A granulometer-based online detection system for processing quality of potato powder

Abstract

The application discloses a kind of based on particle size analyzer's potato whole powder processing quality on-line detection system, it is related to whole powder quality detection technical field, the present application includes particle size data acquisition module, particle size data analysis module, processing quality analysis module, processing quality control module, early warning terminal and database, through the particle size distribution data of real-time acquisition, and then constructs multidimensional quality evaluation index, and presets each abnormal rule, based on the comparison between the two, the processing quality corresponding to potato whole powder is analyzed, and the judgment result is verified, the abnormal grade under abnormal processing quality is analyzed, targeted abnormal control is implemented, to ensure the accuracy and effectiveness of potato whole powder processing quality identification, realize the overall guarantee of potato whole powder processing quality, reduce the unqualified rate of potato whole powder processing quality, while realizing the timely identification of processing false alarm situation, avoid the processing loss caused by false alarm.

Description

Technical Field

[0001] This invention relates to the field of potato flour quality testing technology, specifically to an online potato flour processing quality testing system based on a particle size analyzer. Background Technology

[0002] With the continuous development of food processing technology, more and more foods are being developed in greater depth, thus presenting different appearances. Potato flour is a specific example of this. This paper proposes an online detection system for potato flour processing quality based on a particle size analyzer to effectively ensure the processing quality of potato flour and guarantee its safety and quality.

[0003] Existing technologies, such as the invention patent application CN107991396A, which discloses a rapid identification method for potato flour based on flavor analysis, belong to the field of quality and safety testing of potato staple food raw materials. This invention proposes a qualitative analysis of the flavor components of raw potato flour based on headspace solid-phase microextraction combined with gas chromatography-mass spectrometry (GC-MS), and constructs an intelligent olfactory detection system. By collecting and extracting the sensor response signals of the samples, a discrimination model for raw potato flour is established using pattern recognition methods. This invention, based on the flavor characteristics analysis of potato flour processed at different temperatures, uses intelligent sensory technology to simulate human olfaction to intelligently distinguish dehydrated potato raw materials under different processing conditions. This invention provides a rapid, non-destructive, and relatively accurate method for processing and quality control of potato flour, applicable to multiple stages of raw material procurement, processing, and storage in potato staple food production, providing a theoretical basis for quality assurance of potato staple food products.

[0004] Current technology suffers from the following problems: Traditional testing of potato flour processing quality mostly relies on manual offline sampling. However, the procedures performed by different operators are not entirely consistent, leading to inconsistent data collection. This makes it impossible to accurately assess the processing quality of potato flour, hindering the assurance of quality and exacerbating the problem of substandard products. Furthermore, false alarms occur during potato flour processing, and the inconsistencies in manually sampled data prevent accurate identification of these false alarms. Consequently, the processing quality of potato flour cannot be effectively assessed, the processing flow cannot operate efficiently, and the processing quality and operation cannot be effectively and efficiently guaranteed. Summary of the Invention

[0005] To address the aforementioned technical shortcomings, the present invention aims to provide an online detection system for potato flour processing quality based on a particle size analyzer.

[0006] To solve the above technical problems, the present invention adopts the following technical solution: The present invention provides an online detection system for potato flour processing quality based on a particle size analyzer, comprising: a particle size data acquisition module, used to acquire in real time the particle size distribution data of potato flour samples based on a particle size analyzer installed on the potato flour production line, and to construct a multidimensional quality evaluation index corresponding to potato flour based on the particle size distribution data.

[0007] The particle size data analysis module is used to preset various anomaly rules corresponding to potato flour, and compare the multidimensional quality evaluation indicators corresponding to potato flour with the preset anomaly rules to analyze the processing quality of potato flour.

[0008] The processing quality analysis module is used to obtain the production line process data corresponding to potato flour, and to compare the analysis results of the processing quality corresponding to potato flour with the data in a time series to evaluate the accuracy of the analysis results corresponding to potato flour.

[0009] The processing quality control module is used to analyze the abnormality level of potato flour processing quality and to analyze the control of the abnormality level of potato flour processing quality.

[0010] The early warning terminal is used to issue early warnings when the processing quality of potato flour is substandard or the analysis results of potato flour are inaccurate.

