the sampling participation identifier, and generating an optical path flow sampling dataset; S4, calculating the light scattering equivalent particle mass based on the optical path flow sampling dataset and a preset detection parameter table; S5, calculating the mass closure difference and generating a mass closure identifier, performing dual-path reconstruction on the mass closure difference, and generating an original closure output report.
[0011] Further, the specific process of obtaining batch-specific test data of the extract and performing source binding to generate verification identifiers to obtain the batch-specific test dataset of the extract is as follows: Batch-specific test data of the extract is collected using a cleanliness extract particulate matter detection device; the batch-specific test data includes extraction batch identifier, sample identifier, cleaning mode identifier, extract volume, extract temperature, extract settling time, filter membrane sampling data, filter membrane weighing data, light scattering detection data, flow status data, optical path status data, multi-level time data, and a preset detection parameter table; using the extraction batch identifier and sample identifier in the batch-specific test data as indexes, the filter membrane sampling data, filter membrane weighing data, and light scattering detection data corresponding to the same batch of extract are bound to source; when the filter membrane number in the batch-specific test data is empty, the filter... A batch data missing identifier is generated when the membrane sampling completion identifier is 0, the scattering pulse peak value is empty, the scattering pulse width is empty, the initial filter membrane mass is empty, or the final filter membrane mass is empty. A collection time difference verification identifier is generated when the difference between the device acquisition time and the controller receiving time in the same batch of extraction solution test data is greater than the first time allowable deviation threshold, or the difference between the controller receiving time and the processor writing time is greater than the second time allowable deviation threshold. A filter membrane mixing verification identifier is generated when the same extraction batch identifier corresponds to different filter membrane numbers. A batch attribution verification identifier is generated when the same filter membrane number corresponds to different extraction batch identifiers. Data that do not generate a batch data missing identifier and do not generate a collection time difference verification identifier, a filter membrane mixing verification identifier, or a batch attribution verification identifier are written into the same batch of extraction solution test dataset.
[0012] Furthermore, based on the batch-specific dataset of the extract, the specific process for calculating the filter membrane particle mass difference and generating weighing participation indicators is as follows: The filter membrane sampling completion indicator, drying completion indicator, cooling completion indicator, and weighing stability indicator in the batch-specific dataset of the extract are encoded using binary codes. A value of 1 indicates that the corresponding step has been completed, and a value of 0 indicates that the corresponding step has not been completed. When all three indicators are 1, the difference between the final filter membrane mass and the initial filter membrane mass is calculated to obtain the filter membrane particle mass difference. When the filter membrane particle mass difference is greater than 1, the calculation continues. When the difference between the filter membrane particle mass difference and the filter membrane mass difference is equal to or less than or equal to the upper limit threshold of the filter membrane mass difference, the weighing participation flag is set to 1, and the filter membrane particle mass difference is written into the filter membrane mass difference dataset. When any of the following flags is 0: filter membrane sampling completion flag, drying completion flag, cooling completion flag, and weighing stabilization flag, the weighing participation flag is set to 0, and a weighing precondition verification flag is generated. When the filter membrane particle mass difference is less than 0, the weighing participation flag is set to 0, and a mass difference direction verification flag is generated. When the filter membrane particle mass difference is greater than the upper limit threshold of the filter membrane mass difference, the weighing participation flag is set to 0, and a mass difference upper bound verification flag is generated.
[0013] Furthermore, based on the batch-of-extract detection dataset, the specific process for calculating the sampling window features is as follows: Based on the batch-of-extract detection dataset, the detection pool flow rate sequence, detection pool pressure sequence, light source emission intensity sequence, photoelectric receiving intensity sequence, blank sampling scattering intensity sequence, and extraction liquid temperature sequence within the same sampling window are read according to the sampling window number. The window extreme value difference method is then used to extract extreme values and calculate extreme value differences in the sequences to generate sampling window feature data. The sampling window feature data includes minimum flow rate, maximum flow rate, flow rate fluctuation value, minimum pressure, maximum pressure, pressure fluctuation value, minimum light source intensity, maximum light source intensity, photoelectric receiving peak value, blank scattering intensity, minimum temperature, maximum temperature, and temperature fluctuation value. Extreme values are extracted from the photoelectric receiving intensity sequence, and the maximum value in the photoelectric receiving intensity sequence is taken as the photoelectric receiving peak value. The arithmetic mean of the blank sampling scattering intensity sequence is calculated, and the arithmetic mean is taken as the blank scattering intensity.
[0014] Furthermore, the specific process of determining the sampling participation identifier and generating the optical path circulation sampling dataset is as follows: when the minimum flow rate is greater than or equal to the lower flow rate threshold, the maximum flow rate is less than or equal to the upper flow rate threshold, the flow rate fluctuation value is less than or equal to the flow rate fluctuation threshold, the minimum pressure is greater than or equal to the lower pressure threshold, the maximum pressure is less than or equal to the upper pressure threshold, the pressure fluctuation value is less than or equal to the pressure fluctuation threshold, the minimum light source intensity is greater than or equal to the lower light source intensity threshold, the maximum light source intensity is less than or equal to the upper light source intensity threshold, the photoelectric receiving peak value is less than the receiving saturation threshold, the blank scattering intensity is less than or equal to the blank background threshold, the minimum temperature is greater than or equal to the lower temperature threshold, the maximum temperature is less than or equal to the upper temperature threshold, and the temperature fluctuation value is less than or equal to the temperature fluctuation threshold, the sampling participation identifier is set to 1 and written into the optical path circulation sampling dataset; otherwise, the sampling participation identifier is set to 0 and a sampling condition verification identifier is generated.
[0015] Furthermore, based on the optical path flow sampling dataset and the preset detection parameter table, the specific process for calculating the equivalent particle mass of light scattering is as follows: Subtract the blank scattering intensity from the scattering pulse peak in the optical path flow sampling dataset to obtain the subtracted scattering peak; when the subtracted scattering peak is greater than or equal to the particle identification threshold, perform interval matching and interval assignment statistics on the subtracted scattering peak according to the preset detection parameter table to generate particle size interval counts; when the blank scattering intensity is less than or equal to the blank background threshold, and the extract temperature is greater than or equal to the lower limit threshold and less than or equal to the upper limit threshold... When the particle size interval count is less than or equal to the upper limit threshold of the linear count, the calculation of the light scattering equivalent particle mass is initiated; otherwise, an equivalent mass verification identifier is generated. Using the temperature linear correction method, the median of the particle size interval is multiplied by the temperature correction factor to obtain the corrected median of the particle size interval. The equivalent volume of a single particle is calculated according to the spherical equivalent volume. The equivalent volume of a single particle is multiplied by the particle size interval count to obtain the interval equivalent volume. The interval equivalent volume is multiplied by the particle density parameter to obtain the interval equivalent mass. The interval equivalent mass of each particle size interval is summed to obtain the light scattering equivalent particle mass.
[0016] Further, the specific process for calculating the mass closure difference and generating the mass closure indicator is as follows: When both the weighing participation indicator and the sampling participation indicator are 1, and the equivalent mass verification indicator has not been generated, the absolute value of the difference between the filter membrane particle mass difference and the light scattering equivalent particle mass is calculated to obtain the mass closure difference; when the mass closure difference is less than or equal to the mass closure threshold, the mass closure indicator is set to 1; when the mass closure difference is greater than the mass closure threshold, the mass closure indicator is set to 0, and a mass closure verification indicator is generated; when the result of subtracting the filter membrane particle mass difference from the light scattering equivalent particle mass is greater than the mass closure threshold, a scattering mass excess indicator is generated; when the result of subtracting the light scattering equivalent particle mass difference from the filter membrane particle mass difference is greater than the mass closure threshold, a weighing mass excess indicator is generated; when the weighing participation indicator is 0, or the sampling participation indicator is 0, or the equivalent mass verification indicator has been generated, the mass closure difference is not calculated, and the mass closure indicator is set to 0.
