Automation Control and Information Management System for the New Generation Juice Production Line

Abstract

The present invention relates to the field of data processing and big data technology, and specifically relates to a new generation of automated control and information management system for a fruit juice production line, including: a control module, a data acquisition module, a data analysis and processing module, and a human-machine interaction module. The data acquisition module collects multi-link and multi-dimensional data and transmits it to the analysis module. The analysis module processes the data through steps such as cleaning and standardization, sets short-term and long-term dynamic thresholds according to the sliding window technology, and compares them in real time to trigger early warnings. The control module adjusts the equipment according to the early warnings, and the human-machine interaction module provides managers with operations and information viewing; the purpose of the present invention is to solve the problems of non-intelligent control of fruit juice production line equipment, ineffective management of production data, and difficulty in timely monitoring and early warning of product quality.

Description

Technical Field

[0001] The present invention belongs to the technical field of data processing and big data, and relates to an automated control and information management system for a new generation of fruit juice production lines. Background Art

[0002] In the field of fruit juice production, with the intensification of market competition and the continuous improvement of consumers' requirements for fruit juice quality and safety, fruit juice production enterprises are facing huge challenges. Modern fruit juice production lines cover multiple complex links such as raw material processing, juicing, blending, sterilization, and filling. Each link is interrelated and affects each other. To ensure stable product quality, improve production efficiency, and reduce costs, it is crucial to achieve automated control and efficient information management of the production line. At present, automated technology and information technology are widely used in industrial production, providing technical support for the upgrading and transformation of fruit juice production lines. However, in actual applications, there are still many problems to be solved in fruit juice production lines.

[0003] The equipment control of traditional fruit juice production lines mostly relies on manual operation or simple program setting. Manual operation is prone to errors and it is difficult to adjust the equipment operation parameters in a timely and accurate manner according to the actual production situation. For example, in the juicing process, it is impossible to dynamically adjust the rotation speed of the juicer according to the real-time state of the raw materials, which affects the juice yield and fruit juice quality. At the same time, there is a lack of effective coordinated control between different equipment, and the overall optimization of the production process cannot be achieved, resulting in low production efficiency. Moreover, various types of data generated on the production line, such as raw material information, equipment operation data, and product quality data, are stored separately and in inconsistent formats. These data have not been effectively integrated and deeply analyzed, and cannot provide strong support for production decisions. For example, it is difficult for enterprises to dig out the potential relationship between raw material characteristics and product quality from a large amount of raw material quality data, and it is difficult to optimize the production process targeted.

[0004] The existing systems mainly rely on manual sampling inspection and post-event detection for product quality monitoring, and cannot monitor product quality changes in real time and comprehensively. When product quality problems occur, it is often difficult to detect them in a timely manner and take effective measures, resulting in the production of a large number of unqualified products, increasing production costs, and damaging the enterprise's reputation. For example, in the sterilization process, if the sterilization temperature and time are not properly controlled, the problem of excessive microorganisms may be discovered only after filling, causing a large number of products to be scrapped.

[0005] The current main solutions to the above problems are as follows. For a patent named "A Fine-grained Energy Management Method, System, Internet of Things Cloud Management Server and Its Storage Medium" with the authorization announcement number CN116048023B, by obtaining the historical energy consumption inspection data set, splitting it into multiple management subsets to determine the energy consumption standard, combining the production plan to predict energy usage, obtaining the predicted and real-time energy consumption curves, comparing the differences to determine the control strategy and implementing it, and at the same time revising the energy consumption standard. Using distributed monitoring devices to collect data and uploading it to the management cloud server for analysis and processing to achieve fine-grained management of energy, establishing an energy management model that is not affected by production volume and has high precision, capable of achieving minute-level precision division, automatically giving control recommendation strategies, and continuously iterating and optimizing the energy demand and supply curves to achieve the purpose of energy conservation and carbon reduction, providing an effective means for energy management in the production process.

[0006] However, the above solutions mainly focus on energy management and involve less in aspects such as the intelligence of equipment control, production data management, and product quality monitoring and early warning on the juice production line.

[0007] Another example is a patent named "Product Optimization Production Method and System Combining pH Value Control" with the authorization announcement number CN117311298B. By obtaining the technological process nodes of the target product to be produced, the added components and operating environment information of each node, establishing a component and environment correlation matrix, evaluating the PH fluctuation index, establishing a PH stabilization recognition unit through historical production record data, and judging whether to optimize the process of abnormal nodes according to the fluctuation stabilization probability. For the problem of PH value control in the product production process, through correlation analysis and probability evaluation, the production process is optimized, which can effectively improve product quality and has certain reference significance in product quality control; however, the above solutions only focus on the product quality optimization related to the PH value and cannot comprehensively solve problems such as equipment control, data management, and monitoring and early warning of other quality indicators on the juice production line, and the application scope is relatively narrow.

