Emulsified puffed food processing quality monitoring method

Abstract

The invention relates to the technical field of data processing, in particular to an emulsified puffed food processing quality monitoring method. Obtaining the moisture suspected abnormal degree of the target batch according to the moisture difference characteristics of the target batch and other batches; obtaining a suspected abnormal batch according to the moisture suspected abnormal degree; clustering is carried out according to the distribution characteristics of the suspected abnormal batches, and moisture abnormal distribution characteristic values of clusters are obtained according to interval characteristics of the suspected abnormal batches and adjacent suspected abnormal batches in the clusters; and obtaining an abnormal cluster according to the moisture abnormal distribution characteristic value. According to the method, the moisture change anomaly degree is obtained according to the moisture and hardness change similar characteristics and the moisture and puffing degree change similar characteristics of suspected abnormal batches in abnormal clusters; the real abnormal batch of the emulsified puffed food is obtained according to the abnormal degree of the moisture change, so that the accuracy of moisture monitoring and the processing quality of the puffed food are improved.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, specifically to a method for monitoring the processing quality of emulsified and extruded grains. Background Technology

[0002] Emulsified puffed grains are composite grain products made by rapidly puffing raw materials under high temperature and pressure, characterized by good emulsification stability and a fluffy, crisp texture. During production, quality monitoring is crucial for ensuring the consistency and safety of puffed grain products. Current technologies typically monitor indicators such as moisture content, hardness, and degree of puffing during processing, and judge the presence of quality problems based on empirical values. Abnormal moisture content is a core issue affecting puffing effect, finished product quality, and storage stability, making moisture monitoring paramount. However, the high temperature and humidity environment in actual emulsified puffed grain processing can easily cause data drift in infrared moisture analyzers, leading to errors between the monitored moisture value and the actual moisture content, ultimately affecting the accuracy of moisture monitoring and the processing quality of the emulsified puffed grains. Summary of the Invention

[0003] To address the technical problem mentioned above where moisture monitoring instruments may experience data drift due to high temperature and humidity environments in emulsified and extruded grain processing, affecting the accuracy of moisture monitoring and processing quality, the present invention aims to provide a method for monitoring the processing quality of emulsified and extruded grains. The specific technical solution adopted is as follows: Obtain the moisture content, hardness, and degree of puffing of consecutive batches of emulsified puffed grains from the same processing stage; The degree of suspected moisture abnormality of the target batch is obtained based on the difference in moisture content between the target batch and other batches; suspected abnormal batches are then identified based on the degree of suspected moisture abnormality. Clustering is performed based on the distribution characteristics of the suspected abnormal batches to obtain different clusters; the moisture abnormality distribution characteristic value of the cluster is obtained based on the interval characteristics between the suspected abnormal batches and adjacent suspected abnormal batches in the cluster; and abnormal clusters are obtained based on the moisture abnormality distribution characteristic value. The degree of moisture change abnormality is obtained based on the similarity characteristics of moisture and hardness changes and the similarity characteristics of moisture and puffing degree changes of suspected abnormal batches in the abnormal clusters; the true abnormal batches of emulsified puffed grains are obtained based on the degree of moisture change abnormality.

[0004] Furthermore, the step of obtaining the degree of suspected moisture abnormality of the target batch based on the moisture difference characteristics between the target batch and other batches includes: Calculate and normalize the sum of the absolute values ​​of the moisture differences between the target batch and the two adjacent batches to obtain the adjacent moisture difference degree; calculate and normalize the absolute value of the difference between the moisture of the target batch and the average moisture of all batches to obtain the comprehensive moisture difference degree; calculate the average of the adjacent moisture difference degree and the comprehensive moisture difference degree to obtain the suspected abnormality degree of moisture of the target batch.

[0005] Furthermore, the step of obtaining the suspected abnormal batch based on the degree of suspected moisture abnormality includes: The batches whose moisture content is suspected to be abnormal and exceeds a preset first threshold are designated as the suspected abnormal batches.

[0006] Furthermore, the step of clustering based on the distribution characteristics of the suspected abnormal batches to obtain different clusters includes: The batch distribution of suspected abnormal batches is clustered using the K-means clustering algorithm to obtain different clusters.