[0017] Furthermore, the specific process of dual-path reconstruction of the mass closure difference is as follows: When the scattering mass error indicator has been generated, the scattering-side reconstruction of the light scattering equivalent particle mass is performed based on the background residual reconstruction and particle size boundary reconstruction to obtain the background residual reconstruction mass and particle size boundary reconstruction mass; the background residual reconstruction mass and particle size boundary reconstruction mass are compared with the filter membrane particle mass difference, and the reconstruction mass with the smaller difference is selected as the scattering-side correction mass, and the corresponding scattering-side correction path is recorded; when the weighing mass error indicator has been generated, the light scattering equivalent particle mass is lower than the filter membrane particle mass difference as the correction object, and the light scattering missed detection mass is reconstructed based on window compensation reconstruction and large particle size compensation reconstruction to obtain the window-compensated equivalent mass and the large particle size compensated equivalent mass; the absolute values of the differences between the window-compensated equivalent mass, the large particle size compensated equivalent mass and the filter membrane particle mass difference are calculated respectively to obtain the window compensation correction closure difference and the large particle size compensation correction closure difference; the compensated equivalent mass with a smaller correction closure difference and less than the original mass closure difference is selected as the sampling-side correction mass, and the corresponding sampling-side correction path is recorded.
[0018] Furthermore, the specific process for generating the original closed output report is as follows: When the quality closure flag is 1, the light scattering equivalent particle mass, filter membrane particle mass difference, particle size interval count, and quality closure difference are written into the detection output data to generate the original closed output report; when the quality closure verification flag has been generated, the absolute values of the differences between the scattering side correction quality, the sampling side correction quality, and the filter membrane particle mass difference are calculated respectively, and the correction quality of the side with the smaller difference is selected as the target correction quality, and the corrected quality closure difference is calculated; when the corrected quality closure difference is less than the quality closure difference and less than or equal to the quality closure threshold, the light scattering equivalent particle mass in the detection output data is updated with the target correction quality and written into the correction closure output record; when the corrected quality closure difference is greater than the quality closure threshold, or the corrected quality closure difference is greater than or equal to the quality closure difference, the final corrected particulate matter detection result is not output, only the verification output record is output.
[0019] Furthermore, a second aspect of the present invention provides a cleanliness extract particulate matter detection device based on light scattering, applied to a cleanliness extract particulate matter detection method based on light scattering, comprising: an extract sampling unit for outputting extraction batch identifier, sample identifier, cleaning mode identifier, extract volume, extract temperature, and extract settling time; a filter membrane sampling and weighing unit for outputting filter membrane sampling data, filter membrane number, filter membrane sampling completion identifier, filter membrane initial mass, filter membrane final mass, drying completion identifier, cooling completion identifier, and weighing stability identifier; a light scattering detection unit for outputting scattering pulse peak value, scattering pulse width, and blank sampling scattering intensity; a flow state acquisition unit for outputting a detection cell flow rate sequence and a detection cell pressure sequence; an optical path state acquisition unit for outputting a light source emission intensity sequence and a photoelectric receiving intensity sequence; and a multi-stage time... The intermediate acquisition unit is used to output the device acquisition time, controller reception time, and processor write time; the parameter storage unit is used to store the preset detection parameter table; the processor is connected to the extraction liquid sampling unit, filter membrane sampling and weighing unit, light scattering detection unit, flow status acquisition unit, optical path status acquisition unit, multi-level time acquisition unit, parameter storage unit, and output unit respectively, and is used to receive the data output by each unit, and perform the following operations: binding of the same batch of extraction liquid test data from the same source, generation of verification identifier, calculation of filter membrane particle mass difference, generation of weighing participation identifier, calculation of sampling window characteristics, determination of sampling participation identifier, calculation of light scattering equivalent particle mass, calculation of mass closure difference, generation of mass closure identifier, dual-path reconstruction, and calculation of corrected mass closure difference; the output unit is used to output the original closure output report, the corrected closure output record, or the verification output record.
[0020] Beneficial effects
[0021] The present invention has the following beneficial effects:
[0022] (1) In this invention, the filter membrane sampling completion mark, drying completion mark, cooling completion mark and weighing stability mark are binary encoded, and the difference between the filter membrane termination mass and the filter membrane initial mass is calculated to obtain the filter membrane particle mass difference. This allows the filter membrane weighing result to complete the precondition screening before entering the mass closure calculation, thus avoiding the filter membrane mass data that has not completed sampling, drying, cooling or weighing stability from entering the particulate matter detection result.
[0023] (2) In this invention, the scattering peak value after deduction is obtained by subtracting the blank scattering intensity from the scattering pulse peak value, and the scattering peak value after deduction is statistically assigned to the particle size interval according to the particle size calibration table to generate the particle size interval count. Then, by combining the corrected particle size interval median, the spherical equivalent volume calculation method and the particle density parameter, the light scattering equivalent particle mass is obtained, so that the light scattering output result is converted from the particle number and particle size distribution into mass data that can be compared with the particle mass difference of the filter membrane.
[0024] (3) In this invention, the scattering side quality is corrected by background residual reconstruction and particle size boundary reconstruction. The background residual reconstruction calculates the background residual amount based on the scattering peak value after deduction which is lower than the particle identification threshold. The particle size boundary reconstruction assigns the boundary particles to a lower-level particle size interval based on the particle size interval boundary value and the boundary backtracking coefficient, so that the problem of excessive scattering quality caused by insufficient background deduction and particle size interval assignment can be back-calculated and located.
[0025] (4) In this invention, the sampling side mass is corrected by window compensation reconstruction and large particle size compensation reconstruction. The window compensation reconstruction calculates the undetected volume based on the number of sampling windows with a sampling participation mark of 0. The large particle size compensation reconstruction calculates the number of missing large particle size particles based on the difference between the mass difference of the filter membrane particles and the mass of the light scattering equivalent particles. This allows the problem of excessive weighing caused by the sampling window not participating in the calculation and the large particle size not entering the optical path to be back-calculated and located.
[0026] Of course, any product implementing this invention does not necessarily need to achieve all of the advantages described above at the same time. Attached Figure Description
[0027] Figure 1 This is a flowchart of the method for detecting particulate matter in cleanliness extracts based on light scattering, as described in this invention.
[0028] Figure 2 This is a block diagram of the cleanliness extract particulate matter detection device based on light scattering method of the present invention;
[0029] Figure 3 This is a characteristic diagram of flow rate and pressure state in the sampling window of the present invention. Detailed Implementation
[0030] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.
[0031] Please see Figures 1-3This invention provides a technical solution: a method for detecting particulate matter in cleanliness extracts based on light scattering, comprising: S1, acquiring batch-specific test data of the extract, performing homogeneous binding, generating a verification identifier, and obtaining a batch-specific test dataset of the extract; S2, calculating the mass difference of filter membrane particles based on the batch-specific test dataset of the extract and generating a weighing participation identifier; S3, generating sampling window features based on the batch-specific test dataset of the extract, determining the sampling participation identifier, and generating an optical path flow sampling dataset; S4, calculating the light scattering equivalent particle mass based on the optical path flow sampling dataset and a preset detection parameter table; S5, calculating the mass closure difference and generating a mass closure identifier, performing dual-path reconstruction on the mass closure difference, and generating an original closure output report.