[0011] The beneficial effects of this invention are as follows: 1. This invention provides an online detection system for potato flour processing quality based on a particle size analyzer. By acquiring particle size distribution data in real time, it constructs multi-dimensional quality evaluation indicators and presets various anomaly rules. Based on the comparison between the two, it analyzes the processing quality corresponding to potato flour, verifies the evaluation results, analyzes the anomaly level under abnormal processing quality, and implements targeted anomaly control. This ensures the accuracy and effectiveness of potato flour processing quality identification, comprehensively guarantees the processing quality of potato flour, reduces the non-conforming rate of potato flour processing quality, and enables timely identification of processing false alarms, avoiding processing losses caused by false alarms. This allows potato flour processing to achieve comprehensive and effective quality detection and quality assurance.

[0012] 2. Real-time acquisition of particle size distribution data of potato flour samples, and construction of multi-dimensional quality evaluation indicators for potato flour based on particle size distribution data, providing effective and rich data support for subsequent processing quality assessment of potato flour, so as to ensure the accuracy of processing quality assessment of potato flour and further ensure the processing quality of potato flour. Attached Figure Description

[0013] To more clearly illustrate the technical solutions 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.

[0014] Figure 1 This is a schematic diagram of the system structure connection of the present invention. Detailed Implementation

[0015] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0016] Please see Figure 1 As shown, an online detection system for potato flour processing quality based on a particle size analyzer includes a particle size data acquisition module, a particle size data analysis module, a processing quality analysis module, a processing quality control module, an early warning terminal, and a database.

[0017] The particle size data acquisition module is connected to the particle size data analysis module and the database, respectively. The particle size data analysis module is connected to the processing quality analysis module, the early warning terminal and the database, respectively. The processing quality analysis module is connected to the processing quality control module, the early warning terminal and the database, respectively. The processing quality control module and the database are connected.

[0018] The particle size data acquisition module is used to acquire the particle size distribution data of potato flour samples in real time based on the particle size analyzer installed on the potato flour production line, and to construct multi-dimensional quality evaluation indicators for potato flour based on the particle size distribution data.

[0019] As an optional implementation, the particle size distribution data includes the median particle size (D50), particle size span (Span), proportion of fine powder, and proportion of coarse powder.

[0020] It should be noted that the fine powder percentage is the volume percentage of particles with a diameter smaller than the first preset threshold; the coarse powder percentage is the volume percentage of particles with a diameter greater than 250 mm. The percentage of particle volume.

[0021] As an optional implementation, the construction process of the multidimensional quality indicators corresponding to potato flour is as follows: Based on particle size distribution data, particle size distribution stability indicators, abnormal particle morphology indicators, and powder flowability correlation indicators are constructed respectively. The particle size distribution stability indicator is constructed by using the distribution difference distance between the current particle size distribution curve and the target particle size distribution curve as a quantitative value; the abnormal particle morphology indicator is constructed by using the skewness deviation and kurtosis deviation of the particle size distribution curve relative to the normal distribution curve as quantitative values; and the powder flowability correlation indicator is constructed by using the rate of change of the ratio of fine flour proportion to coarse particle proportion as a quantitative value. Based on the particle size distribution stability indicator, abnormal particle morphology indicator, and powder flowability correlation indicator, a multidimensional quality indicator vector corresponding to the processing quality of potato flour is comprehensively constructed.

[0022] It should be noted that the greater the difference distance, the worse the stability; the larger the deviation, the higher the degree of particle morphological abnormality; the greater the rate of change, the more drastic the flowability fluctuation; skewness coefficient and kurtosis coefficient: Obtain the particle size distribution data of the current sampling batch, where the data includes n particle size channels divided by particle size intervals and the corresponding volume percentage of each channel. The particle size value of the i-th particle size channel is denoted as... The volume percentage of this channel is denoted as The sum of the volume percentages of all channels is normalized to 1, first according to the formula. The weighted average of the particle size was obtained. Then according to the formula The weighted standard deviation of particle size is obtained. Finally, according to the formula Calculate the skewness coefficient ,when When >0, it indicates that the particle size distribution curve is positively skewed, showing a right tail, meaning that a tail exists on the coarse particle side; when When < 0, it indicates that the particle size distribution curve is negatively skewed, showing a left tail, meaning that a tail exists on the fine particle side; when When = 0, it indicates that the distribution is symmetrical; when obtaining the weighted average... and weighted standard deviation Based on the formula Calculate the kurtosis factor ,when When >0, it indicates that the particle size distribution curve is sharper than the normal distribution, indicating that the particle size tends to be highly concentrated; when When < 0, it indicates that the particle size distribution curve is flatter than the normal distribution, indicating a greater degree of particle size dispersion; when When = 0, it indicates that the particle size distribution curve has the same kurtosis as the normal distribution.