[0007] Further, the step of obtaining the moisture anomaly distribution characteristic value of the cluster based on the interval characteristics between suspected abnormal batches and adjacent suspected abnormal batches in the cluster includes: The following steps are performed: First, calculate the sum of the number of suspected anomalous batches and the interval batches between adjacent suspected anomalous batches in the cluster to obtain the adjacent interval characteristic value of the suspected anomalous batches. Second, calculate the average of the adjacent interval characteristic values ​​of all suspected anomalous batches in the cluster to obtain a first value. Third, calculate the reciprocal of the first value to obtain the density. Fourth, calculate the reciprocal of the average of the absolute values ​​of the differences between all adjacent interval characteristic values ​​and the first value to obtain the uniformity. Fifth, calculate the sum of the density and the uniformity to obtain the moisture anomalous distribution characteristic value of the cluster.

[0008] Further, the step of obtaining the abnormal clusters based on the abnormal moisture distribution characteristic values ​​includes: Clusters whose abnormal moisture distribution characteristic values ​​exceed a preset second threshold are designated as abnormal clusters.

[0009] Further, the step of obtaining the degree of moisture change anomaly based on the similarity characteristics of moisture and hardness changes and the similarity characteristics of moisture and expansion degree changes of suspected abnormal batches in the abnormal cluster includes: Construct corresponding moisture, hardness, and puffiness sequences according to the batch order of suspected abnormal batches in the abnormal cluster; calculate the Pearson correlation coefficient between the moisture sequence and the hardness sequence and perform a positive correlation mapping to obtain a first degree of correlation; calculate the Pearson correlation coefficient between the moisture sequence and the puffiness sequence and perform a negative correlation mapping to obtain a second degree of correlation; calculate the sum of the first degree of correlation and the second degree of correlation to obtain the degree of abnormality of moisture change in the abnormal cluster.

[0010] Furthermore, the step of obtaining the true abnormal batches of emulsified extruded grain based on the degree of moisture change anomaly includes: Suspected abnormal batches in the abnormal clusters where the moisture change anomaly exceeds a preset third threshold are considered as real abnormal batches in the emulsified and extruded grain processing process.

[0011] The present invention has the following beneficial effects: In this invention, obtaining the degree of suspected moisture anomaly can characterize whether the moisture monitoring value of the target batch is abnormal, and obtaining suspected abnormal batches can preliminarily identify batches that may have moisture anomalies. Obtaining clusters can be used to further analyze the causes of suspected abnormal batches; obtaining the characteristic value of moisture anomaly distribution can determine the possibility that the suspected abnormal batches are affected by water vapor in the sensor, improving the accuracy of moisture monitoring; obtaining abnormal clusters can reduce the possibility of monitoring errors caused by water vapor affecting the sensor during moisture monitoring. Since sensor monitoring is prone to data drift in the high temperature and high humidity environment of the puffing process, and there is a correlation between moisture, hardness, and degree of puffing during processing, obtaining the degree of moisture change anomaly can further determine the possibility that the abnormal clusters in the moisture monitoring process are affected by sensor data drift; finally, based on the degree of moisture change anomaly, the true abnormal batches of emulsified puffed grain are obtained, reducing the impact of sensor error on the monitoring results during the puffed grain monitoring process, improving the accuracy of moisture monitoring and the processing quality of puffed grain. Attached Figure Description

[0012] To more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0013] Figure 1 This is a flowchart of a method for monitoring the processing quality of emulsified and puffed grains, provided as an embodiment of the present invention. Detailed Implementation

[0014] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of a method for monitoring the processing quality of emulsified and extruded grains according to the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0015] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0016] The following description, in conjunction with the accompanying drawings, details a specific scheme for a method for monitoring the processing quality of emulsified and puffed grains provided by this invention.

[0017] Please see Figure 1 The diagram illustrates a flowchart of a method for monitoring the processing quality of emulsified and extruded grains according to an embodiment of the present invention. The method includes the following steps: Step S1: Obtain the moisture content, hardness, and degree of puffing of emulsified extruded grains from consecutive batches during the same processing.

[0018] In this embodiment of the invention, the implementation scenario involves monitoring the processing of emulsified extruded grains to improve monitoring accuracy and processing quality. First, the moisture content, hardness, and degree of expansion of consecutive batches of emulsified extruded grains within the same processing stage are acquired. During the extruded grain processing, the grains are processed in multiple batches continuously. After each batch completes the extrusion and drying process, the moisture content of that batch is monitored using an infrared moisture meter at the outlet of the extrusion equipment. Simultaneously, the hardness and degree of expansion of each batch are acquired in subsequent processing steps. In this embodiment, the same processing stage includes 100 batches of extruded grains, with each batch corresponding to a set of data on moisture content, hardness, and degree of expansion.