[0008] A patent named "A Deep Processing Method and Processing Equipment of Cherries Based on the Industrial Internet" with the publication number CN119292195A deploys sensors and intelligent devices in the cherry planting, picking and processing production line to collect data, deeply mines the data using big data analysis and machine learning technologies, adjusts the product formula and production process based on market demand, and realizes comprehensive digital management of the processing process through the industrial Internet platform, constructs an industrial Internet platform to achieve intelligent, automated and digital management, and has functions such as digital traceability and personalized customization, improving production efficiency and product quality. However, this patent is for the deep processing of cherries, and its processing technology and equipment are different from those of juice production, and there is insufficient adaptability in terms of equipment control details and data management and quality monitoring and early warning for juice production.

[0009] By comprehensively comparing the above solutions, the new generation of automated control and information management system for fruit juice production lines proposed by the present invention integrates a control module, a data acquisition module, a data analysis and processing module, and a human-machine interaction module according to the characteristics of fruit juice production. The control module realizes intelligent control of equipment; the data acquisition module comprehensively acquires data from multiple links; the data analysis and processing module deeply analyzes data, sets thresholds and issues warnings; the human-machine interaction module facilitates personnel operation and monitoring. This system can effectively solve the problems of non-intelligent equipment control, ineffective management of production data, and difficulty in timely monitoring and warning of product quality in fruit juice production lines, and better meet the actual needs of fruit juice production. Summary of the Invention

[0010] The present invention provides a new generation of automated control and information management system for fruit juice production lines, which solves the problems of non-intelligent equipment control, ineffective management of production data, and difficulty in timely monitoring and warning of product quality in fruit juice production lines.

[0011] To solve the above problems, the technical solution adopted by the invention is:

[0012] The new generation of automated control and information management system for fruit juice production lines includes:

[0013] A control module for starting, stopping and adjusting operation parameters of equipment in each link;

[0014] A data acquisition module is set at each node of the fruit juice production line, including the raw material processing area, the juicing area, the blending area, the sterilization area and the filling area, and is used to collect raw material information, equipment operation status information, and product quality parameter information. The raw material information includes raw material types, weights, and freshness. The equipment operation status information includes equipment rotation speed, temperature, pressure, and energy consumption. The product quality parameter information includes juice concentration, acidity, and microbial content. The data acquisition module is electrically connected to the data analysis module and transmits the collected data to the data analysis module in real time;

[0015] A data analysis and processing module for receiving and analyzing data from the data acquisition module. After data analysis, when the monitored parameters deviate from the preset normal range, an adjustment instruction is sent to the central control unit;

[0016] The specific steps of the data analysis are as follows:

[0017] S01 The data analysis and processing module obtains the original data X from the data acquisition module, and cleans the original data X to remove abnormal values generated due to sensor failures and electromagnetic interference during the acquisition process. The 3σ criterion based on statistics is combined with the adjacent data smoothing comparison method for data processing. If the data at a certain moment fluctuates more than the set smoothing coefficient compared with the data at the previous and subsequent moments, it is marked as abnormal and removed;

[0018] S02 uses the Z-score standardization method to convert the weight, freshness, rotation speed, temperature, pressure, energy consumption, juice concentration, juice acidity, and microbial content data into standard normal distribution data with a mean of 0 and a standard deviation of 1. The formula is: where X is the original data processed in S01, and Z is the data after standardization, is the mean, is the standard deviation; for the raw material types, one-hot encoding is used to convert the raw material types into binary vector forms;

[0019] Based on the sliding window technique, S03 sets a short-term time window and calculates the dynamic threshold for the standardized data within the window. For the equipment rotation speed, calculate the mean and the standard deviation of the rotation speed data within the window, set the upper threshold as and the lower threshold as , where is the sensitivity coefficient preset according to the stability of the production line equipment and process requirements; and short-term dynamic thresholds are established for weight, freshness, temperature, pressure, energy consumption, juice concentration, juice acidity, and microbial content;

[0020] S04 sets a time window for a long period of time, calculates the median M of the data within the window and the standard deviation of the interquartile range IQR. For the juice acidity, set the upper threshold as and the lower threshold as , The value of

[0021] is determined by the strictness of the product quality standard and the historical quality fluctuation situation, and long-term dynamic thresholds are established for weight, freshness, rotation speed, temperature, pressure, energy consumption, juice concentration, and microbial content;

[0022] The human-machine interaction module is bidirectionally communicatively connected to the control module and is used for production management personnel to input control instructions, view real-time status information of the production line, historical production data reports, and receive warning information pushed by the system.

