Intelligent evaluation system for mixing degree of ceramic granular material

Abstract

The present application relates to the technical field of evaluation of the mixing degree of ceramic material, and discloses an intelligent evaluation system for the mixing degree of ceramic material, which comprises a material sampling device, a parameter acquisition unit, a feature analysis module and an evaluation and determination module. The parameter acquisition unit is arranged at each mixing station and comprises a record storage component and a detection assembly. The record storage component stores historical mixing abnormal information (including abnormal time points, color, granularity and vibration information). The detection assembly acquires real-time color, granularity data and vibration information. The feature analysis module divides abnormal and stable time intervals, and determines color, granularity reference values and vibration amplitude reference intervals. The evaluation and determination module adjusts the activation state of the sampling device and the response state of the detection assembly according to the feature time intervals, judges the mixing abnormal tendency through real-time data, and determines whether to change the state of the sampling device in combination with the vibration amplitude. The system improves the accuracy and efficiency of the evaluation of the mixing degree and is suitable for intelligent monitoring of the production of ceramic material.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of evaluation of the mixing degree of ceramic material, in particular to an intelligent evaluation system for the mixing degree of ceramic material. BACKGROUND

[0002] In the production process of ceramic, material mixing is a key link to ensure the uniformity of product quality. Ceramic production usually needs to go through multiple processes such as raw material proportioning, mixing and stirring, calcination, etc., and the uniformity of material mixing at each mixing station directly affects the core performance of the final ceramic, such as strength, density, and water absorption rate. If the mixing is not uniform, it may cause local composition deviation of the ceramic product, trigger quality fluctuations, increase the rate of defective products, and cause waste of raw materials and increase production costs.

[0003] In traditional ceramic production, the evaluation of the mixing degree of material mainly relies on manual sampling detection, which judges the mixing effect by observing the color difference of the material with the naked eye, sieving and measuring the particle size distribution, etc. This method has obvious limitations: on the one hand, manual detection is low in efficiency and difficult to realize real-time monitoring of each mixing station on the production line, which is prone to miss transient mixing abnormalities; on the other hand, the detection results are greatly influenced by human experience and are highly subjective, lacking unified quantitative standards, resulting in insufficient evaluation accuracy. In addition, manual sampling needs to interrupt the production process, affecting the continuity of production, and cannot trace the historical data of mixing abnormalities, making it difficult to form a systematic quality improvement plan.

[0004] With the application of automation technology in manufacturing, some ceramic production lines have begun to introduce sensors for parameter collection, but existing systems mostly use fixed-frequency detection mode, continuously sampling regardless of whether the mixing state is stable or not, causing data redundancy and energy waste. At the same time, existing systems do not fully combine historical abnormal data for dynamic analysis, making it difficult to distinguish between normal fluctuations and abnormal trends, resulting in a high rate of misjudgment. For example, when the color or particle size of the material shows slight fluctuations, the system may directly determine it as a mixing abnormality, triggering unnecessary shutdown inspection and affecting production efficiency; when the mixing equipment causes a decline in mixing effect due to vibration abnormalities, it may delay abnormal handling due to the failure to associate vibration data in time.

[0005] How to build an intelligent evaluation system that can dynamically divide the detection period by combining historical data, analyze the mixing state based on multi-parameter linkage, and realize adaptive adjustment of the sampling strategy has become a key requirement to improve the stability of ceramic production quality and reduce production costs. SUMMARY

[0006] The present application aims to provide an intelligent evaluation system for the mixing degree of ceramic material to solve the problems raised in the background.

[0007] To achieve the above-mentioned purpose, the present application provides the following technical solution: an intelligent evaluation system for the mixing degree of ceramic material, comprising:

[0008] A material sampling device is used to obtain material samples of each mixing station of a ceramsite production line;

[0009] A parameter acquisition unit is arranged at each mixing station of the ceramsite production line, and includes a record storage component for storing historical mixing abnormal information of each mixing station, and a plurality of detection components for acquiring material color data, particle size data and mixing process vibration information of each station, wherein the historical mixing abnormal information includes historical abnormal time points, historical color data, historical particle size data and historical vibration information.

[0010] A feature analysis module is connected to the parameter acquisition unit, and is used to divide a statistical period into a plurality of feature time intervals according to the historical mixing abnormal information of a single mixing station, to determine color reference values and particle size reference values according to historical color data and historical particle size data of a single station, to construct a time-domain vibration waveform diagram according to the vibration information to determine an amplitude distribution atlas of a stable period, and to determine a vibration amplitude reference interval according to the amplitude distribution atlas.

[0011] The feature time intervals include abnormal time intervals and stable time intervals.

[0012] An evaluation and determination module is connected to each material sampling device, the parameter acquisition unit and the feature analysis module, and is used to determine an operation activation state of a material sampling device and a collection response state of a detection component of a corresponding station according to the feature time intervals of a single mixing station, to determine a mixing abnormality tendency according to real-time color data and real-time particle size data of a single mixing station to determine whether to analyze vibration information, and to determine whether to change the operation activation state of a corresponding material sampling device according to the mixing abnormality tendency and a real-time vibration amplitude.

[0013] Preferably, each mixing station is provided with one detection component, and a single detection component includes a color identifier, a particle size measuring instrument and a vibration sensor.

[0014] Preferably, the feature analysis module determines an abnormal occurrence point of a single mixing station according to historical abnormal time points of the single mixing station, determines a fixed time interval centered on the abnormal occurrence point as an abnormal time interval, and determines an absolute complement of the abnormal time interval in a statistical period as a stable time interval.

[0015] If the length of a single stable time interval is less than a set threshold value, the stable time interval and two adjacent abnormal time intervals are merged into one abnormal time interval.