[0023] It should be noted that the ratio of fine powder to coarse particles is calculated as follows: This is achieved by obtaining particle size distribution data output by the particle size analyzer during the current testing period, including multiple particle size intervals arranged in ascending order of particle size and the corresponding particle volume percentage for each interval. A first preset threshold is then set based on the product specifications of the potato flour. Second preset threshold The first preset threshold Less than the second preset threshold ,in Used to define the upper limit of particle size for fine powders Used to define the lower limit of particle size for coarse particles, excluding particles with a size less than or equal to a first preset threshold. The particle volume percentages across all particle size ranges are summed to obtain the fine powder percentage. ,Right now That is, for all ≤ The particle size range, in which The percentage of particle volume in the i-th particle size range. The representative particle size value for the i-th particle size range is then set to a second preset threshold. The percentage of coarse particles is obtained by summing up the particle volume percentages across all particle size ranges. ,Right now That is, for all ≥ The particle size range, in which This represents the particle volume percentage for the j-th particle size range. The representative particle size value for the j-th particle size range is given by the formula. The ratio R of fine powder to coarse particles is obtained. When the R value increases, it indicates that the proportion of fine powder in the powder increases relatively and the powder flowability decreases. When the R value decreases, it indicates that the proportion of coarse particles in the powder increases relatively and the powder flowability increases.

[0024] The particle size data analysis module is used to preset various anomaly rules corresponding to potato flour, and compare the multidimensional quality evaluation indicators corresponding to potato flour with the preset anomaly rules to analyze the processing quality of potato flour.

[0025] As an optional implementation, the specific pre-setting process for the various anomaly rules corresponding to potato flour is as follows: Extract a multi-dimensional quality evaluation index dataset of historical potato flour batches that meet standards from the database; calculate the control interval for each index based on a pre-set confidence interval, and use this as the basic threshold; then, based on the physical parameters of the process equipment stored in the database, set the safety boundary values ​​for each process parameter as mandatory thresholds, with the mandatory thresholds having higher priority than the control intervals; next, establish a mapping relationship library for different product specifications, so that each product model is associated with the corresponding target distribution curve and threshold group; when the product model is switched, retrieve the corresponding threshold group to complete the rule update; finally, based on the processing records of historical potato flour processing anomaly events in the database, adjust the thresholds of each index by offset, thereby completing the setting of the various anomaly rules corresponding to potato flour.

[0026] It should be noted that the control range for each indicator is calculated as follows: A dataset of multidimensional quality evaluation indicators for historical compliant batches is extracted from the database. These historical compliant batches are production batches that have passed laboratory testing and customer acceptance. The multidimensional quality evaluation indicators for each batch include at least the particle size distribution stability index, the skewness coefficient and kurtosis coefficient of abnormal particle morphology, and the coefficient of variation of the fineness-to-coarseness ratio, which is related to powder flowability. The number of batches is denoted as [number missing]. , ≥20, for any evaluation index X, calculate its average value across all N historical batches that met the standards. and standard deviation , where the average =(1 / )× Standard deviation , Let j be the calculated value of this indicator in the j-th batch, and then according to the formula... = +k× To obtain the upper limit of control for this indicator. According to the formula = -k× The control lower limit of the indicator is obtained. Where k is a preset confidence coefficient, and finally, the upper limit is controlled. and control lower limit This constitutes the control range for the indicator. When the calculated value of this indicator falls within [ ] during real-time detection. , When the value is within the specified range, the indicator is considered normal; when the calculated value of the indicator is below [a certain value], the indicator is considered normal. or higher When this happens, the indicator is deemed abnormal.