[0019] Step S2: Obtain the degree of suspected moisture abnormality of the target batch based on the difference characteristics of moisture between the target batch and other batches; obtain the suspected abnormal batches based on the degree of suspected moisture abnormality.

[0020] While the moisture content of extruded grains will not be exactly the same across different batches during processing, it will fluctuate within a normal range. When abnormalities occur during the extrusion and drying process, such as unstable temperature control of the heat source system or unstable extrusion pressure, moisture cannot evaporate effectively, leading to fluctuations in the moisture content of different batches, with differences between adjacent batches exceeding the normal range. Furthermore, abnormalities in temperature and pressure control of the extrusion equipment are relatively rare, and batches with abnormal moisture content occur less frequently than normal batches, showing a certain difference from normal moisture values. Therefore, the degree of suspected moisture abnormality in a target batch can be determined by comparing its moisture content with that of other batches.

[0021] Preferably, in this embodiment of the invention, the step of obtaining the suspected moisture abnormality degree includes: calculating and normalizing the sum of the absolute values ​​of the moisture differences between the target batch and the two adjacent batches to obtain the adjacent moisture difference degree; the larger the adjacent moisture difference degree, the greater the moisture difference between the target batch and the two adjacent batches, and the more likely the moisture of the target batch is to be abnormal; it should be noted that the target batch refers to any batch that meets the calculation conditions in the same processing process. Calculating and normalizing the absolute value of the difference between the moisture of the target batch and the average moisture of all batches to obtain the comprehensive moisture difference degree; the larger the comprehensive moisture difference degree, the greater the difference between the moisture of the target batch and the average moisture of all batches, and the more likely the moisture of the target batch is to be abnormal. It should be noted that the normalization method in this embodiment of the invention is linear normalization. Calculating the average value of the adjacent moisture difference degree and the comprehensive moisture difference degree to obtain the suspected moisture abnormality degree of the target batch. The larger the suspected moisture abnormality degree, the more likely the target batch is to have abnormal moisture content in the extruded grain. The step of obtaining the suspected moisture abnormality degree includes:

[0022] In the formula, R represents the suspected abnormality in moisture content of the target batch. This indicates normalization, and W represents the moisture content of the target batch. This indicates the moisture content of the batch immediately preceding the target batch. This indicates the moisture content of the next batch immediately following the target batch, where L represents the average moisture content across all batches. Indicates the degree of difference in moisture content between adjacent areas. This indicates the overall moisture content variation.

[0023] Furthermore, suspected abnormal batches can be obtained based on the degree of suspected moisture abnormality. Specifically, this includes classifying target batches whose degree of suspected moisture abnormality exceeds a preset first threshold as suspected abnormal batches. In this embodiment of the invention, the value range of the degree of suspected moisture abnormality of the target batch after linear normalization is 0 to 1. The more abnormal the moisture content of the target batch, the closer the degree of suspected moisture abnormality is to 1; the more normal the moisture content of the target batch, the closer the degree of suspected moisture abnormality is to 0. These two cases are distributed at opposite ends of the value range. Therefore, in this embodiment of the invention, the preset first threshold is 0.5, thereby filtering out batches with potentially abnormal moisture content. The implementer can determine the preset first threshold according to the implementation scenario.

[0024] Step S3: Cluster the suspected abnormal batches according to their distribution characteristics to obtain different clusters; obtain the moisture abnormality distribution characteristic value of the clusters according to the interval characteristics between the suspected abnormal batches and adjacent suspected abnormal batches in the clusters; obtain the abnormal clusters according to the moisture abnormality distribution characteristic value.

