Putty powder granularity abnormity early warning system fusing multi-source data

Abstract

The invention relates to the technical field of abnormity early warning, in particular to a putty powder granularity abnormity early warning system fusing multi-source data, which comprises a particle size form extraction module, a form offset identification module, a data weight calibration module, a multi-dimensional risk scoring module and a trend driving grading module. According to the method, the particle size sequences of multiple stages of discharging, mixing, stirring and packaging are combined, the continuous fluctuation characteristic track is constructed based on the time sequence, and the particle size statistical index differences among the procedures are subjected to quantitative comparison, so that the stage identification of the particle size abnormal change trend is realized; meanwhile, performing difference measurement and calculation and correction factor adjustment on a particle size mean value offset section, performing risk quantification by matching particle boundary morphological parameters and particle size variation coefficients, enhancing the multi-dimensional stability of anomaly recognition, and performing driving classification on risk levels through the consistency of a particle size variation trend and a scoring result on the basis of the multi-dimensional stability. And the particle risk identification precision and the timeliness of the early warning result are improved.

Description

Technical Field

[0001] This invention relates to the field of anomaly early warning technology, and in particular to a putty powder particle size anomaly early warning system that integrates multi-source data. Background Technology

[0002] The field of anomaly early warning technology falls under the interdisciplinary application of intelligent monitoring and predictive control. It primarily utilizes methods such as sensor data acquisition, statistical analysis, machine learning modeling, pattern recognition, and multi-dimensional feature correlation calculation to identify and provide early warnings of abnormal states occurring in production processes, equipment operation, quality inspection, or environmental monitoring in real time. Core technologies in this field include multi-source acquisition of anomaly feature data, data noise filtering, trend modeling based on time-series analysis, dynamic threshold setting, anomaly pattern matching, and risk-level response mechanisms. Its applications cover various scenarios such as industrial production quality control, equipment health management, environmental safety monitoring, and energy system stability management, aiming to reduce losses caused by anomalies through intelligent analysis and achieve early intervention and optimized resource allocation.

[0003] The putty powder particle size anomaly early warning system is a quality monitoring and risk management system for the putty powder production process. It is primarily used for real-time detection and early warning of abnormal fluctuations in putty powder particle size during production. The system collects particle size data during production, combines it with set particle size thresholds and statistical analysis to automatically identify deviations from the normal particle size range and issue early warning signals when an anomaly is detected. Its purpose is to improve the quality stability of putty powder production, reduce product defect rates, and help companies respond quickly to anomalies during production, thereby improving overall production efficiency and product consistency.

[0004] Traditional early warning systems lack correlation analysis between multi-channel data from different processes, which can lead to fluctuation anomaly information being limited to a local perspective. In cases of inconsistent particle size changes across multiple stages, it is difficult to accurately identify the source of the anomaly. Furthermore, the lack of quantitative assessment of the dynamic evolution trend of particle size distribution means that some minor trend shifts cannot be responded to in real time. For example, although the particle size fluctuation in the mixing and stirring section may not exceed the threshold of a single point, the edge aggregation phenomenon within a continuous time period may still not be detected in time. This can result in the quality risks not being fully revealed, affecting the comprehensive perception and response to particle anomalies in the production process. Summary of the Invention

[0005] The purpose of this invention is to address the shortcomings of existing technologies by proposing a putty powder particle size anomaly early warning system that integrates multi-source data.

[0006] To achieve the above objectives, the present invention adopts the following technical solution: a putty powder particle size anomaly early warning system integrating multi-source data, the system comprising:

[0007] The particle size extraction module acquires putty powder particle size monitoring data, calls particle size data collected within the same time window, analyzes the distribution of fluctuation intervals within the time period, and combines them according to the time series to form a trajectory, generating a set of particle size statistical feature trajectories.

[0008] The morphological offset recognition module uses the discharge end sensor as the reference channel based on the particle size statistical feature trajectory set. It compares the skewness difference and standard deviation difference in the same time window between the sensors in the mixing zone and the packaging zone to determine the offset state of each channel and generate a channel particle size feature offset mark.

[0009] The data weighting calibration module calls the particle size sequence at the abnormal marker position in the offset channel according to the channel particle size feature offset marker, analyzes the average particle size deviation of the channel in the corresponding time period, applies a correction factor to the sampling point of the offset segment to adjust the value, and generates a calibrated particle size distribution record.

[0010] The multidimensional risk scoring module acquires the grayscale image of the particle image sensor in the mixing zone corresponding to the sampling time period, calls the particle size sequence of the synchronous interval in the calibrated particle size distribution record, calculates the risk score value corresponding to the particle image and particle size synchronization feature, and generates particle risk segment score information.

[0011] The present invention is improved in that the particle size statistical feature trajectory set includes a time series distribution map extracted based on the sampling window, a particle size feature combination structure at each monitoring location, and a three-channel particle size morphology time alignment map. The channel particle size feature offset markers are specifically monitoring channel offset status identification labels, fluctuation anomaly numbers corresponding to offset channels, and morphological index offset records. The calibrated particle size distribution record includes a weighted and corrected particle size sequence, reconstructed interval values ​​of offset segments, and a particle size comparison structure after channel synchronization. The particle risk segment scoring information is specifically a particle image feature aggregation score value, a time period particle size variation combination score, and a risk factor weight mapping table.

[0012] The present invention is improved in that the particle size and morphology extraction module includes:

[0013] The data acquisition submodule acquires putty powder particle size monitoring data, including particle size distribution sequences collected by particle size sensors at the discharge end, mixing zone, and packaging zone. It calls the particle size sampling data of each channel under the same time window and extracts the original particle size point data within three consecutive sampling windows in each channel to generate a multi-channel particle size basic data set.