[0016] Preferably, the feature analysis module determines the color reference value according to the color mean value and the color fluctuation range corresponding to the historical color data in the stable time interval, and determines the granularity reference value according to the granularity mean value and the granularity fluctuation range corresponding to the historical granularity data in the stable time interval.

[0017] In the formula, the color reference value is the sum of the color mean value and the lower limit value of the color fluctuation range, and the granularity reference value is the sum of the granularity mean value and the lower limit value of the granularity fluctuation range.

[0018] Preferably, the feature analysis module determines the vibration amplitude reference interval according to the amplitude range corresponding to the pre-set confidence in the amplitude distribution map of the stable time interval.

[0019] In the formula, the horizontal axis of the amplitude distribution map is the vibration signal intensity, and the vertical axis of the amplitude distribution map is the occurrence frequency.

[0020] Preferably, the evaluation and determination module determines the running activation state of the material sampling device of the corresponding mixing station and the collection response state of the detection component according to the feature time interval of the single mixing station, including:

[0021] If the feature time interval is an abnormal time interval, the corresponding material sampling device is in the activated state, and the collection response state of the detection component is to collect only the color data, the granularity data and the vibration data;

[0022] If the feature time interval is a stable time interval, the corresponding material sampling device is in the dormant state, and the collection response state of the detection component is to collect the color data, the granularity data and the vibration data, and the feature analysis module is controlled to determine the color reference value and the granularity reference value.

[0023] In the formula, the running activation state includes the activated state and the dormant state.

[0024] Preferably, the evaluation and determination module determines the mixing abnormality tendency of the single mixing station according to the determination result that the material sampling device of the single mixing station is in the dormant state, in combination with the real-time color data and the real-time granularity data, including:

[0025] If the real-time color data is less than or equal to the color reference value and the real-time granularity data is less than or equal to the granularity reference value, it is determined that the corresponding mixing station does not have the mixing abnormality tendency;

[0026] If the real-time color data is greater than the color reference value and the real-time granularity data is greater than the granularity reference value, it is determined that the corresponding mixing station has the mixing abnormality phenomenon, and the corresponding material sampling device is adjusted to the activated state.

[0027] Preferably, the evaluation and determination module determines the mixing abnormality tendency of the mixing station according to the determination result of the single mixing station material sampling device being in the dormant state, in combination with the real-time color data and the real-time granularity data, and further comprises:

[0028] If the real-time color data is greater than the color reference value or the real-time granularity data is greater than the granularity reference value, it is determined that the corresponding mixing station has a mixing abnormality tendency and it is determined to analyze the vibration information.

[0029] Preferably, the evaluation and determination module controls the feature analysis module to determine the vibration amplitude reference interval according to the determination result of the mixing abnormality tendency, and determines whether to change the running activation state of the material sampling device corresponding to the mixing station according to the real-time vibration amplitude.

[0030] If the real-time vibration amplitude is not in the vibration amplitude reference interval, it is determined to change the material sampling device corresponding to the mixing station from the dormant state to the active state.

[0031] If the real-time vibration amplitude is in the vibration amplitude reference interval, it is determined not to change the running activation state of the material sampling device corresponding to the mixing station so as to keep it in the dormant state.

[0032] Preferably, when the feature analysis module determines the amplitude distribution atlas of the stable time interval, first, the vibration information in the stable time interval is segmented and sampled, each segment of sampling data has a fixed length, then the amplitude values of each segment of sampling data are frequency counted, and finally the statistical results of each segment are combined to form the amplitude distribution atlas covering the entire stable time interval.

[0033] Compared with the prior art, the present application has the following advantages:

[0034] From the perspective of data acquisition and analysis, the system innovatively introduces historical mixing abnormality information as the analysis basis, divides the statistical period into an abnormal time interval and a stable time interval through the feature analysis module, so that the detection strategy is more targeted. In the abnormal time interval, the material sampling device is activated and the parameter acquisition is intensified to ensure accurate capture of potential problems; and in the stable time interval, the dormant sampling mode is adopted and the color and granularity reference values are dynamically updated, which not only reduces invalid data acquisition, but also ensures the timeliness of the reference values, solving the problems of data redundancy and reference value rigidity caused by traditional fixed frequency sampling.

[0035] In the evaluation decision logic, the system realizes the hierarchical judgment of mixed abnormalities through multi-parameter linkage analysis. First, based on real-time color and particle size data, the system preliminarily identifies the tendency of mixed abnormalities, and then combines vibration information for secondary verification, forming a "color-particle size-vibration" triple verification mechanism, which greatly reduces the probability of misjudgment of a single parameter. For example, when the color or particle size fluctuates, the system does not directly determine the abnormality, but further confirms whether the vibration amplitude exceeds the reference interval, avoiding false operation caused by normal process fluctuations and improving the reliability of evaluation.

[0036] From the perspective of production efficiency and cost control, the system reduces unnecessary sampling detection times through adaptive activation state adjustment of the material sampling device, reduces equipment energy consumption and manual intervention requirements. At the same time, with the help of the vibration amplitude reference interval and the benchmark value updating mechanism constructed by historical data, the system has self-learning ability and can continuously optimize the evaluation model as the production process advances, adapt to changes in the characteristics of different batches of raw materials, and significantly improve the stability of the ceramsite product quality in the long run, reduce the loss of substandard products caused by mixed abnormalities, and indirectly reduce production costs.