[0032] Specifically, the process of obtaining the batch-specific test data of the extract, performing homology binding, generating verification identifiers, and obtaining the batch-specific test dataset of the extract is as follows:
[0033] The cleanliness of the extract solution is assessed using a particulate matter detection device, which collects batch-specific data for the extract solution. This batch-specific data includes batch identifier, sample identifier, cleaning mode identifier, extract volume, extract temperature, extract settling time, filter membrane sampling data, filter membrane weighing data, light scattering detection data, flow status data, optical path status data, multi-level time data, and a preset detection parameter table. Filter membrane sampling data includes the filter membrane number and sampling completion indicator; filter membrane weighing data includes initial mass, final mass, drying completion indicator, cooling completion indicator, and weighing stability indicator; light scattering detection data includes scattering pulse peak value, scattering pulse width, and blank sampling scattering intensity; flow status data includes the detection cell flow rate sequence and detection cell pressure sequence; optical path status data includes the light source emission intensity sequence and photoelectric receiving intensity sequence; multi-level time data includes device acquisition time, controller reception time, and processor write time; and the preset detection parameter table includes the sampling window duration and peak interval boundaries. The following parameters are included: calibration temperature, particle size range boundary value, boundary retreat coefficient, lower limit threshold for large particle size, first time allowable deviation threshold, second time allowable deviation threshold, upper limit threshold for filter membrane quality difference, lower limit threshold for flow rate, upper limit threshold for flow rate, flow fluctuation threshold, lower limit threshold for pressure, upper limit threshold for pressure, pressure fluctuation threshold, lower limit threshold for light source intensity, upper limit threshold for light source intensity, receiver saturation threshold, blank background threshold, lower limit threshold for temperature, upper limit threshold for temperature, temperature fluctuation threshold, particle identification threshold, upper limit threshold for linear counting, median of particle size range, temperature correction coefficient, particle density parameter, and mass closure threshold.
[0034] Each parameter in the preset detection parameter table is set according to its corresponding detection purpose and corresponding detection conditions. The sampling window duration is used to determine the statistical time range of light scattering data; the first and second allowable time deviation thresholds are used to determine whether the time difference between the device acquisition time and the controller reception time, and between the controller reception time and the processor write time, respectively, meets the requirements. The above time parameters are determined in conjunction with the corresponding data acquisition cycle, actual time difference, and data transmission and processing requirements.
[0035] Peak interval boundaries are used to delineate the intervals corresponding to the scattering pulse peaks, particle size interval boundaries are used to delineate particle size intervals, and the median of the particle size interval is determined based on the upper and lower boundaries of the corresponding particle size interval. The particle identification threshold is used to determine whether the subtracted scattering peaks participate in particle identification, and the linear counting upper limit threshold is used to limit the counting range of particle size intervals entering the calculation of the equivalent particle mass of light scattering. The above parameters and thresholds are determined in conjunction with the scattering pulse peak value, scattering pulse width, blank sampling scattering intensity, and corresponding detection requirements.
[0036] The lower limit threshold, upper limit threshold, flow fluctuation threshold, lower limit threshold, upper limit threshold, and pressure fluctuation threshold are determined based on the operating range of the detection cell and its flow and pressure sequences. The lower limit threshold, upper limit threshold, and receiver saturation threshold are determined based on the light source emission intensity sequence, photoelectric receiving intensity sequence, and corresponding equipment operating requirements. The blank background threshold is determined based on the blank sampling scattering intensity and the scattering signal under blank detection conditions. The lower limit threshold, upper limit threshold, and temperature fluctuation threshold are determined based on the extract temperature and corresponding detection conditions.
[0037] The upper limit threshold for filter membrane mass difference is used to determine whether the mass difference of filter membrane particles is within the allowable mass measurement range. It is determined by combining the initial mass of the filter membrane, the final mass of the filter membrane, and the drying completion marker, cooling completion marker, and weighing stability marker during the filter membrane weighing process. The mass closure threshold is used to determine whether the difference between the filter membrane particle mass difference and the light scattering equivalent particle mass meets the mass closure requirements. Its value should correspond to the filter membrane weighing measurement error, the light scattering equivalent particle mass calculation error, and the detection requirements.
[0038] The lower limit threshold for large particle size is used to determine the particle size range for large particle size compensation. It is determined by combining the boundary values of the particle size interval, the median of the particle size interval, and the calculation process of the equivalent particle mass for light scattering. The particle density parameter is used to convert the equivalent volume of the interval into the equivalent mass of the interval, and is determined based on the material and density data of the particles to be detected.
[0039] The temperature correction factor characterizes the degree of temperature correction applied to the median particle size range when the extract temperature deviates from the calibration temperature. When determining the temperature correction factor, the extract type, the particulate matter being detected, and other detection conditions of the detection device are kept consistent. The median particle size range without temperature correction at the calibration temperature is used as a control. The median particle size range without temperature correction and its changes under other temperature conditions are compared. The temperature correction factor is determined based on the correspondence between temperature changes and changes in the median particle size range.
[0040] The boundary retreat coefficient is used to determine the boundary retreat range of particles near the lower boundary of a particle size interval. The boundary retreat coefficient is determined based on the particle size assignment deviation near the lower boundary of the particle size interval. During calibration, standard particles with known particle sizes are used to obtain corresponding light scattering detection data. The assignment deviation range near the lower boundary of the particle size interval is determined based on the difference between the actual particle size of the standard particles and the equivalent particle size obtained from the detection. The boundary retreat coefficient is determined based on the ratio between the determined assignment deviation range and the corresponding particle size interval width.
[0041] Using the extraction batch identifier and sample identifier in the batch detection data of the same batch of extract as an index, a batch data index record is established. This record includes the extraction batch identifier, sample identifier, filter membrane number, and data source field. Filter membrane sampling data, filter membrane weighing data, and light scattering detection data corresponding to the same batch of extract are bound together based on their common source. A batch data missing identifier is generated when the filter membrane number is empty, the filter membrane sampling completion identifier is 0, the scattering pulse peak value is empty, the scattering pulse width is empty, the initial filter membrane mass is empty, or the final filter membrane mass is empty. A batch data missing identifier is generated when the difference between the device acquisition time and the controller reception time in the batch detection data of the same batch of extract is greater than the first time allowable deviation threshold in the preset detection parameter table, or when the controller reception time and the processor write time are different. When the difference between the input and output times exceeds the second allowable time deviation threshold in the preset detection parameter table, a time difference verification identifier is generated, and the device acquisition time, controller reception time, processor writing time, and corresponding time difference are recorded. When the same extraction batch identifier corresponds to different filter membrane numbers, a filter membrane mixing verification identifier is generated. When the same filter membrane number corresponds to different extraction batch identifiers, a batch attribution verification identifier is generated. Data for which no batch data missing identifier, no time difference verification identifier, no filter membrane mixing verification identifier, and no batch attribution verification identifier are generated are written into the extraction liquid batch detection dataset. The binding relationship between the extraction batch identifier, sample identifier, filter membrane number, filter membrane sampling data, filter membrane weighing data, light scattering detection data, and multi-level time data is retained in the extraction liquid batch detection dataset.
[0042] In this implementation plan, a batch-based data index record is established by using extraction batch identifiers and sample identifiers. Filter membrane sampling data, filter membrane weighing data, and light scattering detection are all bound to the same extraction solution detection object. Combined with key field integrity verification, multi-level time difference verification, and verification of the correspondence between filter membrane number and extraction batch identifier, a batch-based detection dataset of the extraction solution is formed. This ensures that the calculation of filter membrane particle mass difference, light scattering equivalent particle mass, and mass closure difference are all performed based on the same batch of extraction solution data, avoiding deviations in detection results caused by mixing data from different batches, time-displaced data in calculations, and incorrect filter membrane attribution.