[0014] The morphology calculation submodule calls the original point data in the sampling window corresponding to each channel according to the multi-channel particle size basic data group, calculates the standard deviation, skewness and range, uses the three statistics as a set of morphological parameters, constructs a particle size morphology time slice sequence according to the channel order, and generates a particle size window statistical parameter sequence.

[0015] The trajectory construction submodule calls the set of morphological parameters arranged by time in each channel according to the sequence of statistical parameters of the particle size window, connects the window morphological change path according to the order of the statistical parameter values, and uniformly constructs a three-channel morphological alignment structure to generate a set of particle size statistical feature trajectories.

[0016] The present invention is improved in that the morphological offset recognition module includes:

[0017] The baseline setting submodule calls the standard deviation and skewness values ​​of the particle size corresponding to the particle size sensor at the discharge end according to the particle size statistical feature trajectory set, and marks them as baseline channel morphology data respectively. It also compares the morphology parameters in the mixing and stirring zone and the packaging zone channel under the same time window to generate a channel comparison morphology parameter group.

[0018] The difference calculation submodule uses the standard deviation and skewness values ​​of the mixing zone and the packaging zone to calculate and obtain the morphological offset value of the offset channel within a unit time window, based on the channel comparison morphological parameter group, and obtains the morphological offset sequence.

[0019] The offset judgment submodule calls the offset value of each channel according to the morphological offset sequence, and performs interval judgment with the set offset tolerance threshold to filter out channels that exceed the limit, assign an identification number, and obtain the channel particle size feature offset mark.

[0020] The present invention is improved in that the data weighting calibration module includes:

[0021] The deviation extraction submodule, based on the channel particle size feature offset marker, calls the particle size sequence of the target channel and the particle size sequence of the reference channel within the same time period at the abnormal marker position, calculates the mean particle size of each channel respectively, performs the difference calculation between the mean particle size of the target channel and the mean particle size of the reference channel, and generates particle size mean deviation information.

[0022] The interval filtering submodule calls the sampling point sequence corresponding to the deviation position in the channel particle size sequence based on the particle size mean deviation information, calculates the three-times standard deviation interval of the channel where each group of sampling points is located as the confidence range, determines whether the particle size mean deviation falls outside the interval range, and filters the sampling point index that deviates from the interval to generate an offset sampling point index set.

[0023] The sampling correction submodule, based on the offset sampling point index set, calls the original particle size value at the corresponding position and the mean value of the reference channel and the mean value of the target channel obtained in the deviation calculation, calculates the ratio of the two as a correction factor, multiplies it with the corresponding value of the offset sampling point, obtains the particle size value sequence after sampling segment correction, and establishes a calibrated particle size distribution record.

[0024] The present invention is improved in that the multidimensional risk scoring module includes:

[0025] The image extraction submodule acquires the grayscale image of the particle image sensor in the mixing zone during the corresponding sampling period, extracts the pixel boundary points of the particle edges in the image, constructs the particle boundary set in sequence, and calculates the contour curvature and edge grayscale gradient of the set to obtain the particle image boundary feature set.

[0026] The feature matching submodule calls the particle size sequence of the corresponding time period in the calibrated particle size distribution record according to the particle image boundary feature set, divides the interval according to the image sampling segment, and combines the contour curvature value and particle size variation coefficient in each interval to obtain the synchronous feature parameter pair under each interval, thus obtaining particle synchronous feature matching information.

[0027] The risk calculation submodule, based on the particle synchronization feature matching information, calls the paired contour curvature value, particle size variation coefficient and edge gradient value in each segment, calculates the degree of combined offset after mapping, calculates the image particle size coupling risk score value of each sampling segment, integrates the risk score values ​​of each segment, and generates particle risk segment score information.

[0028] The present invention has an improvement, wherein the system further includes:

[0029] The trend-driven classification module, based on the particle risk segment scoring information, calls the risk score sequence of each segment and the average slope of particle size in three consecutive time periods in the calibrated particle size distribution record, determines whether the score value is consistent with the direction of change of the particle size slope, divides the warning level label corresponding to each time period, and generates particle risk classification warning output results.

[0030] The particle risk classification and early warning output items include a risk level label sequence, trend score exceeding the limit time point, and an abnormal trend triggering factor record table.

[0031] The present invention is improved in that the trend-driven hierarchical module includes:

[0032] The scoring linkage submodule calls the scoring sequence under each time period and the average particle size in three consecutive time periods in the calibrated particle size distribution record according to the particle risk segment scoring information. It calculates the score change value and the slope of the average particle size between two adjacent time periods in chronological order, and compares the sign of the score change direction with the particle size slope direction to obtain the time period label that is consistent with the score and particle size trend, and generates a scoring trend consistency tag set.

[0033] The trend construction submodule calls the time period number marked as consistent according to the scoring trend consistency mark set, and extracts the score change value and particle size slope amplitude of the time period. It performs the absolute value calculation of the product of the score change value and particle size slope in the segment order to construct the trend factor value sequence and generate segment risk trend information.

[0034] The risk level determination submodule calls the trend factor value and compares it with the risk level threshold value according to the risk trend information of the segment. It obtains the corresponding level label according to the threshold segment, establishes the corresponding risk level number sequence, and generates the granular risk level warning output result.