[0037] The modular design of each module of the system facilitates integration in existing production lines, without the need for large-scale modification of the original equipment, reducing the application threshold. Through independent monitoring and centralized analysis of each mixing station, the system can not only realize precise positioning of single-station abnormalities, but also provide data support for mixed process optimization of the entire production line, helping to transform ceramsite production from experience-driven to data-driven. BRIEF DESCRIPTION OF DRAWINGS

[0038] Figure 1 The working principle diagram of the ceramsite material mixing degree intelligent evaluation system described in the present application;

[0039] Figure 2 The flowchart for feature time interval division;

[0040] Figure 3 The flowchart for color and particle size benchmark value determination;

[0041] Figure 4 The flowchart for state control of the sampling device and detection component;

[0042] Figure 5 The flowchart for generating the amplitude distribution map. DETAILED DESCRIPTION

[0043] With reference to the drawings of the embodiments of the present application, the technical solutions in the embodiments of the present application will be described clearly and completely. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments of the present application, all the other embodiments obtained by a person of ordinary skill in the art without creative work fall within the protection scope of the present application.

[0044] Please refer to Figures 1-5 The present application relates to a kind of ceramic material mixing degree intelligent evaluation system, the system includes: material sampling device, parameter acquisition unit, feature analysis module and evaluation module.Detailed implementation steps are as follows:

[0045] Material sampling device, the device is used to obtain the material sample of each mixing station of ceramic production flow line.

[0046] Parameter acquisition unit is arranged at each mixing station of ceramic production flow line, and it includes record storage component and multiple detection components.Record storage component is used to store the historical mixing abnormal information of each mixing station, and the historical mixing abnormal information includes historical abnormal time point, historical color data, historical granularity data and historical vibration information.Detection component is used to collect the color data, granularity data and mixing process vibration information of each station material.

[0047] Feature analysis module is connected with parameter acquisition unit, and it can divide statistical period into several characteristic time intervals according to the historical mixing abnormal information of single mixing station, and characteristic time interval includes abnormal time interval and stable time interval.Meanwhile, color reference value and granularity reference value are determined according to the historical color data and historical granularity data of single station, vibration amplitude reference interval can also be determined according to the vibration information to construct time domain vibration waveform diagram and determine the amplitude distribution atlas of stable period.

[0048] Evaluation module is connected with each material sampling device, parameter acquisition unit and feature analysis module, and it can determine the running activation state of corresponding station material sampling device and the acquisition response state of detection component according to the characteristic time interval of single mixing station, determine whether to analyze vibration information according to the real-time color data and real-time granularity data of single mixing station to determine mixing abnormal tendency, and determine whether to change the running activation state of corresponding material sampling device according to mixing abnormal tendency and real-time vibration amplitude.

[0049] In this embodiment, each mixing station is provided with a detection assembly, and a single detection assembly includes a color identifier, a particle size measuring instrument, and a vibration sensor. The color identifier uses optical sensing principles to capture and identify the color characteristics of the material by capturing the spectral information reflected from the surface of the material. Specifically, the color identifier is internally provided with a light source emitting module and a spectrum receiving module. The light source emitting module emits light of a specific wavelength range to the surface of the material, and the light is reflected from the surface of the material and received by the spectrum receiving module. The spectrum receiving module converts the received light signal into an electrical signal, and the electrical signal is processed by a signal processing circuit to amplify, filter, etc., and finally the color of the material is converted into quantifiable color data, such as RGB value or HSV value, etc. During the operation of the ceramsite production line, the color identifier continuously works to collect the color of the material at each mixing station in real time at a fixed sampling frequency, ensuring that the changes in the color of the material can be obtained in a timely manner.

[0050] The particle size measuring instrument uses the principle of laser diffraction to accurately measure the size distribution of the material particles. The instrument mainly consists of a laser light source, a sample pool and a detector. When the material passes through the sample pool, the laser beam emitted by the laser light source is incident on the material particles, and diffraction phenomenon occurs. Different sizes of particles produce different diffraction patterns, which can be captured by the detector and converted into an electrical signal. Through analysis and calculation of the electrical signal, the particle size measuring instrument can obtain the particle size distribution data of the material particles, including the average particle size, particle size distribution range and other parameters. In actual application, the particle size measuring instrument will set appropriate measurement parameters such as laser wavelength, sampling time, etc. according to the characteristics of the material and the production process requirements, to ensure that accurate particle size data is obtained. The instrument monitors the particle size of the material at each mixing station in real time, providing key particle size information for subsequent analysis of the mixing degree.

[0051] The vibration sensor uses the piezoelectric effect principle to sense the vibration information during the mixing process and collect vibration signals in real time. The piezoelectric crystal inside the sensor will generate an electric charge when subjected to vibration, and the size of the electric charge is proportional to the amplitude and frequency of the vibration. The vibration sensor converts the generated electric charge signal into a standard electrical signal output after amplification, filtering, etc. This electrical signal contains information such as amplitude, frequency, and phase of the vibration during the mixing process. In the ceramsite production process, the vibration of the mixing equipment is closely related to the mixing state of the material, so the vibration sensor records the vibration during the mixing process in real time, which can provide important basis for judging whether the mixing degree of the material is abnormal.

[0052] The color identifier, the particle size measuring instrument and the vibration sensor in the detection assembly of each mixing station work synchronously during system operation. The color identifier continuously collects material color data, and transmits the color data such as RGB value or HSV value collected each time to the recording storage component of the parameter acquisition unit in real time. The particle size measuring instrument monitors the material particle size in real time, and also transmits the particle size data such as average particle size and particle size distribution range obtained each time to the recording storage component in real time. The vibration sensor records vibration signals in the mixing process in real time, and transmits the electrical signals representing vibration amplitude, frequency and other information to the recording storage component for storage.