[0025] Moisture content is obtained using an infrared moisture analyzer. Therefore, besides actual moisture anomalies in extruded grains, the moisture value may also be affected by the sensor's own condition. Under the high-temperature and high-pressure extrusion environment, water vapor may obstruct the infrared path, causing fluctuations in the values ​​monitored by the infrared moisture analyzer. This can result in suspected abnormal batches with non-genuine moisture anomalies, affecting the accuracy of extruded grain processing quality monitoring. To further distinguish whether the moisture content of suspected abnormal batches is truly abnormal, it is necessary to analyze the distribution characteristics of these suspected abnormal batches. Because sensor errors are affected by water vapor, the probability and degree of error are greater when water vapor concentration is high. Furthermore, the expansion path of water vapor is relatively random and not fixed in any one area of ​​the equipment. Therefore, batches with sensor errors mainly caused by steam concentration exhibit randomness and a significant sporadic characteristic. Genuine moisture anomalies during the extrusion process are caused by unstable temperature or pressure control. When an abnormality occurs in a batch due to equipment malfunction, it indicates a problem with the equipment. Consequently, the probability of abnormalities occurring in subsequent batches increases, leading to a larger number of batches with moisture anomalies and closer batch intervals. Therefore, clustering is performed based on the distribution characteristics of suspected abnormal batches to obtain different clusters. Preferably, in this embodiment of the invention, the batch distribution of suspected abnormal batches is clustered using the K-means clustering algorithm to obtain different clusters. It should be noted that the K-means clustering algorithm is existing technology, and the specific steps will not be described in detail. Clustering can distinguish different batches of suspected abnormal batches and divide them into different clusters, with the batch intervals of suspected abnormal batches within each cluster being relatively close.

[0026] Furthermore, the moisture distribution characteristic value of a cluster can be obtained based on the interval characteristics between suspected abnormal batches and adjacent suspected abnormal batches within the cluster. Preferably, in this embodiment of the invention, the step of obtaining the moisture distribution characteristic value includes: calculating the sum of the number of suspected abnormal batches and the interval batches between adjacent suspected abnormal batches in the cluster to obtain the adjacent interval characteristic value of the suspected abnormal batch; the smaller the adjacent interval characteristic value, the closer the batch interval of the suspected abnormal batches in the cluster, and the more likely it is that the extrusion equipment is malfunctioning. The average value of the adjacent interval characteristic values ​​of all suspected abnormal batches in the cluster is calculated to obtain a first value; the reciprocal of the first value is calculated to obtain the density; the smaller the density, the larger the batch interval, and the more likely it is that the suspected abnormal batches are caused by the sensor being affected by water vapor. The reciprocal of the average of the absolute values ​​of the differences between all adjacent interval characteristic values ​​and the first value is calculated to obtain the uniformity; the smaller the uniformity, the more random the batch distribution of the suspected abnormal batches in the cluster, and the more likely it is that the suspected abnormal batches are caused by the sensor being affected by water vapor. The sum of density and uniformity is calculated to obtain the moisture anomaly distribution characteristic value of the cluster. A smaller moisture anomaly distribution characteristic value indicates that the suspected abnormal batches within the cluster are more likely due to inaccurate moisture monitoring caused by the sensor being affected by water vapor. Furthermore, abnormal clusters can be obtained based on the moisture anomaly distribution characteristic value, specifically by classifying clusters whose moisture anomaly distribution characteristic values ​​exceed a preset second threshold as abnormal clusters. In this embodiment, the preset second threshold is 0.6, which can be determined by the implementer according to the implementation scenario. Suspected abnormal batches within abnormal clusters are more likely to be genuine moisture anomalies.

[0027] Step S4: Obtain the degree of moisture change anomaly based on the similarity characteristics of moisture and hardness changes and the similarity characteristics of moisture and puffing degree changes of suspected abnormal batches in the abnormal cluster; obtain the true abnormal batches of emulsified puffed grain based on the degree of moisture change anomaly.

[0028] In the high-temperature, high-pressure, and excessively humid puffing environment, infrared moisture analyzers may experience data drift. This is because the sensor is affected by the high-temperature and high-humidity environment, impacting its monitoring accuracy. Data drift typically exhibits a persistent characteristic, occurring in clusters of consecutive batches, making it easily identified as a genuine moisture anomaly. Therefore, suspected anomalous batches in anomaly clusters may be caused by sensor data drift. To further improve moisture monitoring and the processing quality of puffed grains, it is necessary to analyze both hardness and puffing degree. Moisture plays a positive role in driving puffing; as water content increases, starch granules absorb water and swell, increasing the absorbed frictional heat and shear force, thus increasing starch maturity and puffing degree. Therefore, when the moisture value of puffed grain increases, its puffing degree should also increase. As moisture content increases, the hardness of puffed grains tends to decrease. During puffing, moisture provides a medium for the breakdown of starch granules; after absorbing water, the starch granules swell, leading to a looser granule structure and reduced hardness. Therefore, when the moisture content of puffed grain increases, its corresponding hardness decreases and its puffing degree increases; however, the moisture anomaly caused by sensor data drift is unrelated to hardness and puffing degree. Thus, the degree of moisture anomaly can be obtained based on the similarity characteristics of moisture and hardness changes and the similarity characteristics of moisture and puffing degree changes in suspected abnormal batches in the abnormal cluster.