[0043] Specifically, the process of calculating the mass difference of filter membrane particles and generating a weighing participation identifier based on the same batch of extraction solution test dataset is as follows:
[0044] The filter membrane sampling completion identifier, drying completion identifier, cooling completion identifier, and weighing stability identifier in the same batch of extraction solution test dataset are encoded using binary codes. A value of 1 indicates that the corresponding step has been completed, and a value of 0 indicates that the corresponding step has not been completed. The filter membrane number has been verified by non-empty verification and batch attribution verification before the formation of the same batch of extraction solution test dataset. The weighing stability identifier is used to indicate that the filter membrane weighing data has reached a readable state. A filter membrane weighing pre-encoding group is formed based on the filter membrane sampling completion identifier, drying completion identifier, cooling completion identifier, and weighing stability identifier. The filter membrane weighing pre-encoding group is used to control whether the filter membrane particle mass difference participates in the subsequent mass closure difference calculation. When the filter membrane sampling completion identifier, drying completion identifier, cooling completion identifier, and weighing stability identifier are all 1, the difference between the filter membrane termination mass and the filter membrane initial mass is calculated. The value is used to obtain the filter membrane particle mass difference. When the filter membrane particle mass difference is greater than or equal to 0 and less than or equal to the upper limit threshold of the filter membrane mass difference, the weighing participation flag is set to 1, and the filter membrane particle mass difference is written into the filter membrane mass difference dataset. Data with the weighing participation flag set to 1 is used for subsequent mass closure difference calculation, and data with the weighing participation flag set to 0 is not used for subsequent mass closure difference calculation. When any of the filter membrane sampling completion flag, drying completion flag, cooling completion flag, and weighing stability flag is 0, the weighing participation flag is set to 0, and a weighing precondition verification flag is generated. When the filter membrane particle mass difference is less than 0, the weighing participation flag is set to 0, and a mass difference direction verification flag is generated. When the filter membrane particle mass difference is greater than the upper limit threshold of the filter membrane mass difference, the weighing participation flag is set to 0, and a mass difference upper bound verification flag is generated.
[0045] In this implementation scheme, a filter membrane weighing pre-encoding group is formed by identifying filter membrane sampling completion, drying completion, cooling completion, and weighing stability. Combined with the direction and upper limit judgment of the filter membrane particle mass difference, it is determined whether the filter membrane particle mass difference is included in the mass closure difference calculation. Filter membrane weighing data that are not sampled, not dried, not cooled, have unreadable weighing data, have negative mass differences, or have mass differences exceeding the upper limit threshold are excluded from the mass closure calculation. This ensures that the filter membrane particle mass difference participating in the closure verification has a clear sampling source, weighing status, and numerical boundary.
[0046] Specifically, the process of calculating the sampling window features based on the batch detection dataset of the extract is as follows:
[0047] Based on the batch-specific dataset of extract solutions, data under the same batch identifier are sorted according to light scattering detection time. The continuous detection time is divided into several sampling windows, with the sampling window duration in the preset detection parameter table used as the window length. Sampling window numbers are generated according to chronological order. Each sampling window contains at least one sample value for the detection cell flow rate, one sample value for the detection cell pressure, one sample value for the light source emission intensity, one sample value for the photoelectric receiving intensity, one sample value for the blank sample scattering intensity, and one sample value for the extract solution temperature. The sequence of detection cell flow rate, detection cell pressure, light source emission intensity, photoelectric receiving intensity, blank sample scattering intensity, and extract solution temperature within the same sampling window is read according to the sampling window number, and then... The window extreme value difference method extracts extreme values and calculates extreme value differences in the sequence to generate sampling window feature data. The window extreme value difference method involves extracting the maximum and minimum values of the sequence data within the same sampling window, and subtracting the minimum value from the maximum value to obtain the fluctuation value of the corresponding sequence. The sampling window feature data includes minimum flow rate, maximum flow rate, flow rate fluctuation value, minimum pressure, maximum pressure, pressure fluctuation value, minimum light source intensity, maximum light source intensity, photoelectric receiving peak value, blank scattering intensity, minimum temperature, maximum temperature, and temperature fluctuation value. Extreme values are extracted from the photoelectric receiving intensity sequence, and the maximum value in the photoelectric receiving intensity sequence is taken as the photoelectric receiving peak value. An arithmetic mean is calculated from the blank sampled scattering intensity sequence, and the arithmetic mean is taken as the blank scattering intensity.
[0048] The flow rate sequence of the detection pool is numerically sorted, with the first value in the sorted sequence taken as the minimum flow rate and the last value taken as the maximum flow rate. The flow rate fluctuation value is obtained by subtracting the minimum flow rate from the maximum flow rate. Similarly, the pressure sequence of the detection pool is numerically sorted, with the first value in the sorted sequence taken as the minimum pressure and the last value taken as the maximum pressure. The pressure fluctuation value is obtained by subtracting the minimum pressure from the maximum pressure. The emission intensity sequence of the light source is numerically sorted, with the first value in the sorted sequence taken as the minimum light source intensity and the last value taken as the maximum light source intensity. The photoelectric receiving intensity sequence is numerically sorted, with the last value taken as the peak photoelectric receiving intensity. The blank sampling scattering intensity sequence is arithmetically averaged to obtain the blank scattering intensity. Finally, the temperature sequence of the extract is numerically sorted, with the first value in the sorted sequence taken as the minimum temperature and the last value taken as the maximum temperature. The temperature fluctuation value is obtained by subtracting the minimum temperature from the maximum temperature.
[0049] like Figure 3 The figure shows the flow rate and pressure characteristics of a sampling window. The horizontal axis represents the sampling window number, the left vertical axis represents the flow rate of the detection pool, and the right vertical axis represents the pressure of the detection pool. The flow rate intervals in the figure represent the range of variation between the minimum and maximum flow rates within the same sampling window, and the pressure intervals represent the range of variation between the minimum and maximum pressure within the same sampling window. The mean flow rate and mean pressure are used to characterize the central level of the flow state within the corresponding sampling window. Figure 3 It is evident that there are differences in the flow rate and pressure ranges between different sampling windows, indicating that the flow state of the extract in the detection cell changes with the sampling window. By jointly observing the flow rate range, pressure range, average flow rate, and average pressure, we can provide data to determine the sampling participation identifier, thereby avoiding the direct inclusion of sampling windows with large fluctuations in flow state into the calculation of light scattering equivalent particle mass.
[0050] In this implementation scheme, the continuous detection time under the same extraction batch identifier is divided into several sampling windows by the sampling window duration. The extreme values, fluctuation values, peak values and average values are extracted from the detection cell flow rate sequence, detection cell pressure sequence, light source emission intensity sequence, photoelectric receiving intensity sequence, blank sampling scattering intensity sequence and extraction liquid temperature sequence within each sampling window to form sampling window feature data. This allows the sampling participation identifier to be judged based on the flow state, optical path state, background scattering state and temperature state within the same sampling window, avoiding sampling windows with flow fluctuations, pressure fluctuations, light source intensity exceeding the limit, photoelectric receiving saturation, high blank background or excessive temperature fluctuations from entering the calculation of light scattering equivalent particle mass.
[0051] Specifically, the process of determining the sampling participation identifier and generating the optical path flow sampling dataset is as follows:
[0052] When the minimum flow rate is greater than or equal to the lower limit threshold of the preset detection parameter table, the maximum flow rate is less than or equal to the upper limit threshold of the preset detection parameter table, the flow fluctuation value is less than or equal to the flow fluctuation threshold of the preset detection parameter table, the minimum pressure is greater than or equal to the lower limit threshold of the preset detection parameter table, the maximum pressure is less than or equal to the upper limit threshold of the preset detection parameter table, the pressure fluctuation value is less than or equal to the pressure fluctuation threshold of the preset detection parameter table, the minimum light source intensity is greater than or equal to the lower limit threshold of the preset detection parameter table, the maximum light source intensity is less than or equal to the upper limit threshold of the preset detection parameter table, and the photoelectric receiving peak value is less than the receiving saturation value of the preset detection parameter table, the following conditions apply: When the threshold, blank scattering intensity is less than or equal to the blank background threshold in the preset detection parameter table, the minimum temperature is greater than or equal to the lower limit threshold in the preset detection parameter table, the maximum temperature is less than or equal to the upper limit threshold in the preset detection parameter table, and the temperature fluctuation value is less than or equal to the temperature fluctuation threshold in the preset detection parameter table, the sampling participation flag is set to 1, and the sampling window number, sampling window feature data, sampling participation flag, and light scattering detection data in the corresponding sampling window are written into the optical path flow sampling dataset; otherwise, the sampling participation flag is set to 0, and a sampling condition verification flag is generated. The sampling condition verification flag is associated with the feature name, corresponding feature value, and corresponding threshold that do not meet the threshold conditions.