[0053] The recording storage component classifies and stores the data transmitted from each detection assembly, and arranges and saves the historical color data, historical particle size data and historical vibration information according to different mixing stations and different time points. In this way, when the mixing state of a certain mixing station needs to be analyzed, the corresponding historical data can be quickly and accurately retrieved. For example, when determining the color reference value and the particle size reference value, the feature analysis module can obtain all the historical color data and historical particle size data in the stable time interval of the mixing station from the recording storage component, thereby providing a data basis for subsequent statistical calculation.

[0054] The three parts of the detection assembly cooperate with each other to complete the acquisition of the related parameters of the material in the mixing station. The color identifier focuses on the color change of the material, the particle size measuring instrument grasps the size distribution of the material particles, and the vibration sensor monitors the mechanical vibration in the mixing process. The data of these three aspects reflects the mixing state of the material from different dimensions, and provides comprehensive information support for the analysis and judgment of the entire intelligent evaluation system. Through the cooperative work of the three components, the system can obtain the material data of each mixing station in real time and accurately, thereby laying a solid foundation for subsequent feature analysis and evaluation and judgment. Whether in the abnormal time interval or in the stable time interval, the detection assembly continuously works to ensure the continuity and integrity of the data, so that the system can timely discover the abnormal situation in the mixing process and take corresponding measures for adjustment.

[0055] In this embodiment, the feature analysis module processes the historical abnormal time points of a single mixing station to determine the abnormal occurrence point, and divides the abnormal time interval and the stable time interval according to the abnormal occurrence point. Specifically, the feature analysis module extracts the historical mixing abnormal information of the mixing station from the recording storage component of the parameter acquisition unit, which includes historical abnormal time points. These historical abnormal time points are specific time points recorded when the system detects that the mixing is abnormal in the past production process, for example, a certain year, month, day, hour, minute and second.

[0056] The feature analysis module determines each historical abnormal time point as an abnormal occurrence point. Centering on the abnormal occurrence point, a fixed time length is extended forward and backward, thereby forming a time interval, which is the abnormal time interval. The fixed time length is a value preset by the system, which can be set to 10 minutes, 15 minutes, etc. The specific time length can be determined according to the process characteristics and actual needs of the ceramsite production. For example, if the abnormal occurrence point is 10:30 on June 25, 2025, and the fixed time length is set to 10 minutes, then the abnormal time interval is from 10:20 to 10:40 on June 25, 2025.

[0057] After determining all abnormal time intervals, the feature analysis module needs to determine the stable time interval in the entire statistical period. The statistical period refers to a time range in which the system performs analysis, which can be one day, one shift, one week, etc. The stable time interval is the part of the statistical period excluding all abnormal time intervals. For example, the statistical period is 24 hours on June 25, 2025, and the mixing station has two abnormal time intervals on this day, which are from 10:20 to 10:40 and from 14:30 to 14:50, respectively. Then the stable time interval is the other time period in 24 hours excluding the two abnormal time intervals.

[0058] In addition, the feature analysis module also needs to judge the length of the stable time interval. If the length of a single stable time interval is less than the set threshold, the time interval needs to be merged. The set threshold is also a time length value preset by the system, which can be set to 5 minutes, 8 minutes, etc. When the length of a stable time interval is less than the set threshold, it means that the stable time interval is too short and may not truly represent the stable state, or it is difficult to reflect the stable mixing state in such a short time. At this time, the feature analysis module will merge the stable time interval and the two adjacent abnormal time intervals into a new abnormal time interval.

[0059] For example, assuming that the length of a stable time interval is 4 minutes, the set threshold is 5 minutes, the previous abnormal time interval of the stable time interval is from 10:20 to 10:40, and the next abnormal time interval is from 11:00 to 11:20. The feature analysis module will merge the two abnormal time intervals and the stable time interval in between to form a new abnormal time interval, i.e., from 10:20 to 11:20. After this processing, the new abnormal time interval covers the original two abnormal time intervals and the short stable time interval in between, which is more consistent with the actual production situation and avoids analysis errors caused by the short stable time interval.

[0060] Throughout the entire processing process, the feature analysis module must accurately extract and process historical anomaly time points to ensure accurate identification of anomaly occurrence points. Furthermore, the setting of fixed durations and thresholds requires comprehensive consideration of factors such as the characteristics of mixing anomalies during ceramsite production, the operating characteristics of the equipment, and production process requirements. For example, different mixing stations may have different fixed durations and thresholds depending on the equipment type and material characteristics.

[0061] The characteristic analysis module divides the statistical period in this way to obtain characteristic time intervals, including abnormal time intervals and stable time intervals. The division of these time intervals is crucial for the subsequent work of the evaluation and judgment module. The evaluation and judgment module uses the characteristic time intervals to determine the operating activation status of the material sampling device and the collection response status of the detection component, thereby realizing intelligent assessment and corresponding control of the ceramsite material mixing degree.

[0062] In actual applications, the feature analysis module continuously processes new historical anomaly time points and updates the division between abnormal and stable time intervals. As production data accumulates, the system's time interval divisions become more accurate, more precisely reflecting the actual operating status of the mixed workstations and providing a more reliable time basis for subsequent analysis and judgment. Throughout the entire process, the feature analysis module strictly follows pre-set rules and procedures to ensure the scientific and rational division of time intervals, thereby guaranteeing the accuracy and effectiveness of the entire intelligent assessment system.

[0063] Embodiment 3: In this embodiment, the feature analysis module determines the color reference value and the granularity reference value based on the historical color data and the historical granularity data corresponding to the stable time interval.

[0064] The feature analysis module retrieves all historical color data for a single mixing station within a stable time interval from the parameter acquisition unit's record storage component. This historical color data is collected and stored by the color recognizer in the detection component during a stable time interval and includes quantifiable color parameters such as RGB values ​​and HSV values ​​at different time points.

