A safety monitoring method and system for the production of compound probiotics

Abstract

The present invention relates to the technical field of mixing or stirring treatment, and particularly relates to a safety monitoring method and system for the production of compound probiotics, including: stirring the mono- and diglycerol fatty acid esters and medium-chain triglycerides or sunflower oil in the compound probiotic formula multiple times, calculating the torque cycle coefficient, stability coefficient and generalization degree of the mixer blades per week to characterize the fluctuations, stability and generalization ability during the stirring process. Through the cluster analysis of the generalization degree of each week, different clusters are formed. When the torque cycle coefficient of the current week exceeds the range of the cluster it belongs to, the heating temperature is automatically adjusted to ensure the stability and quality of the stirring process. The present invention solves the problem of unstable production efficiency during the production process of compound probiotics.

Description

Technical Field

[0001] The present invention relates to the technical field of mixing or stirring treatment. More specifically, the present invention relates to a method and system for safety monitoring in the production of compound probiotics. Background Art

[0002] The existing production process of compound probiotics mainly relies on traditional monitoring and control methods, which are mostly based on physical sensors and simple control systems, and mainly monitor some basic physical parameters such as temperature, pressure and flow rate. These traditional means can ensure the stability of the production process to a certain extent, but they usually cannot cope with the complexity and dynamic changes in the reaction process. For example, in the multi-component reaction process, factors such as the reaction state of raw materials, reaction rate, and conversion rate of raw materials may fluctuate to varying degrees at each moment.

[0003] In addition, traditional monitoring systems mostly rely on simplified mathematical models or empirical formulas for process prediction and control. These models often have limitations when facing complex and variable reaction conditions. They cannot comprehensively consider all variables in the reaction process, such as stirring efficiency, reaction uniformity, and material distribution, and these factors directly affect the sufficiency of the reaction and the final quality of the product. Due to the lack of a comprehensive and accurate model of the reaction process, traditional control systems are often slow to respond when adjusting key parameters such as stirring speed, reaction temperature, and pressure in real time, and it is difficult to perform fine control on the production process based on real-time data, resulting in low stirring efficiency, uneven reaction, and ultimately affecting the quality and yield of compound probiotics.

[0004] However, the existing technical solutions only implement safety monitoring of the production process based on traditional network models, and fail to fully consider the limitations of traditional models. Traditional network models usually rely on fixed rules and relatively simple feature extraction methods, and it is difficult to cope with the variable factors and complex situations that may occur in the production process, such as dynamic changes in the production environment, fluctuations in process parameters, and potential abnormal situations. Traditional models cannot achieve efficient and accurate monitoring when facing different types of safety hazards in the production process of compound probiotics, resulting in unstable production efficiency in the production process of compound probiotics. Summary of the Invention

[0005] To solve the problem of unstable production efficiency in the production process of compound probiotics proposed in the above background art, the present invention provides solutions in the following aspects.

[0006] In a first aspect, the present invention provides a method for monitoring the production safety of a composite probiotic, comprising: stirring mono- and diglycerol fatty acid esters with medium-chain triglycerides or sunflower oil in a composite probiotic formula for multiple times, obtaining a torque sequence of a mixer blade per rotation of one circle during each stirring; calculating the torque sequence of the mixer blade per rotation of the first circle; and calculating the torque sequence of the mixer blade per rotation of the first circle. Torque cycle coefficient The torque cycle coefficient characterizes the rotation of the mixer blade. The degree of fluctuation of the mixer blades Weekly stability factor , , where 、 、 Respectively Week, first Week, first The torque cycle coefficient of the cycle, is the total number of revolutions of the mixer blades; calculate the number of revolutions of the mixer blades. Zhou's generalization , , where To obtain the minimum function, To obtain the maximum value function, To complete During the first mixing process, the mixer blades The set of torque cycle coefficients during rotation, is the hyperbolic tangent function, 、 Complete the first Stir, The first stirring The stability coefficient of the mixer blades, is the total number of stirring times completed; the generalization degree of each week is clustered to obtain multiple clusters; if the torque cycle coefficient of the current week exceeds the range of the cluster to which it belongs, the heating temperature is adjusted until the torque cycle coefficient of the current week is within the range of the cluster to which it belongs.