[0027] It should be noted that the process involves obtaining a complete list of potato flour product specifications produced by the factory. Each product specification corresponds to a unique product model, which must include at least the target median particle size and target particle size distribution width. Then, for each product model, a corresponding target particle size distribution curve P(d) is generated based on its corresponding target median particle size and target particle size distribution width. This target particle size distribution curve is the benchmark particle size distribution curve specified in the quality standard corresponding to the product model. Finally, the target values ​​of each evaluation index corresponding to the target particle size distribution curve P(d) for each product model are calculated, including... The target values ​​for particle size distribution stability index, skewness coefficient, kurtosis coefficient, and coarseness ratio are set. For each product model, a set of thresholds is generated according to the control interval calculation method. The thresholds include the statistical control upper limit and statistical control lower limit of each index. A mapping relationship is established between the product model and the target particle size distribution curve and the thresholds to form a mapping relationship library. When the product model of the current production task is input, the corresponding target particle size distribution curve and thresholds are retrieved from the mapping relationship library according to the product model, and the loading and switching of rules are automatically completed.

[0028] It should be noted that historical anomaly event handling records are extracted from the database, including the measured values ​​of evaluation indicators at the time of the anomaly occurrence, the anomaly type determination results, and the final confirmation results after manual review. The final confirmation results after manual review include two types: confirmed anomalies and false alarms. All events in the records whose final confirmation results are false alarms are filtered out, and the corresponding measured values ​​of evaluation indicators and the deviation of the indicators from the statistical control limits are extracted from each false alarm event. =X_actual measurement- And the percentage of deviation outside the statistical control limits, P=| | / ( - Then, false alarm events are categorized according to product model. Based on each evaluation indicator for each product model, the average deviation μ_Δ and the average distance percentage P in false alarm events are statistically analyzed. Using the formula C=1-(P / 100%), the initial dynamic correction coefficient C for that indicator under that product model is obtained. The calculated value X_real-time of each evaluation indicator obtained from real-time detection is then fused and corrected with the corresponding dynamic correction coefficient C in the following manner: When X_real-time > At that time, the corrected judgment deviation value _corr=(X_real-time- If )×C, If _corr > 0, it is still considered an exception. If _corr≤0, the exception judgment will be corrected to normal; when X_real-time < At that time, the corrected judgment deviation value _corr=( -X_real-time)×C, if If _corr > 0, it is still considered an exception. If _corr≤0, the abnormal judgment will be corrected to normal; that is, the dynamic correction coefficient C acts on the judgment result by reducing the deviation between the measured value and the control limit, so that the judgment boundary shrinks or expands in the direction of the control limit.

[0029] As an optional implementation, the analysis of the processing quality of potato flour is carried out in the following specific process: the particle size distribution stability index, abnormal particle morphology index and powder flowability correlation index of potato flour are compared with preset reference thresholds respectively. When only the particle size distribution stability index exceeds the preset reference stability threshold, the processing quality of potato flour is determined to be unqualified and there are abnormal process parameters.

[0030] When only the abnormal particle morphology index exceeds the preset reference morphology threshold, the processing quality of the potato flour is determined to be unqualified, indicating wear of the crushing equipment or abnormal material characteristics.

[0031] When only the powder flowability-related index exceeds the preset reference threshold, the processing quality of potato flour is deemed unqualified, indicating an abnormality in the screening or grading process.

[0032] When multiple indicators exceed the preset reference thresholds simultaneously, the anomaly type is determined to be a primary and secondary related anomaly based on the combination of indicators exceeding the thresholds; otherwise, the processing quality is determined to be qualified.

[0033] As an optional implementation method, the analysis of the anomaly type corresponding to potato flour and the anomaly type when they are primary and secondary related anomalies is carried out. The specific analysis process is as follows: when the particle size distribution stability index exceeds the preset reference morphology threshold and the skewness of the abnormal particle morphology index exceeds the preset second reference skewness threshold, but the powder flowability related index does not exceed the preset reference related threshold, the primary anomaly type is determined to be a primary crushing anomaly caused by the fluctuation of the feed rate.

[0034] When the particle size distribution stability index exceeds the preset reference morphology threshold and the kurtosis of the abnormal particle morphology index exceeds the preset second reference kurtosis threshold, but the skewness does not exceed the preset second reference skewness threshold, the main abnormality type is determined to be secondary crushing abnormality caused by wear of the crusher hammer or blade.

[0035] When the particle size distribution stability index does not exceed the preset reference morphology threshold, but the powder flowability correlation index exceeds the preset reference correlation threshold and the abnormal particle morphology index exceeds the preset reference morphology threshold, the main anomaly type is determined to be an abnormal deviation of the air pressure or air velocity parameters in the air classification process.