[0065] The feature analysis module statistically processes this historical color data, first calculating the color mean. This is calculated by summing the values ​​for each color channel of all historical color data within a stable time interval and dividing the sum by the number of data points. For example, in the RGB color model, assuming there are n color data points within a stable time interval, each containing values ​​for the R, G, and B channels, the R mean is the sum of all R values ​​divided by n. The G and B mean values ​​are calculated in the same way. The resulting color mean reflects the average color state of the material within that stable time interval.

[0066] After calculating the color mean value, the feature analysis module also needs to determine the color fluctuation range. The color fluctuation range refers to the difference between the maximum value and the minimum value of the historical color data in the stable time interval. Taking the RGB model as an example, the maximum value and the minimum value of the R, G, and B channels are calculated respectively, and the fluctuation range of each channel is the difference between the maximum value and the minimum value of the channel, thereby obtaining the fluctuation range of the entire color data, which reflects the change amplitude of the material color in the stable time interval.

[0067] Based on the color mean value and the color fluctuation range, the feature analysis module determines the color reference value. The calculation method of the color reference value is the sum of the color mean value and the lower limit value of the color fluctuation range. The lower limit value of the color fluctuation range here is the minimum value of the color data, for example, in the RGB model, the lower limit value of the R channel is the minimum value of the channel, and the same applies to the G channel and the B channel. Adding the color mean value to the lower limit value of each channel, the result obtained is the color reference value, which is used to measure whether the subsequent real-time color data is within the normal range.

[0068] For the processing of historical particle size data, the feature analysis module uses a similar method. First, all historical particle size data in the stable time interval are obtained from the record storage component, which are collected by the particle size measuring instrument and contain parameters such as the average particle size and the particle size distribution range of the material particles. The feature analysis module performs statistics on these historical particle size data and calculates the particle size mean value, which is the average value of all particle size data in the stable time interval, reflecting the average level of the material particle size in the stable state.

[0069] The feature analysis module determines the particle size fluctuation range, which is the difference between the maximum value and the minimum value of the historical particle size data in the stable time interval, which reflects the fluctuation degree of the material particle size in the stable time interval. Then, according to the particle size mean value and the lower limit value of the particle size fluctuation range (i.e. the minimum particle size), the particle size reference value is calculated, which is the sum of the particle size mean value and the lower limit value of the particle size fluctuation range, which is used to judge whether the subsequent real-time particle size data is abnormal.

[0070] In actual operation, the feature analysis module needs to ensure that the historical color data and the historical particle size data obtained are from the accurately divided stable time interval. The stable time interval is determined by the feature analysis module processing the historical abnormal time points, i.e. excluding the part of the abnormal time interval in the statistical period, and after the merging processing of the short stable time interval, it is ensured that the stable time interval can truly reflect the stable state of the material mixing.

[0071] When processing data, the feature analysis module needs to filter and verify the data, excluding possible abnormal data points. For example, if a certain historical color data or historical particle size data deviates significantly from other data, it may be due to detection component failure or errors in the transmission process, so the feature analysis module will mark and exclude these data to ensure the accuracy of the calculated color mean, color fluctuation range, particle size mean and particle size fluctuation range.

[0072] The determination of color reference value and particle size reference value provides an important reference for the evaluation and judgment module. After receiving real-time color data and real-time particle size data, the evaluation and judgment module compares them with the corresponding reference values to determine whether there is a mixing abnormality tendency in the mixing station. For example, when the real-time color data is greater than the color reference value or the real-time particle size data is greater than the particle size reference value, the evaluation and judgment module considers that there may be a mixing abnormality and further analyzes the vibration information to confirm.

[0073] In addition, the feature analysis module will periodically update the color reference value and particle size reference value as production data accumulates. For example, set an update period, when the system runs to this period, the feature analysis module will reacquire the latest historical data in the stable time interval, recalculate the color mean, color fluctuation range, particle size mean and particle size fluctuation range, and update the reference value accordingly, to adapt to possible changes in material properties or process adjustments in the production process, ensuring that the reference value can always accurately reflect the current stable state of the material characteristics.

[0074] The entire process of determining the color reference value and particle size reference value strictly follows the pre-set statistical method and calculation rules, ensuring that each operation has a clear logic and basis. Through scientific processing of historical data in the stable time interval, the reference value obtained can accurately reflect the normal state of material mixing, providing a reliable judgment standard for subsequent intelligent evaluation of mixing degree, enabling the system to timely detect abnormal conditions in the mixing process and take appropriate measures to ensure the quality and stability of the ceramsite production.

[0075] In this embodiment, the feature analysis module needs to determine the vibration amplitude reference interval according to the amplitude range corresponding to the pre-set confidence in the amplitude distribution map of the stable time interval, and the construction of the amplitude distribution map needs to go through steps such as segmented sampling, frequency statistics and result merging. Taking a stable time interval of a mixing station as an example, assume that the stable time interval is from 8:00 to 9:00 on June 25, 2025, with a duration of 1 hour. The feature analysis module segments the vibration information in this interval, with a fixed length of 5 minutes for each segment, so the 1-hour stable time interval can be divided into 12 segments, each corresponding to 5 minutes of vibration information.

[0076] In the process of segment sampling, the feature analysis module retrieves the vibration information of the mixing station in the stable time interval from the record storage component of the parameter acquisition unit. These vibration information are collected by the vibration sensors in the detection assembly, stored in the form of electrical signals, and contain parameters such as vibration amplitude and frequency. For each 5-minute sampling data, the feature analysis module extracts the amplitude value, i.e. the numerical value of the vibration signal strength. For example, the first segment of sampling data is from 8:00 to 8:05, and the vibration sensor collects the amplitude value every second during this period, a total of 300 data points, each data point corresponds to an amplitude value, such as 0.5g, 0.6g, etc. (g is the unit of gravitational acceleration).