[0029] Preferably, in this embodiment of the invention, the step of obtaining the degree of moisture change anomaly includes: constructing corresponding moisture, hardness, and puffiness sequences according to the batch order of suspected abnormal batches in the abnormal cluster; calculating the Pearson correlation coefficient between the moisture sequence and the hardness sequence and mapping them negatively to obtain a first degree of correlation; it should be noted that the Pearson correlation coefficient is existing technology, with a value range of -1 to 1. The more positively correlated the changes of the two sequences are, the closer the Pearson correlation coefficient is to 1; the more negatively correlated the changes of the two sequences are, the closer the Pearson correlation coefficient is to -1. The greater the negative correlation between the changes of the moisture sequence and the hardness sequence, the greater the first degree of correlation, and the more likely the suspected abnormal batches in the abnormal cluster are caused by real moisture anomalies. Calculating the Pearson correlation coefficient between the moisture sequence and the puffiness sequence and mapping them positively to obtain a second degree of correlation; the greater the positive correlation between the changes of the moisture sequence and the puffiness sequence, the greater the second degree of correlation, meaning that the suspected abnormal batches in the abnormal cluster are more likely to be caused by real moisture anomalies. The sum of the first and second correlation degrees is calculated to obtain the anomaly score of moisture change for the abnormal cluster. A higher anomaly score indicates that the suspected abnormal batches within that cluster are more likely to be genuine moisture anomalies; a lower anomaly score indicates that the suspected abnormal batches within that cluster are more likely to be caused by sensor data drift. The formula for obtaining the anomaly score includes:

[0030] In the formula, E represents the degree of anomalousness of moisture change in the anomalous cluster. The Pearson correlation coefficient represents the relationship between the moisture content sequence and the puffiness sequence. This indicates the second degree of association after the positive correlation mapping. The Pearson correlation coefficient represents the relationship between the moisture and hardness sequences. This represents an exponential function with the natural constant as its base. This indicates the first degree of association after negative correlation mapping.

[0031] Furthermore, the true abnormal batches of emulsified puffed grain can be obtained based on the degree of moisture change anomaly. Specifically, this includes: identifying suspected abnormal batches in abnormal clusters where the moisture change anomaly exceeds a preset third threshold as true abnormal batches during the processing of emulsified puffed grain. In this embodiment, the moisture change anomaly of the abnormal clusters with true moisture anomalies tends to be 2, while the moisture change anomaly of the abnormal clusters caused by sensor data drift tends to be 0. Since these two are distributed at opposite ends of the value range, in this embodiment, the preset third threshold is set to 1. The implementer can determine this threshold according to the implementation scenario. Suspected abnormal batches in abnormal clusters exceeding this preset third threshold indicate the existence of true moisture anomalies. Consequently, true abnormal batches can be marked during processing to distinguish them from puffed grain batches with normal moisture content. This eliminates sensor interference in the quality monitoring of the puffed grain processing process, improving the accuracy of moisture monitoring and the processing quality of the emulsified puffed grain.

[0032] In summary, this invention provides a method for monitoring the processing quality of emulsified puffed grains. It obtains the degree of suspected moisture abnormality in a target batch based on the difference in moisture content between the target batch and other batches; identifies suspected abnormal batches based on the degree of suspected moisture abnormality; clusters these suspected abnormal batches based on their distribution characteristics; and obtains the moisture abnormality distribution characteristic value of each cluster based on the interval between suspected abnormal batches and adjacent suspected abnormal batches within the cluster. Finally, it identifies abnormal clusters based on the moisture abnormality distribution characteristic value. This invention obtains the degree of moisture change abnormality based on the similarity between moisture and hardness changes, and the similarity between moisture and puffing degree changes, of the suspected abnormal batches within the abnormal clusters. Based on the degree of moisture change abnormality, it identifies the actual abnormal batches of emulsified puffed grains, thus improving the accuracy of moisture monitoring and the processing quality of puffed grains.