[0053] In this implementation scheme, the sampling window feature data is compared with the flow rate threshold, pressure threshold, light source intensity threshold, receiver saturation threshold, blank background threshold, and temperature threshold in the preset detection parameter table to determine whether the sampling window should be written into the optical path flow sampling dataset. This ensures that the sampling window entering the calculation of the light scattering equivalent particle mass simultaneously meets the requirements of flow state, optical path state, background scattering state, and temperature state. For sampling windows that do not meet the threshold conditions, the corresponding feature name, feature value, and threshold are recorded through the sampling condition verification identifier, so that the reason for not participating in the calculation can be associated with the specific sampling window and specific feature field.
[0054] Specifically, based on the optical path sampling dataset and the preset detection parameter table, the specific process for calculating the equivalent particle mass of light scattering is as follows:
[0055] The peak value of the scattered pulse is obtained by subtracting the blank scattering intensity from the peak value of the scattered pulse in the optical path sampling dataset. The peak value of the scattered pulse, the blank scattering intensity, and the peak value of the subtracted scattering are all dimensionless intensity values after photoelectric signal normalization. When the peak value of the subtracted scattering is greater than or equal to the particle identification threshold, interval matching and interval assignment statistics are performed on the peak value of the subtracted scattering according to the preset detection parameter table to generate a particle size interval count. The preset detection parameter table includes the peak interval boundary and the corresponding particle size interval median. The interval matching adopts the left-closed and right-open rule, that is, when the peak value of the subtracted scattering is greater than or equal to the lower boundary of the current peak interval and less than the upper boundary of the current peak interval, it is assigned to the current peak interval; when the peak value of the subtracted scattering is equal to the upper boundary of the maximum peak interval, it is assigned to the maximum peak interval. The particle size interval count is the number of particles assigned to the same particle size interval.
[0056] When the blank scattering intensity is less than or equal to the blank background threshold, the extract temperature is greater than or equal to the lower limit threshold and less than or equal to the upper limit threshold, and the particle size interval count is less than or equal to the upper limit threshold of the linear count, the calculation of the equivalent particle mass for light scattering is initiated; otherwise, an equivalent mass verification identifier is generated. A temperature linear correction method is used, multiplying the median of the particle size interval by the temperature correction factor to obtain the corrected median of the particle size interval. Both the median of the particle size interval and the corrected median of the particle size interval are in micrometers. The temperature correction factor is a dimensionless value, equal to the sum of 1 and the temperature correction coefficient multiplied by the difference between the extract temperature and the calibration temperature. The unit of the temperature correction coefficient is per... The units for Celsius, extraction solution temperature, and calibration temperature are all Celsius. The calibration temperature and temperature correction factor are derived from the preset detection parameter table. The equivalent volume of a single particle is calculated based on the spherical equivalent volume, which is calculated using the formula V=πd³ / 6. The equivalent volume of a single particle is multiplied by the particle size interval count to obtain the interval equivalent volume. The interval equivalent volume is multiplied by the particle density parameter to obtain the interval equivalent mass. The particle density parameter is derived from the preset detection parameter table, with units of milligrams per cubic micrometer, and corresponds to the sample category to be tested. The unit of the interval equivalent mass is milligrams. The interval equivalent mass of each particle size interval is summed to obtain the light scattering equivalent particle mass.
[0057] In this implementation scheme, the blank scattering intensity is subtracted from the peak value of the scattering pulse, and the left-closed and right-open intervals are statistically assigned based on the peak interval boundaries and particle size interval medians in the preset detection parameter table to obtain the particle size interval count. Then, by combining the temperature linear correction method, the spherical equivalent volume calculation method, and the particle density parameter, the dimensionless peak data obtained by light scattering detection is converted into light scattering equivalent particle mass with mass units. This allows the light scattering detection results to be further converted from particle number and particle size distribution into interval equivalent mass and total equivalent mass, providing a data basis with a unified mass dimension for calculating the mass closure difference with the filter membrane particle mass difference.
[0058] Specifically, the process of calculating the mass closure difference and generating the mass closure indicator is as follows:
[0059] When both the weighing participation flag and the sampling participation flag are 1, and the equivalent mass verification flag is not generated, it is determined that both the filter membrane particle mass difference and the light scattering equivalent particle mass meet the mass closure calculation conditions. The absolute value of the difference between the filter membrane particle mass difference and the light scattering equivalent particle mass is calculated to obtain the mass closure difference. When the mass closure difference is less than or equal to the mass closure threshold, the mass closure flag is set to 1; when the mass closure difference is greater than the mass closure threshold, the mass closure flag is set to 0, and a mass closure verification flag is generated. When the light scattering equivalent particle mass is reduced by the filter membrane particle mass difference... When the result is greater than the mass closure threshold, a scattering mass error is generated; when the result of the filter membrane particle mass difference minus the light scattering equivalent particle mass is greater than the mass closure threshold, a weighing mass error is generated; when the weighing participation flag is 0, or the sampling participation flag is 0, or the equivalent mass verification flag has been generated, the mass closure difference is not calculated, the mass closure verification flag is not generated, the mass closure not calculated flag is set to 1, and the reason for not calculating is recorded; the reasons for not calculating include the weighing participation flag being 0, the sampling participation flag being 0, or the equivalent mass verification flag being generated.
[0060] In this implementation scheme, the calculation entry point for the mass closure difference is defined by the weighing participation identifier, the sampling participation identifier, and the equivalent mass verification identifier. The absolute value of the difference between the filter membrane particle mass difference and the light scattering equivalent particle mass is used as the mass closure difference, so that the filter membrane weighing result and the light scattering equivalent mass result can be judged for closure in the same mass dimension. When the mass closure difference exceeds the mass closure threshold, a scattering mass deviation identifier or a weighing mass deviation identifier is generated based on the magnitude relationship between the two types of mass data, providing a clear deviation direction for scattering side reconstruction or sampling side reconstruction. For data that does not meet the preconditions for calculation, the reason for not calculating is recorded by the mass closure not calculated identifier, avoiding confusion between the uncalculated situation and the calculated but not closed situation.