[0077] The feature analysis module performs frequency statistics on the amplitude values of each segment of sampling data. Taking the first 5-minute sampling data as an example, assuming that the amplitude value ranges from 0.4g to 0.8g, the feature analysis module divides the range into several amplitude intervals, for example, with an interval of 0.1g, it is divided into 0.4-0.5g, 0.5-0.6g, 0.6-0.7g, 0.7-0.8g. Then count the number of data points in each amplitude interval, i.e. the frequency of occurrence. For example, there are 120 data points in the 0.5-0.6g interval, so the frequency of this interval is 120 / 300=40%. In this way, the frequency statistics of this segment of sampling data are completed, and the amplitude interval and its corresponding frequency of occurrence are obtained.

[0078] After completing the frequency statistics of all 12 segments of sampling data, the feature analysis module combines the statistical results of each segment to form an amplitude distribution map covering the entire stable time interval. When combining, taking the amplitude value as the horizontal axis (unit: g) and the frequency of occurrence as the vertical axis (unit: %), the statistical results of each segment are superimposed or accumulated in the map. For example, for the amplitude interval 0.5-0.6g, the frequency of the first segment is 40%, the frequency of the second segment is 35%, the frequency of the third segment is 38%……the frequency of the twelfth segment is 42%, after combining, the total frequency of this interval in the entire amplitude distribution map can be calculated by weighted average or direct accumulation, depending on the system's preset combination rule.

[0079] After obtaining the amplitude distribution map, the feature analysis module determines the vibration amplitude reference interval according to the pre-set confidence. The pre-set confidence is a probability value preset by the system, for example, 90% or 95%, which represents the confidence of the vibration amplitude within a certain range. Taking the confidence of 95% as an example, the feature analysis module finds the amplitude range covering 95% frequency in the amplitude distribution map, which is the vibration amplitude reference interval. For example, from the amplitude distribution map, it can be seen that when the amplitude value is between 0.55g and 0.75g, the corresponding frequency accumulates to 95%, so 0.55g to 0.75g is the vibration amplitude reference interval of the stable time interval.

[0080] In practical applications, the fixed duration of segmented sampling can be adjusted according to the frequency of vibration information changes and system requirements. If the vibration frequency of the mixing device is high, the segmented duration can be shortened, such as 1 minute, to capture vibration changes more finely; if the vibration frequency is low, the segmented duration can be appropriately extended, such as 10 minutes, to reduce the amount of calculation. The selection of the pre-set confidence level also needs to be combined with the production process requirements and device characteristics, for example, for processes with higher requirements for mixing uniformity, a higher confidence level, such as 95%, can be set to more strictly define the normal vibration amplitude range.

[0081] In addition, when constructing the amplitude distribution map, the feature analysis module needs to filter abnormal amplitude data. For example, if there are amplitude values in a certain segment of sampling data that are significantly deviating from the normal range, it may be due to vibration sensor failure or external interference, and the feature analysis module will exclude these data according to the pre-set abnormal value judgment rule (such as values exceeding ±3 times the standard deviation of the mean) to ensure the accuracy of the amplitude distribution map.

[0082] After the vibration amplitude reference interval is determined, it will be used for the analysis of real-time vibration amplitude by the evaluation and judgment module. For example, when the evaluation and judgment module determines that there is a mixing abnormality tendency according to the real-time color data or particle size data, it will retrieve the current real-time vibration amplitude and compare it with the vibration amplitude reference interval. If the real-time vibration amplitude is not within the interval, it indicates that the vibration state during mixing is abnormal, which may indicate a problem with the mixing degree, and the evaluation and judgment module will adjust the running activation state of the material sampling device accordingly, such as switching from a dormant state to an active state, to further sample and analyze.

[0083] As the production process continues, the feature analysis module will periodically update the amplitude distribution map and the vibration amplitude reference interval. For example, at the end of each day or production shift, the feature analysis module will re-sample, statistically analyze frequencies, and merge based on the latest vibration information within the stable time interval to generate a new amplitude distribution map, and update the vibration amplitude reference interval accordingly, to adapt to changes in vibration state caused by factors such as equipment wear and tear and changes in material properties, ensuring that the vibration amplitude reference interval accurately reflects the current normal vibration state at all times.

[0084] Throughout the process, the feature analysis module systematically processes vibration information within the stable time interval, from segmented sampling to frequency statistics, to map construction and reference interval determination, each step following clear logic and pre-set rules to ensure the scientificity and reliability of the vibration amplitude reference interval. This analysis method based on actual production data can effectively capture the characteristics of the vibration state during mixing, providing key vibration characteristic references for intelligent evaluation of the mixing degree, enabling the system to discover potential mixing abnormalities in a timely manner through changes in vibration information, thereby achieving accurate evaluation and control of the mixing degree of ceramsite materials.

[0085] In this embodiment, the evaluation and decision module needs to control the running activation state of the material sampling device and the collection response state of the detection assembly according to the characteristic time interval, real-time color data, real-time particle size data, and real-time vibration amplitude, so as to realize intelligent evaluation of the mixing degree of the ceramsite material. The implementation manner is described in detail below in combination with specific examples.

[0086] Suppose that the statistical cycle of a mixing station on June 25, 2025, 10:00, the characteristic analysis module determines that the abnormal time interval of the station is from 10:15 to 10:25, and the remaining time is the stable time interval. When the time is 10:16, that is, in the abnormal time interval, the evaluation and decision module sets the material sampling device of the corresponding station to the activated state, and controls the collection response state of the detection assembly to collect only color data, particle size data, and vibration data. At this time, the material sampling device is started, and the color identifier, particle size measuring instrument, and vibration sensor in the detection assembly start working, and real-time collection of color data, particle size data, and vibration information of the material is performed, and these data are transmitted to the evaluation and decision module.