[0035] Compared with the prior art, the advantages and positive effects of the present invention are as follows:

[0036] In this invention, by combining particle size sequences from multiple stages of material discharge, mixing, and packaging, a continuous fluctuation characteristic trajectory is constructed based on time series. The differences in particle size statistical indicators between each process are quantitatively compared to achieve phased identification of abnormal particle size change trends. At the same time, difference calculation and correction factor adjustment are performed on the particle size mean deviation segment to normalize the particle size distribution in the fluctuation anomaly segment. Furthermore, risk quantification is performed by pairing particle boundary morphology parameters with particle size variation coefficients to enhance the multidimensional stability of anomaly identification. Based on this, the consistency between particle size change trends and scoring results drives the risk level classification, making the warning levels in different time periods continuous and directional, thereby improving the accuracy of particle risk identification and the timeliness of warning results. Attached Figure Description

[0037] Figure 1 This is a system flowchart of the present invention;

[0038] Figure 2 This is a flowchart of the particle size and morphology extraction module of the present invention;

[0039] Figure 3 This is a flowchart of the morphology offset recognition module of the present invention;

[0040] Figure 4 This is a flowchart of the data weighting calibration module of the present invention;

[0041] Figure 5 This is a flowchart of the multidimensional risk scoring module of the present invention;

[0042] Figure 6 This is a flowchart of the trend-driven hierarchical module of the present invention. Detailed Implementation

[0043] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0044] In the description of this invention, it should be understood that the terms "length," "width," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," and "outer," etc., indicating orientation or positional relationships, are based on the orientation or positional relationships shown in the accompanying drawings and are only for the convenience of describing the invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the invention. Furthermore, in the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0045] Please see Figure 1 The present invention provides a technical solution: a putty powder particle size anomaly early warning system that integrates multi-source data, the system including a particle size and morphology extraction module, a morphology offset recognition module, a data weighting calibration module, a multi-dimensional risk scoring module and a trend-driven grading module;

[0046] The particle size extraction module acquires putty powder particle size monitoring data, including particle size distribution sequences collected by particle size sensors at the discharge end, mixing zone, and packaging zone. It calls the particle size data collected within the same time window, calculates the particle size standard deviation, skewness, and range under three consecutive sampling windows, obtains the fluctuation interval distribution information within the time period, and combines them according to the time series to form a trajectory, generating a set of particle size statistical feature trajectories.

[0047] The morphological offset recognition module uses the discharge end sensor as the reference channel based on the particle size statistical feature trajectory set. It compares the difference in skewness and the difference in standard deviation between the sensors in the mixing zone and the packaging zone at the same time window to calculate the morphological offset. Based on the set offset tolerance threshold, it determines whether the morphological offset of each channel exceeds the limit, obtains the identification status of the offset channel, and generates a channel particle size feature offset mark.

[0048] The data weighting calibration module calls the particle size sequence at the abnormal marker position in the offset channel based on the channel particle size feature offset marker, calculates the difference between the particle size mean in the corresponding time window in the reference channel, obtains the particle size mean deviation of the channel in the corresponding time period, and determines whether it coincides with the confidence interval. If it exceeds the interval, a correction factor is applied to the sampling points of the offset segment for numerical adjustment. The correction factor is the ratio of the reference channel mean to the target channel mean, and generates a calibrated particle size distribution record.

[0049] The multidimensional risk scoring module acquires the grayscale image of the particle image sensor in the mixing zone corresponding to the sampling period, calls the particle size sequence of the synchronous interval in the calibrated particle size distribution record, calculates the contour curvature and edge gradient of the particle boundary in the image, and pairs it with the coefficient of variation of particle size in the time period to calculate the risk score value corresponding to the particle image and particle size synchronization feature, and generates particle risk segment scoring information.

[0050] The trend-driven classification module, based on the particle risk segment scoring information, calls the risk score sequence of each segment and the average slope of particle size in three consecutive time periods in the calibrated particle size distribution record, determines whether the score value is consistent with the direction of change of the particle size slope, constructs a trend factor according to the score gradient and slope amplitude, compares it with the risk classification threshold, divides the warning level label corresponding to each time period, and generates particle risk classification warning output results.

[0051] The slope of the average particle size represents the rate of increase or decrease in particle size change per unit time;

[0052] The particle size statistical feature trajectory set includes a time series distribution map extracted based on the sampling window, a particle size feature combination structure at each monitoring location, and a three-channel particle size morphology time alignment map. The channel particle size feature offset markers specifically include monitoring channel offset status identification labels, the fluctuation anomaly number corresponding to the offset channel, and morphological index offset records. The calibrated particle size distribution records include the weighted and corrected particle size sequence, the reconstructed interval value of the offset segment, and the particle size comparison structure after channel synchronization. The particle risk segment scoring information specifically includes particle image feature aggregation score values, time period particle size variation combination scores, and risk factor weight mapping tables. The particle risk classification early warning output items include risk level label sequences, trend score exceeding the limit time points, and abnormal trend triggering factor record tables.

[0053] Please see Figure 2 The particle size and morphology extraction module includes:

[0054] The data acquisition submodule acquires putty powder particle size monitoring data, including particle size distribution sequences collected by particle size sensors at the discharge end, mixing zone, and packaging zone. It calls the particle size sampling data of each channel under the same time window and extracts the original particle size point data within three consecutive sampling windows in each channel to generate a multi-channel particle size basic data set.

[0055] Putty powder particle size monitoring data is obtained using online laser particle size analyzers deployed at the discharge end (channel 0), mixing zone (channel 1), and packaging zone (channel 2). to During the production cycle, The sampling interval is 10 consecutive sampling windows. The data includes particle size data, where each channel collects 5 raw particle size data points within each window, in micrometers (µm). This forms the original particle size sequence shown in Table 1. During execution, the first channel (discharge end) is first called in the first to third consecutive sampling windows. The original particle size data within the range is as follows: Window , Window and Window Subsequently, the raw particle size data from the second channel (mixing zone) within the same three sampling windows are retrieved, specifically... Window , Window and Window Finally, the raw particle size data from the third channel (packaging area) within the same three sampling windows are retrieved, specifically as follows: Window , Window and Window The data sets of all raw points acquired in three consecutive sampling windows from these three channels are used to generate a multi-channel particle size baseline data set.

[0056] Table 1 Original Particle Size Monitoring Data

[0057]

[0058] As shown in Table 1, this table lists the raw particle size data collected by the three channels within three consecutive sampling windows.