[0007] The above technical solution can achieve precise monitoring and control of the stirring process by systematically analyzing the torque data of the mixer blades rotating weekly during the production process of compound probiotics, calculating the torque cycle coefficient, stability coefficient, and generalization degree, and using the method of ordered sample clustering to identify abnormal fluctuations during the stirring process. When the torque cycle coefficient of the current week exceeds the clustering cluster range, the abnormality is quickly corrected by adjusting the heating temperature to keep the stirring process stable and ensure the consistency of the production quality of compound probiotics. At the same time, by analyzing the historical data of multiple stirrings, this method significantly improves the dynamic adaptability to the production process, provides data support for optimizing the production process, and improves production efficiency and the stability of the finished product.

[0008] Further, the torque cycle coefficient of the rotation of the mixer blades in the th week , , where is the number of torques in the torque sequence of the th week, is the number of torques in the torque sequence of the th week, is the mutual information coefficient between the torque sequence of the th week and the torque sequence of the th week.

[0009] The above technical solution can more precisely quantify the correlation between the torque sequences of each week of the rotation of the mixer blades by introducing the mutual information coefficient, thereby reflecting the degree of fluctuation during the stirring process. The mutual information coefficient can effectively capture the similarity or difference between different weeks, making the calculation of the torque cycle coefficient more detailed and reliable. This method not only enhances the monitoring ability of the dynamic changes in the stirring process but also improves the ability to identify and intervene in potential abnormal fluctuations at an early stage, thus further ensuring the stability of the production process and the accuracy of quality control.

[0010] Further, the torque cycle coefficient of the rotation of the mixer blades in the th week , , where is the number of torques in the torque sequence of the th week, is the number of torques in the torque sequence of the th week, is the similarity between the torque sequence of the th week and the torque sequence of the th week.

[0011] The above technical solution calculates the similarity between the torque sequences of each week and the adjacent weeks through the dynamic time warping algorithm, which is used to determine the torque cycle coefficient of each week of the rotation of the mixer fan blades, and can more accurately capture the subtle changes and non-linear correspondence relationships between different weeks. The dynamic time warping algorithm has the ability to adapt to the differences in the length of time series and non-stationary characteristics, making the calculation of similarity more in line with the complex dynamic characteristics in the actual mixing process. By combining the proportional relationship between similarity and the number of sequences, the description accuracy of the torque cycle coefficient is further improved, which helps to more sensitively identify abnormal fluctuations, optimize the control of mixing parameters, and thus achieve higher-precision monitoring and quality assurance of the production process.

[0012] Further, it also includes removing the torque sequences with a stability coefficient greater than the set threshold.

[0013] The above technical solution can effectively filter out abnormal data in the mixing process by removing the torque sequences with a stability coefficient greater than the set threshold, avoiding the influence of these abnormal values on the torque cycle coefficient, generalization degree, and clustering results, thereby improving the accuracy and reliability of data analysis. This not only improves the accuracy of mixing process monitoring but also further optimizes the parameter control strategy, helps to maintain the stability of the production process, reduces the risk of quality fluctuations caused by abnormal fluctuations, and ultimately ensures the stability and consistency of the compound probiotic product.

[0014] Further, the similarity is obtained through the dynamic time warping algorithm.

[0015] Further, a torque sensor is used to collect the torque sequence of each rotation of the mixer fan blade during the mixing of mono- and diglycerides and medium-chain triglycerides or sunflower oil in the compound probiotic formula.

[0016] The above technical solution can reflect the dynamic mechanical characteristics during the mixing of mono- and diglycerides and medium-chain triglycerides or sunflower oil in the compound probiotic formula in real time and accurately by using a torque sensor to collect the torque sequence of each rotation of the mixer fan blade, providing high-precision and continuous basic data support for subsequent data analysis and process control. This high-frequency and high-precision data collection method helps to comprehensively monitor the changes in the force on the fan blade during the mixing process, timely identify abnormal fluctuations, and provide a reliable basis for optimizing the mixing process parameters and improving the production quality, thereby enhancing the stability and efficiency of compound probiotic production.

[0017] Further, the torque sensor is a strain type torque sensor or a rotary torque sensor.

[0018] Further, it also includes performing data cleaning and data normalization processing on the torque sequence.

[0019] Further, the clustering is an ordered sample clustering.

[0020] In a second aspect, the present invention provides a safety monitoring system for compound probiotic production, including a memory and a processor. Computer program instructions are stored in the memory, and when the computer program instructions are executed by the processor, a safety monitoring method for compound probiotic production as described in any one of the above is implemented.