[0087] When the time is 10:30, in the stable time interval, the evaluation and decision module switches the material sampling device to the dormant state, and the collection response state of the detection assembly is still to collect color data, particle size data, and vibration data, while controlling the characteristic analysis module to determine the color reference value and the particle size reference value. For example, the characteristic analysis module obtains the historical color data and the historical particle size data in the stable time interval from the record storage component, calculates the color mean value as RGB(150, 160, 170), the lower limit value of the color fluctuation range as RGB(140, 150, 160), and the color reference value as RGB(150+140, 160+150, 170+160) that is RGB(290, 310, 330) (here, only an example calculation manner is shown, and the calculation of the color value in practice needs to be based on a specific model); the particle size mean value is 50 μm, the lower limit value of the particle size fluctuation range is 40 μm, and the particle size reference value is 50+40=90 μm.

[0088] When the material sampling device is in the dormant state, the evaluation and decision module determines the mixing abnormality tendency of the station in combination with the real-time color data and the real-time particle size data. Suppose that at 10:40, the real-time color data is RGB(300, 320, 340), and the real-time particle size data is 95 μm. Since the real-time color data is greater than the color reference value RGB(290, 310, 330), and the real-time particle size data is greater than the particle size reference value 90 μm, the evaluation and decision module determines that there is a mixing abnormality phenomenon in the mixing station, and adjusts the material sampling device to the activated state to sample and analyze the material.

[0089] For another example, at 10:45, the real-time color data is RGB(295, 315, 335) and the real-time granularity data is 85 μm. At this time, the real-time color data is greater than the color reference value, while the real-time granularity data is less than the granularity reference value, the evaluation and determination module determines that the mixing station has a tendency of mixing abnormality, and determines to analyze the vibration information.

[0090] When the evaluation and determination module determines that there is a tendency of mixing abnormality, the feature analysis module is controlled to determine the vibration amplitude reference interval. Assuming that the feature analysis module determines the vibration amplitude reference interval to be 0.6 g to 0.8 g at a confidence level of 95% according to the amplitude distribution graph of the stable time interval. At this time, the evaluation and determination module obtains the real-time vibration amplitude. If the real-time vibration amplitude is 0.5 g, which is not in the reference interval, the evaluation and determination module determines to change the material sampling device from the dormant state to the activated state. If the real-time vibration amplitude is 0.7 g, which is in the reference interval, the evaluation and determination module determines not to change the running activated state of the material sampling device, so as to keep the material sampling device in the dormant state.

[0091] In another example, the material sampling device of a mixing station is in the dormant state in the stable time interval, and the detection assembly continuously collects data. Assuming that the real-time color data is RGB(280, 300, 320), which is less than the color reference value RGB(290, 310, 330), and the real-time granularity data is 80 μm, which is less than the granularity reference value 90 μm, the evaluation and determination module determines that the mixing station does not have a tendency of mixing abnormality, the material sampling device keeps in the dormant state, and the detection assembly continues to collect data normally.

[0092] During the working process of the evaluation and determination module, the change of the feature time interval is monitored in real time. If the feature analysis module updates the feature time interval at a certain time, the evaluation and determination module will immediately adjust the running activated state of the material sampling device and the collection response state of the detection assembly according to the new feature time interval. For example, if the original stable time interval is re-divided into an abnormal time interval, the evaluation and determination module will immediately switch the material sampling device from the dormant state to the activated state, and the collection response state of the detection assembly becomes only collecting data.

[0093] In addition, the evaluation and determination module also dynamically adjusts the state of the material sampling device according to the change of the real-time vibration amplitude. When the material sampling device is in the activated state, if the real-time color data and the real-time granularity data are both less than or equal to the reference value after a period of sampling analysis, and the real-time vibration amplitude is in the reference interval, the evaluation and determination module determines that the mixing abnormality disappears, and switches the material sampling device back to the dormant state, so as to save energy and equipment loss.

[0094] During the whole evaluation and judgment process, the evaluation and judgment module strictly follows the preset rules and logic to make judgments and control, ensuring the accurate and timely evaluation of the state of the mixing station. Through the analysis of the characteristic time interval, the comparison of the real-time data with the reference value and the auxiliary judgment of the vibration information, the evaluation and judgment module can accurately identify the mixed abnormal tendency and abnormal phenomenon, and through the adjustment of the running state of the material sampling device, it can realize the intelligent evaluation and control of the mixing degree of the ceramsite material, and ensure the stability of the ceramsite production process and the product quality.

[0095] It should be noted that, in this document, relational terms such as first and second and the like can be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Moreover, the terms "comprises", "comprising", or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but can include other elements not expressly listed or inherent to such process, method, article, or apparatus.

[0096] While embodiments of the application have been shown and described, it is to be understood that the embodiments described are merely exemplary and that changes in form and detail can be made without departing from the spirit and scope of the application. The scope of the application is defined by the appended claims and their equivalents.