[0061] Specifically, the process of reconstructing the quality closure error using a dual-path approach is as follows:
[0062] When a marker indicating excessive scattering quality has been generated, the scattering-side reconstruction of the equivalent particle quality based on background residual reconstruction and particle size boundary reconstruction is performed to obtain the background residual reconstruction quality and particle size boundary reconstruction quality. Background residual reconstruction involves: extracting a set of peaks below the particle identification threshold from the subtracted scattering peaks; using the average value of this set as the background residual; setting the background residual to 0 when the set is empty; subtracting the background residual from each subtracted scattering peak to obtain the residual subtracted scattering peak; and re-performing interval matching, particle size interval counting statistics, and interval equivalent volume calculation using the residual subtracted scattering peak. The background residual reconstruction quality is obtained by summing the equivalent mass of the intervals. The particle size boundary reconstruction is as follows: based on the particle size interval boundary value and boundary retreat coefficient in the preset detection parameter table, the boundary retreat range is calculated. The boundary retreat range is the product of the current particle size interval width and the boundary retreat coefficient. The boundary retreat coefficient is a dimensionless value. When the equivalent particle size of a single particle is located between the lower boundary of the current particle size interval and the lower boundary plus the boundary retreat range, the particle is assigned to the adjacent low particle size interval. Particles in the smallest particle size interval are not subject to low particle size interval retreat. The particle size interval count is recounted and the interval equivalent mass is summarized to obtain the particle size boundary reconstruction quality. The reconstructed quality of the background residue and the reconstructed quality of the particle size boundary are compared with the mass difference of the filter membrane particles. The reconstructed quality with the smaller difference is selected as the scattering-side correction quality, and the corresponding scattering-side correction path is recorded. When the overweight indicator has been generated, the light scattering equivalent particle mass is lower than the mass difference of the filter membrane particles as the correction object. Based on window compensation reconstruction and large particle size compensation reconstruction, the light scattering missed detection quality is reconstructed on the sampling side to obtain the equivalent mass after window compensation and the equivalent mass after large particle size compensation. Window compensation reconstruction is as follows: count the number of sampling windows with a sampling participation indicator of 0 and the total number of sampling windows, multiply the extraction volume by the proportion of the number of sampling windows with a sampling participation indicator of 0 to the total number of sampling windows to obtain the undetected volume; according to the extraction... The volume of the extract, the total number of sampling windows, and the number of sampling windows with a sampling participation identifier of 1 are taken to calculate the sampling volume. The sampling volume is calculated by multiplying the extract volume by the ratio of the number of sampling windows with a sampling participation identifier of 1 to the total number of sampling windows. The equivalent mass of light-scattering particles is divided by the sampling volume to obtain the equivalent mass per unit volume. The units of the extract volume, the undetected volume, and the sampling volume are consistent, and the unit of the equivalent mass per unit volume is mass units divided by volume units. The undetected volume is multiplied by the equivalent mass per unit volume to obtain the undetected window compensation mass. The undetected window compensation mass is added to the equivalent mass of light-scattering particles to obtain the window compensation reconstruction mass, which is then used as the equivalent mass after window compensation.
[0063] The large particle size compensation reconstruction is as follows: Based on a preset detection parameter table, the lower limit threshold of the large particle size is read, and the particle size intervals whose lower boundary is greater than or equal to the lower limit threshold are determined as large particle size intervals. Within the large particle size intervals, the interval with the largest median value is selected as the maximum particle size compensation interval. Based on the median value of the particle size interval corresponding to the maximum particle size, particle density parameters, and the spherical equivalent volume calculation method, the equivalent mass of a single particle in the maximum particle size interval is obtained. The difference between the filter membrane particle mass difference and the light scattering equivalent particle mass is divided by the equivalent mass of a single particle in the maximum particle size interval to obtain the number of missing large particle sizes. The number of missing large particle sizes is determined to be an integer according to a rounding rule. The smallest integer not less than the result of the division is used; the number of missing large-diameter particles is multiplied by the equivalent mass of a single particle in the largest particle size range to obtain the mass of the missing large-diameter particles; the mass of the missing large-diameter particles is added to the mass of the light scattering equivalent particles to obtain the large-diameter compensation and reconstruction mass, and the large-diameter compensation and reconstruction mass is used as the equivalent mass after large-diameter compensation; the absolute values of the differences between the equivalent mass after window compensation, the equivalent mass after large-diameter compensation, and the mass difference between the filter membrane particles are calculated to obtain the window compensation correction closure error and the large-diameter compensation correction closure error; the equivalent mass after compensation with a smaller correction closure error and less than the original mass closure error is selected as the sampling-side correction mass, and the corresponding sampling-side correction path is recorded.
[0064] In this implementation scheme, insufficient background subtraction is back-calculated through background residual reconstruction, and the boundary particle attribution offset is back-calculated through particle size boundary reconstruction. When the weighing mass is too large, the missed mass corresponding to the sampling window is compensated through window compensation reconstruction, and the mass of large-diameter particles missing in light scattering detection is compensated through large-diameter compensation reconstruction. This allows the mass closure deviation to be processed according to the deviation direction and correction path, and the correction mass and corresponding correction path are determined by comparing the difference between each reconstructed mass and the mass difference of the filter membrane particles.
[0065] Specifically, the process of generating the original closed output report is as follows:
[0066] When the quality closure flag is 1, the equivalent particle mass of light scattering, the particle mass difference of the filter membrane, the particle size interval count, and the quality closure difference are written into the detection output data to generate the original closure output report. When the quality closure verification flag has been generated, the absolute values of the differences between the scattering-side corrected quality, the sampling-side corrected quality, and the particle mass difference of the filter membrane are calculated to obtain the scattering-side corrected closure difference and the sampling-side corrected closure difference. When the scattering-side corrected closure difference is less than or equal to the sampling-side corrected closure difference, the scattering-side corrected quality is taken as the target corrected quality, and the target corrected side is recorded as the scattering side. When the sampling-side corrected closure difference is less than the scattering-side corrected closure difference, the sampling-side corrected quality is taken as the target corrected quality, and the target corrected side is recorded as the sampling side. The corrected quality closure difference is the absolute value of the difference between the particle mass difference of the filter membrane and the target corrected quality. When the corrected quality closure difference is less than the quality closure difference and smaller than the target corrected quality, the corrected quality closure difference is calculated as follows: When the target correction side is equal to or greater than the mass closure threshold, the correction closure output record is written according to the target correction side. When the target correction side is the scattering side, the light scattering equivalent particle mass in the detection output data is replaced with the target correction mass. When the target correction side is the sampling side, the original light scattering equivalent particle mass remains unchanged, and the target correction mass is written into the sampling side compensation equivalent mass field, along with the corrected mass closure difference and the corresponding correction path. When the corrected mass closure difference is greater than the mass closure threshold, or when the corrected mass closure difference is greater than or equal to the mass closure difference, the final corrected particulate matter detection result is not output; only the verification output record is output. The verification output record includes the filter membrane particle mass difference, light scattering equivalent particle mass, scattering side correction mass, sampling side correction mass, scattering side correction closure difference, sampling side correction closure difference, corrected mass closure difference, target correction side, and corresponding correction path.
[0067] In this implementation scheme, the generation conditions of the original closure output report are determined by the quality closure identifier. When the quality closure verification identifier has been generated, the closure difference is calculated for the scattering side correction quality and the sampling side correction quality respectively. The side that is closer to the particle quality difference of the filter membrane is selected as the target correction quality, so that the correction closure output record can simultaneously retain the target correction quality, the target correction side, the corrected quality closure difference, and the corresponding correction path. When the corrected quality closure difference does not decrease or still exceeds the quality closure threshold, only the verification output record is output to avoid the correction result that does not meet the closure condition being output as the final particulate matter detection result.