[0036] The processing quality analysis module is used to obtain the production line process data corresponding to potato flour, and to compare the analysis results of the processing quality corresponding to potato flour with the data in a time series to evaluate the accuracy of the analysis results corresponding to potato flour.

[0037] As an optional implementation, the time-series correlation comparison to evaluate the accuracy of the analysis results corresponding to potato flour is performed as follows: Based on the judgment time of the processing quality anomaly corresponding to potato flour, a time analysis window is constructed by extending a preset time width forward and backward. The trend characteristics of the data sequence of each key process parameter and the abnormal characteristics of the particle size distribution data within the window are extracted respectively. First, a direction consistency test is performed to check whether the change direction of the process parameter is consistent with the change direction of the abnormal particle size. If they are inconsistent, it is judged as a measurement pseudo-anomaly. If they are consistent, an amplitude matching test is further performed. The measured ratio of the change amplitude of the process parameter to the change amplitude of the abnormal particle size is matched with the reference ratio in the preset mapping rule library. If the match is successful, the anomaly type corresponding to the processing quality anomaly of potato flour is judged to be correctly matched. If the match fails, the anomaly type is judged to be incorrectly matched, and it will be reconfirmed.

[0038] It should be noted that when it is determined that there is a problem with the processing quality of potato flour, the production line process data during the abnormal period is obtained, including the speed of the crusher, the air pressure of the air classifier, the pulse frequency of the dust collector, and the feeding rate. False anomalies are false alarms, in which there is actually no abnormality or quality problem.

[0039] The processing quality control module is used to analyze the abnormality level of potato flour processing quality and to analyze the control of the abnormality level of potato flour processing quality.

[0040] As an optional implementation, the specific analysis process for determining the abnormality level of potato flour processing quality is as follows: The deviation of each abnormal indicator when potato flour processing quality is abnormal is extracted. The deviation of the potato flour processing quality is compared with the reference deviation range corresponding to each processing quality problem level stored in the database. If the deviation of the potato flour processing quality exceeds the reference deviation range corresponding to a certain processing quality problem level stored in the database, the processing quality abnormality is determined to be a Level 1 abnormality. If the deviation of the potato flour processing quality is included within the reference deviation range corresponding to a certain processing quality problem level stored in the database, the processing quality abnormality is determined to be a Level 2 abnormality. If the deviation of the potato flour processing quality does not reach the reference deviation range corresponding to a certain processing quality problem level stored in the database, the processing quality abnormality is determined to be a Level 3 abnormality.

[0041] It should be noted that the deviation is calculated as (|measured value - threshold boundary value|) / (upper control limit - lower control limit); Level 1 anomalies are serious quality problems, Level 2 anomalies are moderate quality problems, and Level 3 anomalies are minor quality problems.

[0042] As an optional implementation, the analysis of the control of the processing quality anomaly level corresponding to potato flour is specifically as follows: Based on the anomaly type and anomaly level corresponding to potato flour, the control actions are compared with the corresponding control actions in the potato processing strategy parameter and quality mapping relationship library stored in the database to obtain the matching control actions. The matching control actions are then converted into equipment executable instructions, which include four types of parameter adjustment instructions, three types of early warning instructions, three types of maintenance instructions, and four types of emergency shutdown instructions. The output priority is four types > three types > one types > two types. When it is a level one anomaly, the four types or three types of instructions are output; when it is a level two anomaly, the one type of instruction is output; and when it is a level three anomaly, the two types of instructions are output.

[0043] The database is used to store particle size distribution data, multidimensional quality evaluation index datasets, physical parameters, historical records of handling abnormal events in potato flour processing, production line process data, reference deviation ranges and control actions corresponding to each processing quality problem level.

[0044] It should be noted that the preset thresholds, preset rule bases, preset datasets, and preset intervals involved in this invention can all be configured and flexibly adjusted by engineers in the field in conjunction with the processing quality standards, demand standards, and quality and safety standards corresponding to whole powder processing and the actual needs of scheme comparison. This application does not impose a unique limitation on the specific values ​​of each parameter.