[0021] The beneficial effects of the present invention are as follows:

[0022] By collecting the torque sequence during the stirring process in real time and combining various advanced calculation models, such as the torque cycle coefficient, stability coefficient, generalization degree, etc., the present invention effectively reflects the mechanical fluctuations and stability of the mixer fan blade during each rotation. By dynamically adjusting the heating temperature to ensure that the torque cycle coefficient is within the specified range, the present invention can achieve real-time optimization and anomaly detection of the production process, avoiding the risks of production fluctuations and quality fluctuations. Especially by introducing data cleaning, normalization processing, and the dynamic time warping algorithm for accurate calculation of similarity, the quality and accuracy of the monitoring data are improved, the adaptability to parameter fluctuations in different production environments is enhanced, thereby improving the production efficiency of compound probiotics and the consistency of products. The present invention also provides intelligent and automated monitoring and control means for the production process through big data clustering analysis, ensuring the stability of the production process and the high quality of the final product. Description of the Drawings

[0023] Figure 1 is a flowchart schematically showing a safety monitoring method for compound probiotic production according to an embodiment of the present invention;

[0024] Figure 2 is a block diagram schematically showing the structure of a safety monitoring system for compound probiotic production according to an embodiment of the present invention. Detailed Embodiments

[0025] An embodiment of a safety monitoring method for compound probiotic production.

[0026] As Figure 1 shown, a flowchart of a safety monitoring method for compound probiotic production according to an embodiment of the present invention includes the following steps:

[0027] S1: Stir the mono- and diglycerol fatty acid esters in the compound probiotic formula with medium-chain triglycerides or sunflower oil multiple times to obtain the torque sequence of each rotation of the mixer fan blade each time.

[0028] In one embodiment, during the stirring process of mono- and diglycerides and medium-chain triglycerides or sunflower oil in the compound probiotic formula, torque is one of the important parameters to measure the operating state of the stirrer. Its change directly reflects the rheological properties of the materials during the stirring process and the working load of the stirrer. Through multiple stirrings, a torque sequence for each rotation of the stirrer blades can be obtained. This sequence provides detailed torque fluctuation information and can reveal various dynamic changes during the stirring process. To ensure the accuracy and reliability of the collected data, high-precision torque sensors, such as strain torque sensors or rotary torque sensors, are selected during the implementation process. These sensors have very high sensitivity and accuracy, can work stably in a complex dynamic stirring environment, monitor the change of torque in real time, and convert it into a digital signal;

[0029] The strain torque sensor can calculate the torque magnitude by measuring the strain change caused by torque, has very high precision and strong anti-interference ability, and is very suitable for the strict requirements of precision in the high-load stirring process; while the rotary torque sensor can directly measure the torque of the rotating shaft and can achieve continuous dynamic monitoring, which is suitable for the detection of high-speed rotation and high-frequency changes.

[0030] However, during the actual acquisition process, the data quality may be affected by various factors, such as environmental noise, sensor error, mechanical vibration, etc. These factors may cause outliers or inconsistencies in the collected torque data. Therefore, in the data processing stage, strict data cleaning and normalization processing must be performed on the original torque sequence. The purpose of data cleaning is to eliminate noise, remove invalid or abnormal data points, and ensure that only the data representing the normal working state of the stirrer is retained. Data cleaning can not only eliminate meaningless fluctuations caused by instantaneous failures or interferences, but also effectively prevent data deviation caused by extreme values or errors, thereby improving the representativeness and accuracy of the data.

[0031] Since the absolute value of torque may be different in different stirring cycles, the normalization process adjusts the torque data of different stirring processes to a unified scale range. This process helps to eliminate data deviation caused by different stirring conditions, such as different materials, stirring speeds, etc., and ensures the comparability and consistency of the data.

[0032] S2: Calculate the torque cycle coefficient, stability coefficient and generalization degree for each rotation of the stirrer blades.

[0033] In one embodiment, calculate the torque cycle coefficient for the th rotation of the stirrer blades , , where is the number of torques in the torque sequence for the th week, For the The number of torques in the torque sequence of the week, For the The torque sequence of the first week is Mutual information coefficient of Zhou's torque series.

[0034] By calculating the torque cycle coefficient of each rotation cycle of the mixer blade and incorporating the mutual information coefficient between the torque sequences of each cycle and its adjacent cycles into the calculation formula, the correlation and information sharing between the two cycles can be more comprehensively quantified. By combining the ratio of the number of torque sequences, the calculation results not only reflect the correlation between the sequences, but can also dynamically adapt to changes in the length of the torque data, thereby improving the accuracy of the characterization of the fluctuation characteristics during the mixing process. This method enhances the ability to identify abnormal fluctuations, provides a solid data foundation for the refined monitoring and parameter optimization of the mixing process, and further improves the quality and efficiency of compound probiotic production.