[0068] Specifically, the second aspect of the present invention provides a cleanliness extract particulate matter detection device based on light scattering, applied to a cleanliness extract particulate matter detection method based on light scattering, comprising: an extract sampling unit for outputting extraction batch identifier, sample identifier, cleaning mode identifier, extract volume, extract temperature, and extract settling time; a filter membrane sampling and weighing unit for outputting filter membrane sampling data, filter membrane number, filter membrane sampling completion identifier, filter membrane initial mass, filter membrane final mass, drying completion identifier, cooling completion identifier, and weighing stability identifier; a light scattering detection unit for outputting scattering pulse peak value, scattering pulse width, and blank sampling scattering intensity; a flow state acquisition unit for outputting detection cell flow sequence and detection cell pressure sequence; an optical path state acquisition unit for outputting light source emission intensity sequence and photoelectric receiving intensity sequence; a multi-level time acquisition unit for outputting device acquisition time, controller receiving time, and processor writing time; a parameter storage unit for storing a preset detection parameter table; and a processor that communicates with the extract sampling unit, filter membrane sampling and weighing unit, light scattering detection unit, and flow state acquisition unit respectively. The processor is connected to an optical path status acquisition unit, a multi-level time acquisition unit, a parameter storage unit, and an output unit. It receives data from each unit and performs the following operations: binding of extraction solution batch detection data to the same source, generating verification identifiers, calculating filter membrane particle mass difference, generating weighing participation identifiers, calculating sampling window characteristics, determining sampling participation identifiers, calculating light scattering equivalent particle mass, calculating mass closure difference, generating mass closure identifiers, dual-path reconstruction, and calculating corrected mass closure difference. The processor also generates and stores batch data missing identifiers, acquisition time difference verification identifiers, filter membrane mixing verification identifiers, batch attribution verification identifiers, weighing participation identifiers, sampling participation identifiers, equivalent mass verification identifiers, mass closure identifiers, mass closure verification identifiers, mass closure not calculated identifiers, scattering mass too large identifiers, and weighing mass too large identifiers. The processor also records correction paths, which include one of the following: background residual reconstruction path, particle size boundary reconstruction path, window compensation reconstruction path, and large particle size compensation reconstruction path. The output unit outputs the original closure output report, the corrected closure output record, or the verification output record. The original closure output report includes light scattering equivalent particle mass, filter membrane particle mass difference, particle size range count, mass closure difference, and mass closure indicator; the calibration closure output record includes target calibration mass, target calibration side, post-calibration mass closure difference, calibration path, and updated detection output data; the verification output record includes filter membrane particle mass difference, light scattering equivalent particle mass, mass closure difference, mass closure verification indicator, mass closure not calculated indicator, scattering side calibration mass, sampling side calibration mass, post-calibration mass closure difference, and calibration path.
[0069] In this implementation scheme, the extraction liquid sampling unit, filter membrane sampling and weighing unit, light scattering detection unit, flow status acquisition unit, optical path status acquisition unit, and multi-level time acquisition unit respectively output the batch data, weighing data, scattering data, flow data, optical path data, and time data required for detection. The parameter storage unit provides a preset detection parameter table, enabling the processor to perform source binding, pre-verification, weighing participation judgment, sampling participation judgment, equivalent mass calculation, mass closure judgment, and dual-path reconstruction based on data sources within the same device. At the same time, the processor stores various verification identifiers, participation identifiers, closure identifiers, overestimation identifiers, and correction paths. The output unit outputs the corresponding fields according to the original closed output report, the corrected closed output record, and the verification output record, so that the detection results, correction results, and verification reasons can be output according to a unified data structure.
[0070] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.
[0071] The preferred embodiments of the present invention disclosed above are merely illustrative of the invention. These preferred embodiments do not exhaustively describe all details, nor do they limit the invention to the specific implementations described. Clearly, many modifications and variations can be made based on the content of this specification. This specification selects and specifically describes these embodiments to better explain the principles and practical applications of the invention, thereby enabling those skilled in the art to better understand and utilize the invention. The invention is limited only by the claims and their full scope and equivalents.
Claims
1. A method for detecting particulate matter in cleanliness extracts based on light scattering, characterized in that, Includes the following steps: S1. Obtain the batch detection data of the extract and perform homologous binding to generate a verification identifier to obtain the batch detection dataset of the extract. S2, based on the same batch of extraction solution test dataset, calculate the mass difference of filter membrane particles and generate weighing participation identifier; S3, based on the batch detection dataset of the extract, generate sampling window features, determine sampling participation identifiers, and generate optical path flow sampling dataset; S4, based on the optical path sampling dataset and the preset detection parameter table, calculate the equivalent particle mass of light scattering; S5 calculates the quality closure difference and generates a quality closure identifier. It then performs dual-path reconstruction on the quality closure difference and generates the original closure output report.
2. The method for detecting particulate matter in cleanliness extracts based on light scattering as described in claim 1, characterized in that: The specific process of obtaining the batch-specific test data of the extract, performing source binding, generating verification identifiers, and obtaining the batch-specific test dataset of the extract is as follows: The cleanliness of the extract is measured by a particulate matter detection device, which collects batch-specific data of the extract. The batch-specific data includes the batch identifier, sample identifier, cleaning mode identifier, extract volume, extract temperature, extract settling time, filter membrane sampling data, filter membrane weighing data, light scattering detection data, flow status data, optical path status data, multi-level time data, and a preset detection parameter table. Using the extraction batch identifier and sample identifier in the batch detection data of the same batch of extract as an index, the filter membrane sampling data, filter membrane weighing data, and light scattering detection data corresponding to the same batch of extract are bound together. When the filter membrane number is empty, the filter membrane sampling completion identifier is 0, the scattering pulse peak value is empty, the scattering pulse width is empty, the filter membrane initial mass is empty, or the filter membrane termination mass is empty in the batch detection data of the same batch of extract, a batch data missing identifier is generated. When the difference between the device acquisition time and the controller receiving time in the batch detection data of the same batch of extract is greater than the first time allowable deviation threshold, or the difference between the controller receiving time and the processor writing time is greater than the second time allowable deviation threshold, an acquisition time difference verification identifier is generated. When the same extraction batch identifier corresponds to different filter membrane numbers, a filter membrane mixing verification identifier is generated. When the same filter membrane number corresponds to different extraction batch identifiers, a batch attribution verification identifier is generated. Data that did not generate batch data missing identifiers, and did not generate acquisition time difference verification identifiers, filter membrane mixing verification identifiers, and batch attribution verification identifiers, will be written into the same batch of extraction solution detection dataset.
3. The method for detecting particulate matter in cleanliness extracts based on light scattering as described in claim 1, characterized in that: The specific process for calculating the mass difference of filter membrane particles and generating a weighing participation identifier based on the same batch of extraction solution test dataset is as follows: For the filter membrane sampling completion, drying completion, cooling completion, and weighing stability indicators in the same batch of extraction solution test data, binary encoding is used. A value of 1 indicates that the corresponding step has been completed, and a value of 0 indicates that the corresponding step has not been completed. When all three indicators are 1, the difference between the final mass and the initial mass of the filter membrane is calculated to obtain the filter membrane particle mass difference. When the filter membrane particle mass difference is greater than or equal to 0 and less than or equal to the upper limit threshold of the filter membrane mass difference, ... The weighing participation flag is set to 1, and the filter membrane particle mass difference is written into the filter membrane mass difference dataset. When any of the filter membrane sampling completion flag, drying completion flag, cooling completion flag, and weighing stability flag is 0, the weighing participation flag is set to 0, and a weighing precondition verification flag is generated. When the filter membrane particle mass difference is less than 0, the weighing participation flag is set to 0, and a mass difference direction verification flag is generated. When the filter membrane particle mass difference is greater than the filter membrane mass difference upper limit threshold, the weighing participation flag is set to 0, and a mass difference upper limit verification flag is generated.
4. The method for detecting particulate matter in cleanliness extracts based on light scattering as described in claim 1, characterized in that: The specific process for calculating the sampling window features based on the same batch of extraction solution detection dataset is as follows: Based on the batch-specific dataset of the extract, the flow rate sequence, pressure sequence, light source emission intensity sequence, photoelectric receiving intensity sequence, blank sampling scattering intensity sequence, and extract temperature sequence within the same sampling window are read according to the sampling window number. The window extreme value difference method is then used to extract extreme values and calculate extreme value differences in the sequences, generating sampling window feature data. The sampling window feature data includes minimum flow rate, maximum flow rate, flow rate fluctuation, minimum pressure, maximum pressure, pressure fluctuation, average flow rate, average pressure, minimum light source intensity, maximum light source intensity, photoelectric receiving peak value, blank scattering intensity, minimum temperature, maximum temperature, and temperature fluctuation. Extreme value extraction is performed on the photoelectric receiving intensity sequence, and the maximum value in the photoelectric receiving intensity sequence is taken as the photoelectric receiving peak value; The arithmetic mean of the blank sampled scattering intensity sequence is taken and used as the blank scattering intensity.