Claims

1. A system for intelligently evaluating the degree of mixing of a ceramisite material, characterized in that, The application relates to a ceramic production line parameter analysis system. The system comprises: a material sampling device for obtaining material samples of each mixing station of the ceramic production line; a parameter acquisition unit arranged at each mixing station of the ceramic production line, comprising a record storage component for storing historical mixing abnormal information of each mixing station, and a plurality of detection components for acquiring material color data, particle size data and mixing process vibration information of each mixing station, wherein the historical mixing abnormal information comprises historical abnormal time points, historical color data, historical particle size data and historical vibration information; a feature analysis module connected to the parameter acquisition unit, for dividing a statistical period into a plurality of feature time intervals according to the historical mixing abnormal information of a single mixing station, determining color reference values and particle size reference values according to historical color data and historical particle size data of a single mixing station, constructing a time-domain vibration waveform diagram according to the historical vibration information to determine an amplitude distribution atlas of a stable period, and determining a vibration amplitude reference interval according to the amplitude distribution atlas. The feature time intervals comprise abnormal time intervals and stable time intervals.

2. The intelligent evaluation system for the mixing degree of the ceramsite material according to claim 1, characterized in that, an evaluation and determination module connected to each material sampling device, parameter acquisition unit and feature analysis module, for determining the operation activation state of the material sampling device and the acquisition response state of the detection components of a corresponding mixing station according to the feature time intervals of a single mixing station, determining a mixing abnormality tendency according to real-time color data and real-time particle size data of a single mixing station to determine whether to analyze historical vibration information, and determining whether to change the operation activation state of the corresponding material sampling device according to the mixing abnormality tendency and real-time vibration amplitude.

3. The intelligent evaluation system for the mixing degree of ceramsite material according to claim 1, characterized in that, Each mixing station is provided with one detection component, and a single detection component comprises a color identifier, a particle size measuring instrument and a vibration sensor. The feature analysis module determines the abnormal occurrence point of a single mixing station according to the historical abnormal time points, determines a fixed time interval centered on the abnormal occurrence point as an abnormal time interval, and determines the absolute complement of the abnormal time interval in the statistical period as a stable time interval.

4. The intelligent evaluation system for the mixing degree of ceramsite material according to claim 1, characterized in that, If the length of a single stable time interval is less than a set threshold, the stable time interval and the two adjacent abnormal time intervals are merged into one abnormal time interval. The feature analysis module determines color reference values according to the historical color data corresponding to the stable time interval, and determines particle size reference values according to the historical particle size data corresponding to the stable time interval.

5. The intelligent evaluation system for the mixing degree of ceramsite material according to claim 1, characterized in that, The color reference values are the sum of the color mean value and the lower limit value of the color fluctuation range, and the particle size reference values are the sum of the particle size mean value and the lower limit value of the particle size fluctuation range. The feature analysis module determines the vibration amplitude reference interval according to the amplitude range corresponding to a preset confidence in the amplitude distribution atlas of the stable time interval. The horizontal axis of the amplitude distribution atlas is the vibration signal intensity, and the vertical axis of the amplitude distribution atlas is the occurrence frequency.

6. The intelligent evaluation system for the mixing degree of ceramsite material according to claim 1, characterized in that, The evaluation decision module determines the operation activation state of the corresponding work station material sampling device and the collection response state of the detection assembly according to the characteristic time interval of the single mixing work station, including: If the characteristic time interval is an abnormal time interval, the corresponding work station material sampling device is in an activated state and the collection response state of the detection assembly is only collecting color data, particle size data and vibration data; If the characteristic time interval is a stable time interval, the corresponding work station material sampling device is in a dormant state and the collection response state of the detection assembly is collecting color data, particle size data and vibration data, and the characteristic analysis module is controlled to determine the color reference value and the particle size reference value. The operation activation state includes an activated state and a dormant state.

7. The intelligent evaluation system of the ceramsite material mixing degree according to claim 6, characterized in that, The evaluation decision module determines the mixing abnormality tendency of the work station according to the determination result that the single mixing work station material sampling device is in a dormant state, in combination with the real-time color data and the real-time particle size data, including: If the real-time color data is less than or equal to the color reference value and the real-time particle size data is less than or equal to the particle size reference value, it is determined that there is no mixing abnormality tendency in the corresponding mixing work station; If the real-time color data is greater than the color reference value and the real-time particle size data is greater than the particle size reference value, it is determined that there is a mixing abnormality phenomenon in the corresponding mixing work station and the corresponding material sampling device is adjusted to an activated state.

8. The intelligent evaluation system of the ceramsite material mixing degree according to claim 7, characterized in that, The evaluation decision module determines the mixing abnormality tendency of the work station according to the determination result that the single mixing work station material sampling device is in a dormant state, in combination with the real-time color data and the real-time particle size data, further including: If the real-time color data is greater than the color reference value or the real-time particle size data is greater than the particle size reference value, it is determined that there is a mixing abnormality tendency in the corresponding mixing work station and it is determined to analyze the vibration information.

9. The intelligent evaluation system of the ceramsite material mixing degree according to claim 8, characterized in that, The evaluation decision module controls the characteristic analysis module to determine the vibration amplitude reference interval according to the determination result that there is a mixing abnormality tendency, and determines whether to change the operation activation state of the material sampling device corresponding to the mixing work station according to the real-time vibration amplitude; If the real-time vibration amplitude is not in the vibration amplitude reference interval, it is determined that the material sampling device corresponding to the mixing work station is changed from a dormant state to an activated state; If the real-time vibration amplitude is in the vibration amplitude reference interval, it is determined that the operation activation state of the material sampling device corresponding to the mixing work station is not changed to keep it in a dormant state.

10. The intelligent evaluation system for the mixing degree of ceramsite material according to claim 5, characterized in that, When the characteristic analysis module determines the amplitude distribution map of the stable time interval, first, the vibration information in the stable time interval is segmented and sampled, the length of each segment of sampling data is a fixed time length, then the frequency of the amplitude values of each segment of sampling data is counted, and finally the statistical results of each segment are combined to form an amplitude distribution map covering the entire stable time interval.

Images