[0059] The morphology calculation submodule calls the original point data in the sampling window of each channel according to the multi-channel particle size basic data group, calculates the standard deviation, skewness and range, uses the three statistics as a set of morphological parameters, constructs a particle size morphology time slice sequence according to the channel order, and generates a particle size window statistical parameter sequence.

[0060] Based on the multi-channel particle size baseline dataset, channel 0 is first invoked in the sampling window. Original point data within Calculate its mean as And based on this, the standard deviation value is calculated. The skewness value is obtained by calculating the ratio of the third central moment to the cube of the standard deviation. Calculate the maximum value and minimum value The difference is obtained by the range value. These three statistics As channel 0 in the window The set of morphological parameters, then channel 1 is called in the sampling window. Original point data within Calculate its mean as And based on this, the standard deviation value is calculated. skewness value The range is , constitutes the set of morphological parameters Next, channel 2 is invoked in the sampling window. Original point data within Calculate its mean as And based on this, the standard deviation value is calculated. skewness value The range is , constitutes the set of morphological parameters Following this process, the set of morphological parameters for all channels within all sampling windows is calculated sequentially, and then...

[0061] ;

[0062] By analyzing the channel order and time order, a sequence of particle size morphology time slices consisting of standard deviation, skewness, and range is constructed, generating a sequence of statistical parameters for the particle size window.

[0063] The trajectory construction submodule calls the set of morphological parameters arranged by time in each channel according to the sequence of statistical parameters of the particle size window, connects the window morphological change path according to the order of the statistical parameter values, and uniformly constructs a three-channel morphological alignment structure to generate a set of particle size statistical feature trajectories.

[0064] Based on the particle size window statistical parameter sequence, the morphological parameter set arranged by time in channel 0 is first called, specifically the window... set ,window set and windows set The standard deviations of these three sets , , Connecting them in chronological order forms the standard deviation variation path of channel 0, and the skewness values... , , Connecting them in chronological order forms the skewness variation path of channel 0, which will then represent the range. , , Connecting them chronologically, the range variation path for channel 0 is constructed. Subsequently, using the same method, the morphological parameter sets arranged chronologically in channels 1 and 2 are connected to construct the variation paths for each statistical parameter in channels 1 and 2. Finally, the standard deviation variation paths, skewness variation paths, and range variation paths of the three channels are structurally aligned. Specifically, the standard deviation variation path of channel 0... Path of standard deviation variation with channel 1 and the standard deviation variation path of Channel 2 Alignment is performed along the time dimension, and so on, to perform a unified structured alignment of all morphological parameter paths, generating a set of particle size statistical feature trajectories.

[0065] Please see Figure 3 The shape offset recognition module includes:

[0066] The baseline setting submodule calls the standard deviation and skewness values ​​of the particle size corresponding to the particle size sensor at the discharge end based on the particle size statistical feature trajectory set, and marks them as baseline channel morphology data. It then compares the morphology parameters in the mixing zone and the packaging zone channels under the same time window to generate a channel comparison morphology parameter group.

[0067] Based on the particle size statistical feature trajectory set, the particle size standard deviation trajectory corresponding to the discharge end (channel 0) is called. skewness value trajectory These two trajectories are labeled as baseline channel morphology data. Then, the mixing zone (channel 1) is compared within the first time window. The set of morphological parameters below The set of morphological parameters of the baseline channel within the same time window Then compare the packaging area (channel 2) in the first time window. The set of morphological parameters below The set of morphological parameters of the baseline channel within the same time window Following this method, the morphological parameters of the mixing and packaging zones in all subsequent time windows are compared with the morphological parameters of the baseline channel in the corresponding time window to generate a channel comparison morphological parameter group.

[0068] The difference calculation submodule, based on the channel comparison morphological parameter group, calls the standard deviation and skewness values ​​of the mixing and packaging zones, and combines them with the kurtosis of the particle distribution in each channel, using the following formula:

[0069] ;

[0070] The morphological offset value of the offset channel within a unit time window is obtained through calculation, resulting in a morphological offset sequence.

[0071] in, Indicates the first The morphological offset value of each channel reflects the degree of deviation of that channel from the reference channel in terms of particle size statistical morphology. Indicates the first The normalized value of the standard deviation of particle size in each channel is obtained by dividing the original standard deviation by the maximum standard deviation of that channel under standard operating conditions. This represents the normalized value of the standard deviation of the reference channel particle size. This represents the normalized value indicating the length of the sampling time window, obtained by dividing the current sampling window length by the maximum sampling period. Indicates the first The normalized value of the particle size skewness for each channel is derived from the real-time sampling sequence output by the particle size sensor. It is calculated by dividing the third central moment in the skewness definition formula by the cube of the standard deviation, and then normalizing by dividing by the maximum skewness value over the entire channel cycle. This represents the normalized value of the reference channel particle size skewness. Indicates the first The normalized value of particle size kurtosis for each channel is obtained by dividing the fourth central moment of the particle size sequence by the fourth power of the standard deviation. Then, the kurtosis value is divided by the maximum kurtosis value in the channel's historical operation. The calculation source is a fourth central moment data structure, used to characterize the sharpness of the particle size distribution curve. The offset tolerance threshold is a static preset value, set based on taking all morphological offset sequences of each channel under stable operating conditions. The upper limit obtained by adding three times the mean of the values ​​to its standard deviation is used as the judgment threshold.