[0045] The examples described in this invention are not limited to the specific embodiments listed above. The examples are merely illustrative to facilitate understanding of the invention and do not constitute a limitation on the scope of protection of this invention. Any modifications, equivalent substitutions, etc., made within the spirit and principles of this invention should be included within the scope of protection.

[0046] The above description is merely an example and illustration of the concept of the present invention. Those skilled in the art can make various modifications or additions to the specific embodiments described or use similar methods to replace them, as long as they do not deviate from the concept of the invention or exceed the scope defined in this specification, they should all fall within the protection scope of the present invention.

Claims

1. An online detection system for potato flour processing quality based on a particle size analyzer, characterized in that, Includes the following modules: The particle size data acquisition module is used to acquire the particle size distribution data of potato flour samples in real time based on the particle size analyzer installed on the potato flour production line, and to construct multi-dimensional quality evaluation indicators for potato flour based on the particle size distribution data. The particle size data analysis module is used to preset various anomaly rules corresponding to potato flour, and compare the multidimensional quality evaluation indicators corresponding to potato flour with the preset anomaly rules to analyze the processing quality of potato flour. The processing quality analysis module is used to obtain the production line process data corresponding to potato flour, and to compare the analysis results of the processing quality corresponding to potato flour with the data in a time series to evaluate the accuracy of the analysis results corresponding to potato flour. The processing quality control module is used to analyze the abnormality level of potato flour processing quality and to analyze the control of the abnormality level of potato flour processing quality. The early warning terminal is used to issue early warnings when the processing quality of potato flour is substandard or the analysis results of potato flour are inaccurate.

2. The online detection system for potato flour processing quality based on a particle size analyzer according to claim 1, characterized in that, The particle size distribution data includes the median particle size (D50), particle size span (Span), proportion of fine powder, and proportion of coarse powder.

3. The online detection system for potato flour processing quality based on a particle size analyzer according to claim 2, characterized in that, The specific construction process for the multidimensional quality indicators corresponding to potato flour is as follows: Based on particle size distribution data, particle size distribution stability index, abnormal particle morphology index, and powder flowability correlation index are constructed respectively. The particle size distribution stability index is constructed by using the distribution difference distance between the current particle size distribution curve and the target particle size distribution curve as the quantitative value; the abnormal particle morphology index is constructed by using the skewness deviation and kurtosis deviation of the particle size distribution curve relative to the normal distribution curve as the quantitative value; and the powder flowability correlation index is constructed by using the rate of change of the ratio of fine powder to coarse particles as the quantitative value. Based on the particle size distribution stability index, abnormal particle morphology index, and powder flowability correlation index, a multi-dimensional quality index vector corresponding to the processing quality of potato flour is comprehensively constructed.

4. The online detection system for potato flour processing quality based on a particle size analyzer according to claim 3, characterized in that, The specific pre-setting process for the various anomaly rules corresponding to the preset potato flour is as follows: A multidimensional quality evaluation index dataset of historical potato flour batches that meet standards is extracted from the database. Control intervals for each index are calculated based on preset confidence intervals and used as basic thresholds. Next, safety boundary values ​​for each process parameter are set as mandatory thresholds based on the physical parameters of the process equipment stored in the database, with the mandatory thresholds having higher priority than the control intervals. Then, a mapping relationship library for different product specifications is established so that each product model is associated with the corresponding target distribution curve and threshold group. When the product model is switched, the corresponding threshold group is retrieved to complete the rule update. Finally, the thresholds of each index are offset and corrected based on the processing records of historical potato flour processing anomalies in the database, thereby completing the setting of various anomaly rules for potato flour.

5. The online detection system for potato flour processing quality based on a particle size analyzer according to claim 4, characterized in that, The analysis of the processing quality of potato flour is as follows: The particle size distribution stability index, abnormal particle morphology index, and powder flowability correlation index of potato flour are compared with preset reference thresholds. When only the particle size distribution stability index exceeds the preset reference stability threshold, the processing quality of potato flour is deemed unqualified and there are abnormal process parameters. When only the abnormal particle morphology index exceeds the preset reference morphology threshold, the processing quality of potato flour is determined to be unqualified, indicating wear of the crushing equipment or abnormal material characteristics. When only the powder flowability-related index exceeds the preset reference correlation threshold, the processing quality of potato flour is deemed unqualified, indicating an abnormality in the screening or grading process. When multiple indicators exceed the preset reference thresholds simultaneously, the anomaly type is determined to be a primary and secondary related anomaly based on the combination of indicators exceeding the thresholds; otherwise, the processing quality is determined to be qualified.