[0035] Calculate the number of rotations of the mixer blades Weekly stability factor , , where 、 、 Respectively Week, first Week, first The torque cycle coefficient of the cycle, is the total number of revolutions of the mixer blades, Rotate the blender blades Torque cycle coefficient of the week;

[0036] By calculating the stability coefficient of each blender blade rotation, the stability of the blending process can be dynamically quantified. By weighing current fluctuations against historical cumulative characteristics, this method not only improves sensitivity to the steady state of the blending process but also enhances adaptability to long-term trends, helping to quickly identify potential abnormal fluctuations. As a result, the present invention further optimizes the ability to accurately monitor and control the blending process, improving production efficiency and the quality consistency of composite probiotic products.

[0037] Calculate the rotation speed of the mixer blade Zhou's generalization , , where To obtain the minimum function, To obtain the maximum value function, To complete During the first mixing process, the mixer blades The set of torque cycle coefficients during rotation, is the hyperbolic tangent function, 、 Complete the first Stir, The first stirring The stability coefficient of the mixer blades, is the total number of times stirring is completed;

[0038] By calculating the degree of generalization for each cycle of the mixer's blade rotation and combining the ratio of the minimum to maximum torque cycle coefficient to determine the dynamic trend of the stability coefficient during multiple mixing cycles, and using a hyperbolic tangent function for smoothing, this method comprehensively quantifies the impact of different mixing cycles on the current mixing process. This method effectively integrates historical and current data from multiple mixing cycles, capturing both short-term fluctuations and long-term stability trends. This improves the overall assessment of the mixing state, providing an important basis for optimizing and controlling the mixing process, and significantly enhancing production quality and the level of intelligent process control.

[0039] In another embodiment, the number of rotations of the blender blades is calculated. Torque cycle coefficient , , where For the The number of torques in the torque sequence of the week, For the The number of torques in the torque sequence of the week, For the The torque sequence of the first week is The similarity of the torque sequences of the weeks is obtained by the dynamic time warping algorithm.

[0040] By introducing a dynamic time warping algorithm to calculate the similarity of torque sequences and combining it with the ratio of the number of sequences to determine the torque cycle coefficient for each rotation of the blender blade, the authors were able to effectively address the issues of torque sequence length differences and nonlinear variations, significantly improving the ability to accurately characterize changes in mechanical properties during the blending process. The dynamic time warping algorithm can capture the complex dynamic correspondence between torque sequences at different rotational times, making the calculated results more consistent with the changing characteristics of the actual blending process. This enhances the sensitivity for detecting abnormal fluctuations, optimizes the monitoring accuracy and parameter control strategies of the blending process, and provides reliable data support for improving the quality and efficiency of compound probiotic production.

[0041] S3: Clustering the generalization degree of each week to obtain multiple clusters; if the torque cycle coefficient of the current week exceeds the range of the cluster to which it belongs, adjusting the heating temperature until the torque cycle coefficient returns to the range of the cluster to which it belongs.

[0042] In one embodiment, the clustering is ordered sample clustering. Ordered sample clustering analysis is performed on the generalization degree of each week. Using the clustering algorithm, the torque cycle coefficients of different weeks are divided into multiple clusters, such as , and . If the torque cycle coefficient of the current week exceeds the range of the cluster to which it belongs, the rheological properties and mechanical response of the material during stirring are changed by adjusting the heating temperature, for example, increasing the heating temperature, until the torque cycle coefficient returns to the range of the cluster to which it belongs. This method can effectively avoid process out-of-control caused by abnormal torque fluctuations through real-time monitoring and dynamic regulation, thereby significantly improving the stability of the stirring process and the consistency of the final product quality. This method can effectively avoid process out-of-control caused by abnormal torque fluctuations through real-time monitoring and dynamic regulation, thereby significantly improving the stability of the stirring process and the consistency of the final product quality.

[0043] The solution of the present invention can monitor the operating state of the mixer in the production process of compound probiotics in real time by introducing a big data processing method based on torque sequence analysis, effectively identify the law of torque fluctuations and accurately calculate the stability and generalization degree. By dynamically calculating and clustering the torque cycle coefficients of each week, and combining the dynamic time warping algorithm to obtain the similarity, it can automatically identify abnormal fluctuations in the stirring process and adjust the heating temperature in time, thereby ensuring the stability and efficiency of the stirring process. By data cleaning and normalization processing, abnormal data that does not meet the stability requirements is removed, making the analysis results more accurate. This method can significantly improve the intelligent monitoring level of the production process, reduce human intervention, improve the consistency of product quality, and at the same time improve production efficiency and reduce energy consumption.