[0072] Based on the channel comparison morphological parameter group, the offset tolerance threshold is first set based on the morphological offset values ​​of 100 consecutive sampling windows for each channel under standard operating conditions. This forms an offset sequence containing 200 values. For example, the average offset sequence for channel 1 is... The standard deviation is The mean of the offset sequence in channel 2 is The standard deviation is Then the offset tolerance threshold for channel 1 is set to The offset tolerance threshold for channel 2 is set to This threshold is a static preset value. Subsequently, the mixing zone (channel 1) is invoked in the window. standard deviation Skewness value And combined with the kurtosis value of the particle distribution in this channel. Normalization was performed, where the maximum standard deviation of channel 1 under standard operating conditions was _____. The maximum skewness value is The maximum kurtosis value is Then the normalized standard deviation Normalized skewness value Normalized kurtosis value The reference channel (channel 0) is in the window. The standard deviation is skewness value Its normalization method is the same as that of channel 1, and the maximum value is set to be the same as that of channel 1. , Normalized value of sampling time window length The method for obtaining this is to set the duration of the current sampling window. Divide by the set maximum sampling period duration ,get Substitute these values ​​into the formula:

[0073] ;

[0074] This formula is used to quantify the morphological offset value of the offset channel within a unit time window, where... Indicates the first The shape offset value of each channel. Indicates the first Normalized value of the standard deviation of the particle size of each channel. This represents the normalized value of the standard deviation of the reference channel particle size. The normalized value representing the length of the sampling time window. Indicates the first Normalized value of the particle size skewness of each channel. This represents the normalized value of the reference channel particle size skewness. Indicates the first The normalized value of the kurtosis of each channel particle size is calculated using the following logic: first, the normalized difference of the standard deviation is calculated. and divide by the normalized time window length. The rate of change of standard deviation is obtained, and then multiplied by a factor determined by kurtosis. A logarithmic amplification factor is determined, which amplifies the effect of the rate of change of the standard deviation when the kurtosis deviates from the norm (i.e., the distribution is sharper or flatter). This result is then squared and the absolute difference between it and the skewness-normalized value. The summation and square root calculation combine the rate of change of standard deviation and the absolute difference in skewness—two different dimensions of morphological deviation—to obtain a comprehensive morphological offset value. The calculation process is as follows:

[0075] ;

[0076] The calculation result That is, channel 1 in the window The morphological offset values ​​are then used to calculate the offset values ​​of all channels across all time windows, resulting in a morphological offset sequence. The advantage of this formula lies in the introduction of kurtosis. As a logarithmic amplification factor, when the particle distribution morphology is abnormally sharp or flat, even if the absolute change in the standard deviation is not large, the effect of its rate of change will be significantly amplified. At the same time, combined with the direct difference in skewness, it has a higher sensitivity to potential process anomalies (such as agglomeration or over-grinding).

[0077] The offset judgment submodule calls the offset value of each channel according to the morphological offset sequence, and performs interval judgment with the set offset tolerance threshold to filter out channels that exceed the limit, assign an identification number, and obtain the channel particle size feature offset mark.

[0078] Based on the morphological offset sequence, the mixing zone (channel 1) is invoked in the window. Offset value below And the offset tolerance threshold of channel 1 set above. Perform interval judgment, because If the value falls outside the limit range, the channel is filtered out and assigned an identifier number. For example, channel 1 is in the window. Marked as "Offset -01", the wrapper area (channel 2) is then invoked in the window. The offset value below is set to the calculated value. The offset tolerance threshold of channel 2 Perform interval judgment, because If the value is not in the out-of-limit range, no filtering or marking is performed. Following this process, the offset value of each channel in the morphological offset sequence is judged and filtered one by one to obtain the channel particle size feature offset mark.

[0079] Please see Figure 4 The data reweighting calibration module includes:

[0080] The deviation extraction submodule, based on the channel particle size feature offset marker, calls the particle size sequence of the target channel and the particle size sequence of the reference channel within the same time period at the abnormal marker position, calculates the mean particle size of each channel, performs the difference calculation between the mean particle size of the target channel and the mean particle size of the reference channel, and generates particle size mean deviation information.

[0081] Based on the channel particle size characteristic offset marker, the position marked "offset-01" is called, which is the mixing zone (channel 1) within the time window. Target channel particle size sequence And call the particle size sequence of the baseline channel (channel 0) within the same time period. First, the average particle size of the target channel is calculated. Then calculate the average particle size of the baseline channel. Then, the difference between the target channel average particle size and the reference channel average particle size is calculated, i.e. This generates information on the average particle size deviation.

[0082] The interval filtering submodule calls the sampling point sequence corresponding to the deviation position in the channel particle size sequence based on the particle size mean deviation information, calculates the three-times standard deviation interval of the channel where each group of sampling points is located as the confidence range, determines whether the particle size mean deviation falls outside the interval range, and filters the sampling point index that deviates from the interval to generate an offset sampling point index set.

[0083] Based on the particle size mean deviation information, channel 1 is called in the window. Particle size sequence With channel 0 in the window Particle size sequence First, calculate the standard deviation of the channel 1 sequence. Then its three-standard-deviation interval is ,Right now Then the standard deviation of the channel 0 sequence was calculated as follows: Then its three-standard-deviation interval is ,Right now Then, determine the deviation of the average particle size obtained in the aforementioned steps. Whether it falls outside these two confidence ranges is determined by judging the mean of the target channel. Is it within the confidence interval of the baseline channel? The mean of the inner and base channels. Is it within the confidence interval of the target channel? Inside, due to exist Within the interval, and Also there Within the interval, the mean deviation does not exceed the confidence range. If it does exceed the range, the main sampling points causing the mean deviation are selected. For example, if a sampling point in channel 1 is... If the mean deviates significantly, the index of the sampling point (such as the first point) is selected to generate an offset sampling point index set.

[0084] The sampling correction submodule calls the original particle size value at the corresponding position and the mean value of the reference channel and the mean value of the target channel obtained in the deviation calculation according to the offset sampling point index set. It calculates the ratio of the two as a correction factor, multiplies it with the corresponding value of the offset sampling point, obtains the particle size value sequence after sampling segment correction, and establishes a calibrated particle size distribution record.