6. The online detection system for potato flour processing quality based on a particle size analyzer according to claim 5, characterized in that, The analysis of anomalies related to potato flour, specifically when anomalies are primary and secondary, is as follows: When the particle size distribution stability index exceeds the preset reference morphology threshold and the skewness of the abnormal particle morphology index exceeds the preset second reference skewness threshold, but the powder flowability correlation index does not exceed the preset reference correlation threshold, the main abnormality type is determined to be a primary crushing abnormality caused by feed rate fluctuation. When the particle size distribution stability index exceeds the preset reference morphology threshold and the kurtosis of the abnormal particle morphology index exceeds the preset second reference kurtosis threshold, but the skewness does not exceed the preset second reference skewness threshold, the main abnormality type is determined to be secondary crushing abnormality caused by wear of the crusher hammer or blade. When the particle size distribution stability index does not exceed the preset reference morphology threshold, but the powder flowability correlation index exceeds the preset reference correlation threshold and the abnormal particle morphology index exceeds the preset reference morphology threshold, the main anomaly type is determined to be an abnormal deviation of the air pressure or air velocity parameters in the air classification process.

7. The online detection system for potato flour processing quality based on a particle size analyzer according to claim 6, characterized in that, The time-series correlation comparison is performed to evaluate the accuracy of the analysis results corresponding to potato flour. The specific analysis process is as follows: Based on the judgment time of the processing quality anomaly corresponding to potato flour, a time analysis window is constructed with a preset time width before and after. The trend characteristics of the data sequence of each key process parameter and the abnormal characteristics of the particle size distribution data are extracted in the window. First, a direction consistency test is performed to check whether the change direction of the process parameter is consistent with the change direction of the abnormal particle size. If they are inconsistent, it is judged as a measurement pseudo-anomaly. If they are consistent, an amplitude matching test is further performed. The measured ratio of the change amplitude of the process parameter to the change amplitude of the abnormal particle size is matched with the reference ratio in the preset mapping rule library. If the match is successful, the anomaly type of the processing quality anomaly corresponding to potato flour is judged to be correctly matched. If the match fails, the anomaly type is judged to be incorrectly matched, and it will be reconfirmed.

8. The online detection system for potato flour processing quality based on a particle size analyzer according to claim 7, characterized in that, The analysis of the abnormality level of potato flour processing quality is as follows: When potato flour processing quality is abnormal, the deviation of each abnormal indicator is extracted. The quality deviation of potato flour processing is compared with the reference deviation range corresponding to each processing quality problem level stored in the database. If the quality deviation of potato flour processing exceeds the reference deviation range corresponding to a certain processing quality problem level stored in the database, the processing quality abnormality is determined to be a level one abnormality. If the quality deviation of potato flour processing is included in the reference deviation range corresponding to a certain processing quality problem level stored in the database, the processing quality abnormality is determined to be a level two abnormality. If the quality deviation of potato flour processing does not reach the reference deviation range corresponding to a certain processing quality problem level stored in the database, the processing quality abnormality is determined to be a level three abnormality.

9. The online detection system for potato flour processing quality based on a particle size analyzer according to claim 7, characterized in that, The analysis of the control of the processing quality abnormality level of potato flour is as follows: Based on the anomaly type and level corresponding to potato flour, the corresponding control actions are compared with the corresponding control actions in the potato processing strategy parameter and quality mapping relationship database stored in the database to obtain the matching control actions. The matching control actions are then converted into equipment executable instructions, which include four types of instruction execution: parameter adjustment instructions, early warning instructions, maintenance instructions, and emergency shutdown instructions. The output priority is four > three > one > two. When it is a level one anomaly, the four or three type instructions are output; when it is a level two anomaly, the one type instruction is output; and when it is a level three anomaly, the two type instructions are output.

10. The online detection system for potato flour processing quality based on a particle size analyzer according to claim 1, characterized in that, It also includes a database for storing particle size distribution data, multidimensional quality evaluation index datasets, physical parameters, historical records of handling abnormal events in potato flour processing, production line process data, reference deviation ranges and control actions corresponding to each processing quality problem level.

Images