5. The method for detecting particulate matter in cleanliness extracts based on light scattering as described in claim 1, characterized in that: The specific process of determining the sampling participation identifier and generating the optical path flow sampling dataset is as follows: When the minimum flow rate is greater than or equal to the lower flow rate threshold, the maximum flow rate is less than or equal to the upper flow rate threshold, the flow rate fluctuation value is less than or equal to the flow rate fluctuation threshold, the minimum pressure is greater than or equal to the lower pressure threshold, the maximum pressure is less than or equal to the upper pressure threshold, the pressure fluctuation value is less than or equal to the pressure fluctuation threshold, the minimum light source intensity is greater than or equal to the lower light source intensity threshold, the maximum light source intensity is less than or equal to the upper light source intensity threshold, the photoelectric receiving peak value is less than the receiving saturation threshold, the blank scattering intensity is less than or equal to the blank background threshold, the minimum temperature is greater than or equal to the lower temperature threshold, the maximum temperature is less than or equal to the upper temperature threshold, and the temperature fluctuation value is less than or equal to the temperature fluctuation threshold, the sampling participation flag is set to 1 and written into the optical path flow sampling dataset; otherwise, the sampling participation flag is set to 0 and a sampling condition verification flag is generated.
6. The method for detecting particulate matter in cleanliness extracts based on light scattering as described in claim 1, characterized in that: The specific process for calculating the equivalent particle mass of light scattering based on the optical path sampling dataset and the preset detection parameter table is as follows: Subtract the blank scattering intensity from the scattering pulse peak in the optical path sampling dataset to obtain the subtracted scattering peak; when the subtracted scattering peak is greater than or equal to the particle identification threshold, perform interval matching and interval assignment statistics on the subtracted scattering peak according to the preset detection parameter table to generate particle size interval count; When the blank scattering intensity is less than or equal to the blank background threshold, the extract temperature is greater than or equal to the lower temperature threshold and less than or equal to the upper temperature threshold, and the particle size interval count is less than or equal to the linear count upper threshold, the calculation of the light scattering equivalent particle mass begins; otherwise, an equivalent mass verification identifier is generated. Using the temperature linear correction method, the median of the particle size interval is multiplied by the temperature correction factor to obtain the corrected median of the particle size interval. The equivalent volume of a single particle is calculated according to the spherical equivalent volume. The equivalent volume of a single particle is multiplied by the particle size interval count to obtain the interval equivalent volume. The interval equivalent volume is multiplied by the particle density parameter to obtain the interval equivalent mass. The interval equivalent mass of each particle size interval is summed to obtain the light scattering equivalent particle mass.
7. The method for detecting particulate matter in cleanliness extracts based on light scattering as described in claim 1, characterized in that: The specific process for calculating the mass closure difference and generating the mass closure identifier is as follows: When both the weighing participation flag and the sampling participation flag are 1, and the equivalent mass verification flag has not been generated, the absolute value of the difference between the filter membrane particle mass difference and the light scattering equivalent particle mass is calculated to obtain the mass closure difference. When the mass closure difference is less than or equal to the mass closure threshold, the mass closure flag is set to 1. When the mass closure difference is greater than the mass closure threshold, the mass closure flag is set to 0, and a mass closure verification flag is generated. When the result of subtracting the filter membrane particle mass difference from the light scattering equivalent particle mass is greater than the mass closure threshold, a scattering mass excess flag is generated. When the result of subtracting the light scattering equivalent particle mass difference from the filter membrane particle mass difference is greater than the mass closure threshold, a weighing mass excess flag is generated. When the weighing participation flag is 0, or the sampling participation flag is 0, or the equivalent mass verification flag has been generated, the mass closure difference is not calculated, and the mass closure flag is set to 0.
8. The method for detecting particulate matter in cleanliness extracts based on light scattering as described in claim 7, characterized in that: The specific process for reconstructing the quality closure error using a dual-path method is as follows: When an indicator indicating excessive scattering mass has been generated, the scattering-side reconstruction of the equivalent particle mass for light scattering is performed based on background residual reconstruction and particle size boundary reconstruction, yielding the background residual reconstruction mass and particle size boundary reconstruction mass. The background residual reconstruction mass and particle size boundary reconstruction mass are compared with the difference in filter membrane particle mass, and the reconstruction mass with the smaller difference is selected as the scattering-side correction mass, and the corresponding scattering-side correction path is recorded. When an indicator indicating excessive weighing mass has been generated, the correction object is the equivalent particle mass for light scattering being lower than the difference in filter membrane particle mass. The sampling-side reconstruction of the light scattering missed mass is performed based on window compensation reconstruction and large particle size compensation reconstruction, yielding the equivalent mass after window compensation and the equivalent mass after large particle size compensation. The absolute values of the differences between the equivalent mass after window compensation, the equivalent mass after large particle size compensation, and the mass difference between the filter membrane particles are calculated respectively to obtain the window compensation correction closure error and the large particle size compensation correction closure error. The equivalent mass after compensation with a smaller correction closure error than the original mass closure error is selected as the sampling-side correction mass, and the corresponding sampling-side correction path is recorded.
9. The method for detecting particulate matter in cleanliness extracts based on light scattering as described in claim 1, characterized in that: The specific process for generating the original closed output report is as follows: When the mass closure flag is 1, the light scattering equivalent particle mass, filter membrane particle mass difference, particle size range count, and mass closure difference are written into the detection output data to generate the original closure output report. When the quality closure verification mark has been generated, calculate the absolute value of the difference between the scattering side correction quality, the sampling side correction quality and the filter membrane particle quality difference, respectively. Select the correction quality of the side with the smaller difference as the target correction quality, and calculate the corrected quality closure difference. When the corrected quality closure difference is less than the quality closure difference and less than or equal to the quality closure threshold, update the light scattering equivalent particle quality in the detection output data with the target correction quality, and write it into the correction closure output record. When the corrected mass closure error is greater than the mass closure threshold, or when the corrected mass closure error is greater than or equal to the mass closure error, the final corrected particulate matter detection result will not be output; only the verification output record will be output.
10. A device for detecting particulate matter in cleanliness extracts based on light scattering, employing the method for detecting particulate matter in cleanliness extracts based on light scattering as described in any one of claims 1-9, characterized in that, include: The extraction liquid sampling unit is used to output extraction batch identifier, sample identifier, cleaning mode identifier, extraction liquid volume, extraction liquid temperature, and extraction liquid settling time. The filter membrane sampling and weighing unit is used to output filter membrane sampling data, filter membrane number, filter membrane sampling completion indicator, filter membrane initial mass, filter membrane final mass, drying completion indicator, cooling completion indicator, and weighing stability indicator; The light scattering detection unit is used to output the scattering pulse peak value, scattering pulse width, and blank sample scattering intensity; The flow status acquisition unit is used to output the flow rate sequence and pressure sequence of the detection pool; The optical path status acquisition unit is used to output the light source emission intensity sequence and the photoelectric reception intensity sequence; The multi-level time acquisition unit is used to output the device acquisition time, controller reception time, and processor write time; The parameter storage unit is used to store the preset detection parameter table; The processor is connected to the extraction liquid sampling unit, the filter membrane sampling and weighing unit, the light scattering detection unit, the flow state acquisition unit, the optical path state acquisition unit, the multi-level time acquisition unit, the parameter storage unit, and the output unit, respectively. It is used to receive the data output by each unit and perform the following operations: binding of the same batch of extraction liquid test data from the same source, generation of verification identifier, calculation of filter membrane particle mass difference, generation of weighing participation identifier, calculation of sampling window characteristics, determination of sampling participation identifier, calculation of light scattering equivalent particle mass, calculation of mass closure difference, generation of mass closure identifier, dual-path reconstruction, and calculation of corrected mass closure difference. The output unit is used to output the original closed output report, the corrected closed output record, or the verification output record.
Citation Information
Patent Citations
Particle measurement system and method
CN111247418B
Particle sensor
CN112585447B