[0085] Based on the offset sampling point index set, the selected offset sampling points are set as channel 1 in the window. The first point has an original particle size value of And call the baseline channel mean obtained in the deviation calculation. With the target channel mean First, calculate the ratio of the two as a correction factor, that is... Then, the correction factor is multiplied by the corresponding value of the offset sampling point to obtain the corrected value. Following this process, all indexed offset sampling points are multiplied and adjusted, and the corrected particle size values ​​replace the original values, resulting in a sequence of corrected particle size values ​​for the sampling segment. For example, channel 1 in the window... The sequence is updated to All the corrected sequences were integrated to establish a calibrated particle size distribution record.

[0086] Please see Figure 5 The multidimensional risk scoring module includes:

[0087] The image extraction submodule acquires the grayscale image of the particle image sensor in the mixing zone during the corresponding sampling period, extracts the pixel boundary points of the particle edges in the image, constructs the particle boundary set in sequence, and calculates the contour curvature and edge grayscale gradient of the set to obtain the particle image boundary feature set.

[0088] Based on the grayscale image of the particle image sensor in the mixing zone corresponding to the sampling time period, specifically the window... A picture collected within the time period For a grayscale image of pixels, edge detection is first performed to extract the pixel boundary points of all particles in the image. For a typical particle, its set of boundary points is as follows: The boundary set of the particle is constructed in clockwise order. Then, for this boundary set, the contour curvature of each boundary point is calculated using the three-point method. For example, for point... and its adjacent points and Calculate the geometric angles formed by these points and derive the curvature values ​​to obtain the curvature sequence of the particle. Then, calculate the pixel grayscale gradient at each boundary point, for example, point... The gray value of the particle is 150, and the gray values ​​of the surrounding pixels are 145 and 155 respectively. Therefore, its gradient is 10, and the edge gray gradient sequence of the particle is obtained. The contour curvature sequence of all particles is integrated with the edge gray gradient sequence to obtain the particle image boundary feature set.

[0089] The feature pairing submodule calls the particle size sequence of the corresponding time period in the calibrated particle size distribution record according to the particle image boundary feature set, divides the interval according to the image sampling segment, and combines the contour curvature value and particle size variation coefficient in each interval to obtain the synchronous feature parameter pair under each interval, thus obtaining particle synchronous feature matching information.

[0090] Based on the particle image boundary feature set, channel 1 in the aforementioned calibrated particle size distribution record is called in the window. Particle size sequence over time period The image was divided into five sampling segments, each corresponding to a particle size sampling point. For example, the mean value of the curvature of all particle contours in the first segment was... Then, this value is combined with the particle size variation coefficient calculated from the first particle size sampling point, which is the value obtained by dividing the standard deviation by the mean, to form a synchronous feature parameter pair. Following this process, the contour curvature value and the particle size variation coefficient are combined for each segment in the image to obtain the synchronous feature parameter pair for each segment, thus obtaining the particle synchronous feature matching information.

[0091] The risk calculation submodule, based on particle synchronization feature matching information, calls the paired contour curvature values, particle size variation coefficients, and edge gradient values ​​in each segment to calculate the combined offset degree after mapping, using the formula:

[0092] ;

[0093] The image particle size coupling risk score value of each sampling segment is obtained by calculation, and the risk score values ​​of each segment are integrated to generate particle risk segment score information.

[0094] in, Indicates the first The image grain size coupling risk score for each sampling segment is used to characterize the degree of cooperative offset between the image boundary and the grain size statistical features of that segment. Indicates the first The normalized value of the particle contour curvature of each segment is derived from the ratio of the curvature value derived from the geometric angle constructed by the three-point method of image boundary to the maximum value of all curvatures. Indicates the first The normalized value of the gray-level gradient at the edge of each image segment is derived from the maximum rate of change of image gray-level by dividing the average gradient value of the corresponding image gray-level histogram. Indicates the first The normalized value of the coefficient of variation of particle size for each segment is calculated by dividing the standard deviation of particle size in that segment by the mean particle size and then normalizing to [0, 1]. Indicates the first The normalized value of the image distribution uniformity index for each segment is defined as the ratio of the information entropy value of the gray-level histogram within the image segment to the maximum information entropy, reflecting the density dispersion of particle distribution in that segment of the image. is the base of the natural logarithm;

[0095] Based on particle synchronization feature matching information, the risk grading threshold is set by statistically analyzing the particle risk segment scoring information of a large number of standard production batches. The scoring value sequence is divided according to percentiles. For example, a score value below the 70th percentile is defined as "low risk" (label 1), between the 70th and 95th percentiles is defined as "medium risk" (label 2), and above the 95th percentile is defined as "high risk" (label 3). This establishes the risk grading threshold. Subsequently, the paired contour curvature value in the first segment is used. Particle size variation coefficient Mean of the edge gradient of this section And the image distribution uniformity index (information entropy value) of this segment. Normalization is performed; for example, the maximum value of all curvatures is... The maximum gradient is The maximum coefficient of variation is The maximum information entropy is The normalized value is , , , Substitute these values ​​into the formula:

[0096] ;

[0097] This formula is used to quantify the image grain size coupling risk score for each sampling segment, where, Indicates the first Risk score of each sampling segment, Indicates the first Normalized values ​​of particle profile curvature in each segment Indicates the first The normalized value of the gray-level gradient at the edge of each image segment. Indicates the first Normalized values ​​of the coefficient of variation of particle size in each segment Indicates the first The normalized value of the image distribution uniformity index for each segment. Given the base of the natural logarithm, the formula's operational logic is to use the contour curvature, which characterizes the irregularity of particle morphology, as the base. Edge gradients characterizing the clarity of particle surface states Multiply by, then multiply by a factor determined by the uniformity of the image distribution. The determined exponential amplification factor increases with increasing particle size distribution to amplify the risk score. Finally, this result is divided by a factor proportional to the particle size variation coefficient. The relevant denominator, whose value increases with the more uneven the particle size distribution, plays a moderating role in the risk score. This calculation integrates the microscopic morphology of the particles, macroscopic distribution statistics, and image clarity to obtain a coupled risk score. The calculation process of the example is as follows:

[0098] ;

[0099] The calculation result This is the image grain size coupling risk score for the first sampling segment. Following this process, the risk scores for all segments are calculated and integrated to generate particle risk segment score information. The advantage of this formula lies in the use of an exponential term... The introduction of this factor makes the uniformity of particle distribution in the image a sensitive amplification factor, which can nonlinearly amplify the local aggregation or sparsity problems caused by uneven mixing in the risk score, thus providing a higher risk warning capability than analyzing particle size or morphology alone.

[0100] Please see Figure 6 The trend-driven grading module includes:

[0101] The scoring linkage submodule calls the scoring sequence for each time period and the average particle size in three consecutive time periods in the calibrated particle size distribution record based on the particle risk segment scoring information. It calculates the score change value and the slope of the average particle size between two adjacent time periods in chronological order, and compares the sign of the score change direction with the particle size slope direction to obtain the time period label that is consistent with the score and particle size trend, and generates a scoring trend consistency tag set.

[0102] Based on particle risk segment scoring information, call the window The rating sequence over a time period is set as follows: And call the window in the calibrated particle size distribution record. The average particle size over the three time periods was set as follows: First, calculate the change in scores between two adjacent time periods in chronological order. and Then calculate the slope of the average particle size between two adjacent time periods, i.e. and Then, for the first time period, if the direction of score change is positive and the direction of particle size slope is positive, and the signs are the same, then the time period is marked as consistent. For the second time period, if the direction of score change is negative and the direction of particle size slope is negative, and the signs are the same, then the time period is also marked as consistent. The time period labels that are consistent with the score and particle size trends are obtained, and a score trend consistency label set is generated.

[0103] The trend construction submodule calls the time period number marked as consistent based on the scoring trend consistency tag set, and extracts the score change value and particle size slope amplitude of the time period. It then calculates the absolute value of the product of the score change value and particle size slope in the segment order to construct the trend factor value sequence and generate segment risk trend information.

[0104] Based on the rating trend consistency marker set, the first time period marked as consistent is retrieved, and its rating change value is extracted. With particle size slope amplitude The trend factor value is calculated by performing the absolute value of the product of the score change value and the particle size slope in the segment order. Subsequently, the second marked time period was retrieved, and its rating change value was extracted. With particle size slope amplitude Calculate its trend factor value All calculated trend factor values ​​are arranged in chronological order to construct a trend factor value sequence and generate segment risk trend information.

[0105] The risk level determination submodule calls the trend factor value and compares it with the risk level threshold value according to the risk trend information of the segment. It obtains the corresponding level label according to the threshold segment, establishes the corresponding risk level number sequence, and generates the granular risk level warning output result.

[0106] Based on the segment risk trend information, the first trend factor value is called. And compare it with the aforementioned risk classification threshold value within a range, setting the risk classification threshold as: low risk (0,0.02] Medium risk (0.02,0.08] High risk ,because lie in (0.02,0.08] If the range is specified, the corresponding risk level label is "Medium Risk" (number 2), and then the second trend factor value is called. Compared with the threshold value, because lie in (0,0.02] If the range is specified, the risk level label is "low risk" (number 1). All the obtained risk level labels are arranged in order to establish the corresponding risk level number sequence and generate the particle risk classification early warning output result.

[0107] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention in any other way. Any person skilled in the art may make changes or modifications to the above-disclosed technical content to create equivalent embodiments that can be applied to other fields. However, any simple modifications, equivalent changes, and modifications made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the protection scope of the present invention.

Claims

1. A putty powder particle size anomaly early warning system integrating multi-source data, characterized in that, The system includes: The particle size extraction module acquires putty powder particle size monitoring data, calls particle size data collected within the same time window, analyzes the distribution of fluctuation intervals within the time period, and combines them according to the time series to form a trajectory, generating a set of particle size statistical feature trajectories. The morphological offset recognition module uses the discharge end sensor as the reference channel based on the particle size statistical feature trajectory set. It compares the skewness difference and standard deviation difference in the same time window between the sensors in the mixing zone and the packaging zone to determine the offset state of each channel and generate a channel particle size feature offset mark. The data weighting calibration module calls the particle size sequence at the abnormal marker position in the offset channel according to the channel particle size feature offset marker, analyzes the average particle size deviation of the channel in the corresponding time period, applies a correction factor to the sampling point of the offset segment to adjust the value, and generates a calibrated particle size distribution record. The multidimensional risk scoring module acquires the grayscale image of the particle image sensor in the mixing zone corresponding to the sampling time period, calls the particle size sequence of the synchronous interval in the calibrated particle size distribution record, calculates the risk score value corresponding to the particle image and particle size synchronization feature, and generates particle risk segment score information.

2. The putty powder particle size anomaly early warning system integrating multi-source data according to claim 1, characterized in that, The particle size statistical feature trajectory set includes a time series distribution map extracted based on the sampling window, a particle size feature combination structure at each monitoring location, and a three-channel particle size morphology time alignment map. The channel particle size feature offset markers specifically include a monitoring channel offset status identification label, a fluctuation anomaly number corresponding to the offset channel, and a morphological index offset record. The calibrated particle size distribution record includes a weighted and corrected particle size sequence, a reconstructed interval value of the offset segment, and a particle size comparison structure after channel synchronization. The particle risk segment scoring information specifically includes a particle image feature aggregation score value, a time period particle size variation combination score, and a risk factor weight mapping table.