[0033] It should be noted that the order of the above embodiments of the present invention is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0034] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

Claims

1. A method for monitoring the processing quality of emulsified and extruded grains, characterized in that, The method includes the following steps: Obtain the moisture content, hardness, and degree of puffing of consecutive batches of emulsified puffed grains from the same processing stage; The degree of suspected moisture abnormality of the target batch is obtained based on the difference in moisture content between the target batch and other batches; suspected abnormal batches are then identified based on the degree of suspected moisture abnormality. Clustering is performed based on the distribution characteristics of the suspected abnormal batches to obtain different clusters; the moisture abnormality distribution characteristic value of the cluster is obtained based on the interval characteristics between the suspected abnormal batches and adjacent suspected abnormal batches in the cluster; and abnormal clusters are obtained based on the moisture abnormality distribution characteristic value. The degree of moisture change abnormality is obtained based on the similarity characteristics of moisture and hardness changes and the similarity characteristics of moisture and puffing degree changes of suspected abnormal batches in the abnormal cluster; the true abnormal batches of emulsified puffed grain are obtained based on the degree of moisture change abnormality.

2. The method for monitoring the processing quality of emulsified and extruded grains according to claim 1, characterized in that, The step of obtaining the suspected moisture abnormality level of the target batch based on the moisture difference characteristics between the target batch and other batches includes: Calculate and normalize the sum of the absolute values ​​of the moisture differences between the target batch and the two adjacent batches to obtain the adjacent moisture difference degree; calculate and normalize the absolute value of the difference between the moisture of the target batch and the average moisture of all batches to obtain the comprehensive moisture difference degree; calculate the average of the adjacent moisture difference degree and the comprehensive moisture difference degree to obtain the suspected abnormality degree of moisture of the target batch.

3. The method for monitoring the processing quality of emulsified and extruded grains according to claim 1, characterized in that, The step of obtaining the suspected abnormal batch based on the degree of suspected moisture abnormality includes: The batches whose moisture content is suspected to be abnormal and exceeds a preset first threshold are designated as the suspected abnormal batches.

4. The method for monitoring the processing quality of emulsified and extruded grains according to claim 1, characterized in that, The step of clustering based on the distribution characteristics of the suspected abnormal batches to obtain different clusters includes: The batch distribution of suspected abnormal batches is clustered using the K-means clustering algorithm to obtain different clusters.

5. The method for monitoring the processing quality of emulsified and extruded grains according to claim 1, characterized in that, The step of obtaining the moisture anomaly distribution characteristic value of the cluster based on the interval characteristics between suspected abnormal batches and adjacent suspected abnormal batches in the cluster includes: The following steps are performed: First, calculate the sum of the number of suspected anomalous batches and the interval batches between adjacent suspected anomalous batches in the cluster to obtain the adjacent interval characteristic value of the suspected anomalous batches. Second, calculate the average of the adjacent interval characteristic values ​​of all suspected anomalous batches in the cluster to obtain a first value. Third, calculate the reciprocal of the first value to obtain the density. Fourth, calculate the reciprocal of the average of the absolute values ​​of the differences between all adjacent interval characteristic values ​​and the first value to obtain the uniformity. Fifth, calculate the sum of the density and the uniformity to obtain the moisture anomalous distribution characteristic value of the cluster.

6. The method for monitoring the processing quality of emulsified and extruded grains according to claim 1, characterized in that, The step of obtaining the abnormal clusters based on the abnormal moisture distribution characteristic values ​​includes: Clusters whose abnormal moisture distribution characteristic values ​​exceed a preset second threshold are designated as abnormal clusters.

7. The method for monitoring the processing quality of emulsified and extruded grains according to claim 1, characterized in that, The step of obtaining the degree of moisture change anomaly based on the similarity characteristics of moisture and hardness changes and the similarity characteristics of moisture and puffing degree changes of suspected abnormal batches in the abnormal cluster includes: Construct corresponding moisture, hardness, and puffiness sequences according to the batch order of suspected abnormal batches in the abnormal cluster; calculate the Pearson correlation coefficient between the moisture sequence and the hardness sequence and perform a positive correlation mapping to obtain a first degree of correlation; calculate the Pearson correlation coefficient between the moisture sequence and the puffiness sequence and perform a negative correlation mapping to obtain a second degree of correlation; calculate the sum of the first degree of correlation and the second degree of correlation to obtain the degree of abnormality of moisture change in the abnormal cluster.

8. The method for monitoring the processing quality of emulsified and extruded grains according to claim 1, characterized in that, The step of obtaining the true abnormal batches of emulsified extruded grain based on the degree of moisture change anomaly includes: Suspected abnormal batches in the abnormal clusters where the moisture change anomaly exceeds a preset third threshold are considered as real abnormal batches in the emulsified and extruded grain processing process.

Images