[0044] An embodiment of a compound probiotic production safety monitoring system:

[0045] As Figure 2 shown, a structural block diagram of a compound probiotic production safety monitoring system according to an embodiment of the present invention includes a processor and a memory.

[0046] The present invention also provides a compound probiotic production safety monitoring system. As Figure 2 shown, the system includes a processor and a memory, and the memory stores computer program instructions. When the computer program instructions are executed by the processor, a compound probiotic production safety monitoring method according to the above of the present invention is implemented.

[0047] The compound probiotic production safety monitoring system further includes other components well known to those skilled in the art such as a communication interface, and its settings and functions are known in the art, so they will not be described herein again.

[0048] In the present invention, the aforementioned memory may be any tangible medium that contains or stores a program, which can be used by or in conjunction with an instruction execution system, apparatus, or device. For example, the computer-readable storage medium may be any suitable magnetic storage medium or magneto-optical storage medium, such as resistive random access memory (RRAM), dynamic random access memory (DRAM), static random-access memory (SRAM), enhanced dynamic random access memory (EDRAM), high-bandwidth memory (HBM), hybrid memory cube (HMC), etc., or any other medium that can be used to store the required information and can be accessed by an application program, module, or both. Any such computer storage medium may be part of the device or accessible or connectable to the device. Any application or module described in the present invention may be implemented using computer-readable / executable instructions stored or otherwise maintained by such a computer-readable medium.

[0049] In the description of this specification, the meanings of "a plurality of" and "several" are at least two, for example, two, three, or more, etc., unless otherwise specifically defined.

[0050] Although this specification has shown and described multiple embodiments of the present invention, it will be apparent to those skilled in the art that such embodiments are provided by way of example only. Those skilled in the art will think of many changes, alterations, and alternative ways without departing from the spirit and concept of the present invention. It should be understood that various alternatives to the embodiments of the present invention described herein may be employed in the practice of the present invention.

Claims

1. A safety monitoring method for the production of compound probiotics, characterized in that, include: The mono- and diglycerol fatty acid esters in the composite probiotic formula are stirred with medium-chain triglycerides or sunflower oil multiple times to obtain a torque sequence per rotation of the blender blade during each stirring operation; Calculate the torque cycle coefficient of the blender blade rotation for the th week , where the torque cycle coefficient characterizes the fluctuation degree of the blender blade rotation for the th week. Among them or . In the formula, is the number of torques in the torque sequence for the th week, is the number of torques in the torque sequence for the th week, is the mutual information coefficient between the torque sequence for the th week and the torque sequence for the th week, is the similarity between the torque sequence for the th week and the torque sequence for the th week, and the similarity is obtained by the dynamic time warping algorithm; Calculate the stability coefficient of the blender blade rotation for the th week , . In the formula, , , are the torque cycle coefficients for the th week, the th week, and the th week respectively, is the total number of weeks of the blender blade rotation; Calculate the generalization degree of the blender blade rotation at the th week , , where is the minimum value function, is the maximum value function, is to complete times of the torque cycle coefficient set of the blender blade during the th week rotation in the stirring process, is the hyperbolic tangent function, , are respectively the stability coefficients of the blender blade at the th stirring and the th stirring at the th week, is the total number of times of completing the stirring; The generalization degree of each week is clustered to obtain multiple clusters; if the torque cycle coefficient of the current week exceeds the range of the cluster to which it belongs, the heating temperature is adjusted until the torque cycle coefficient of the current week is within the range of the cluster to which it belongs.

2. The safety monitoring method for the production of a composite probiotic according to claim 1, characterized in that, The method further includes eliminating torque sequences whose stability coefficient is greater than a set threshold.

3. The safety monitoring method for the production of a composite probiotic according to claim 1, characterized in that, The torque sensor was used to collect the torque sequence of the blender blade during each rotation of the blender during the blending process of mono- and diglycerol fatty acid esters with medium-chain triglycerides or sunflower oil in the compound probiotic formula.

4. The safety monitoring method for the production of a composite probiotic according to claim 3, characterized in that, The torque sensor is a strain gauge torque sensor or a rotational torque sensor.

5. A safety monitoring method for the production of a composite probiotic according to claim 1, characterized in that, The method also includes performing data cleaning and data normalization processing on the torque sequence.

6. The safety monitoring method for the production of a composite probiotic according to claim 1, characterized in that, The clustering is ordered sample clustering.

7. A safety monitoring system for the production of compound probiotics, characterized in that, The method comprises a memory and a processor, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, a composite probiotic production safety monitoring method according to any one of claims 1 to 6 is implemented.

Images