3. The putty powder particle size anomaly early warning system integrating multi-source data according to claim 2, characterized in that, The particle size and morphology extraction module includes: The data acquisition submodule acquires putty powder particle size monitoring data, including particle size distribution sequences collected by particle size sensors at the discharge end, mixing zone, and packaging zone. It calls the particle size sampling data of each channel under the same time window and extracts the original particle size point data within three consecutive sampling windows in each channel to generate a multi-channel particle size basic data set. The morphology calculation submodule calls the original point data in the sampling window corresponding to each channel according to the multi-channel particle size basic data group, calculates the standard deviation, skewness and range, uses the three statistics as a set of morphological parameters, constructs a particle size morphology time slice sequence according to the channel order, and generates a particle size window statistical parameter sequence. The trajectory construction submodule calls the set of morphological parameters arranged by time in each channel according to the sequence of statistical parameters of the particle size window, connects the window morphological change path according to the order of the statistical parameter values, and uniformly constructs a three-channel morphological alignment structure to generate a set of particle size statistical feature trajectories.

4. The putty powder particle size anomaly early warning system integrating multi-source data according to claim 3, characterized in that, The morphological offset recognition module includes: The baseline setting submodule calls the standard deviation and skewness values ​​of the particle size corresponding to the particle size sensor at the discharge end according to the particle size statistical feature trajectory set, and marks them as baseline channel morphology data respectively. It also compares the morphology parameters in the mixing and stirring zone and the packaging zone channel under the same time window to generate a channel comparison morphology parameter group. The difference calculation submodule uses the standard deviation and skewness values ​​of the mixing zone and the packaging zone to calculate and obtain the morphological offset value of the offset channel within a unit time window, based on the channel comparison morphological parameter group, and obtains the morphological offset sequence. The offset judgment submodule calls the offset value of each channel according to the morphological offset sequence, and performs interval judgment with the set offset tolerance threshold to filter out channels that exceed the limit, assign an identification number, and obtain the channel particle size feature offset mark.

5. The putty powder particle size anomaly early warning system integrating multi-source data according to claim 4, characterized in that, The data weighting calibration module includes: The deviation extraction submodule, based on the channel particle size feature offset marker, calls the particle size sequence of the target channel and the particle size sequence of the reference channel within the same time period at the abnormal marker position, calculates the mean particle size of each channel respectively, performs the difference calculation between the mean particle size of the target channel and the mean particle size of the reference channel, and generates particle size mean deviation information. The interval filtering submodule calls the sampling point sequence corresponding to the deviation position in the channel particle size sequence based on the particle size mean deviation information, calculates the three-times standard deviation interval of the channel where each group of sampling points is located as the confidence range, determines whether the particle size mean deviation falls outside the interval range, and filters the sampling point index that deviates from the interval to generate an offset sampling point index set. The sampling correction submodule, based on the offset sampling point index set, calls the original particle size value at the corresponding position and the mean value of the reference channel and the mean value of the target channel obtained in the deviation calculation, calculates the ratio of the two as a correction factor, multiplies it with the corresponding value of the offset sampling point, obtains the particle size value sequence after sampling segment correction, and establishes a calibrated particle size distribution record.

6. The putty powder particle size anomaly early warning system integrating multi-source data according to claim 5, characterized in that, The multidimensional risk scoring module includes: The image extraction submodule acquires the grayscale image of the particle image sensor in the mixing zone during the corresponding sampling period, extracts the pixel boundary points of the particle edges in the image, constructs the particle boundary set in sequence, and calculates the contour curvature and edge grayscale gradient of the set to obtain the particle image boundary feature set. The feature matching submodule calls the particle size sequence of the corresponding time period in the calibrated particle size distribution record according to the particle image boundary feature set, divides the interval according to the image sampling segment, and combines the contour curvature value and particle size variation coefficient in each interval to obtain the synchronous feature parameter pair under each interval, thus obtaining particle synchronous feature matching information. The risk calculation submodule, based on the particle synchronization feature matching information, calls the paired contour curvature value, particle size variation coefficient and edge gradient value in each segment, calculates the degree of combined offset after mapping, calculates the image particle size coupling risk score value of each sampling segment, integrates the risk score values ​​of each segment, and generates particle risk segment score information.

7. The putty powder particle size anomaly early warning system integrating multi-source data according to claim 6, characterized in that, The system also includes: The trend-driven classification module, based on the particle risk segment scoring information, calls the risk score sequence of each segment and the average slope of particle size in three consecutive time periods in the calibrated particle size distribution record, determines whether the score value is consistent with the direction of change of the particle size slope, divides the warning level label corresponding to each time period, and generates particle risk classification warning output results. The particle risk classification and early warning output items include a risk level label sequence, trend score exceeding the limit time point, and an abnormal trend triggering factor record table.

8. The putty powder particle size anomaly early warning system according to claim 7, characterized in that, The trend-driven hierarchical module includes: The scoring linkage submodule calls the scoring sequence under each time period and the average particle size in three consecutive time periods in the calibrated particle size distribution record according to the particle risk segment scoring information. It calculates the score change value and the slope of the average particle size between two adjacent time periods in chronological order, and compares the sign of the score change direction with the particle size slope direction to obtain the time period label that is consistent with the score and particle size trend, and generates a scoring trend consistency tag set. The trend construction submodule calls the time period number marked as consistent according to the scoring trend consistency mark set, and extracts the score change value and particle size slope amplitude of the time period. It performs the absolute value calculation of the product of the score change value and particle size slope in the segment order to construct the trend factor value sequence and generate segment risk trend information. The risk level determination submodule calls the trend factor value and compares it with the risk level threshold value according to the risk trend information of the segment. It obtains the corresponding level label according to the threshold segment, establishes the corresponding risk level number sequence, and generates the granular risk level warning output result.

Images