[0023] The principle and advantages of this solution are as follows: Through the collaborative work of the control module, data acquisition module, and data analysis and processing module, the automated control and information management of the juice production line are realized. The control module is responsible for starting, stopping the equipment in each link and adjusting the operation parameters to ensure the normal progress of the production process. The data acquisition module is distributed at various key nodes of the production line, collecting information in multiple aspects such as raw materials, equipment operation status, and product quality parameters, and transmitting these data to the data analysis and processing module in real time. This module cleans the collected raw data, removes outliers, and then through standardization and coding processing, makes the data more convenient for analysis. Then, based on the sliding window technology, short-term and long-term time windows are respectively set, the dynamic thresholds of various parameters are calculated, the collected parameters are compared with the thresholds in real time, different levels of warnings are triggered according to the comparison results, and adjustment instructions are sent to the control module, so as to realize the real-time monitoring and adjustment of the production process.

[0024] Compared with the prior art, in the early warning mechanism, it usually only relies on a single short-term or long-term dynamic threshold, which has obvious defects. If only the short-term dynamic threshold is adopted, although it is sensitive to the immediate changes in data and can quickly capture the instantaneous fluctuations, because it is too sensitive, it is extremely vulnerable to accidental factors and generates false alarms frequently. For example, in actual production, when the equipment is running, occasionally due to a short-term power fluctuation, the rotation speed of the equipment may instantaneously exceed the short-term threshold, thus triggering an early warning. However, the production line as a whole is actually in a stable operation state, which will undoubtedly interfere with the normal production order without reason and consume manpower and time to verify these false alarms. On the contrary, if only the long-term dynamic threshold is set, although it can effectively grasp the long-term trend of data, it is slow to respond when facing sudden abnormal changes in the short term and is difficult to detect quickly. For example, in the raw material supply link, if the freshness of a batch of fruits suddenly drops, since the long-term threshold reflects the average situation over a period of time, it may not be able to capture this short-term abnormality in time, resulting in the discovery of product quality problems only in the subsequent production process. By then, certain production losses have already occurred and the best adjustment opportunity has been missed. This solution innovatively combines the short-term and long-term dynamic thresholds to construct a more perfect and accurate early warning system. When the collected parameter is higher than the short-term upper threshold or lower than the short-term lower threshold, the system can quickly detect the immediate abnormality and immediately trigger a first-level early warning, timely reminding the staff to pay attention to the current production situation and quickly investigate the possible sudden problems; if the parameter deviates from the long-term threshold at the same time, the system will upgrade the early warning to the second level, which means that the problem not only has immediacy but may also involve systematic hidden dangers accumulated in the production line for a long time, and in-depth investigation and comprehensive rectification are required. Taking the monitoring of the acidity of fruit juice as an example, through the double comparison of the short-term and long-term dynamic thresholds, the system can accurately judge whether it is a short-term process fluctuation caused by an instantaneous mixing error or a quality hidden danger formed by the accumulation of deviations in the production process for a long time, effectively avoiding the problems of frequent production stoppages or product quality out-of-control caused by misjudgment, and greatly improving the accuracy and reliability of the early warning.

[0025] In the prior art, there are often problems of incomplete coverage and insufficient analysis in monitoring the operating status of equipment and product quality parameters. This solution realizes the comprehensive and in-depth collection of multi-dimensional information on raw materials, equipment, and products through the data acquisition module. Through the data analysis and processing module, the potential fault risks of the equipment can be accurately identified. Taking the monitoring of equipment energy consumption as an example, the equipment energy consumption is continuously monitored through long-term dynamic thresholds. If it is found that the equipment energy consumption is higher than the threshold for a long time, it is very likely that there are problems such as component wear and circuit aging inside the equipment. The system issues an early warning in advance to remind the staff to arrange maintenance in time, thus effectively ensuring the stable operation of the equipment, greatly reducing the risk of production interruption caused by sudden equipment failures, and reducing the time and economic costs brought by equipment maintenance. At the same time, in terms of product quality control, taking the key parameter of microbial content as an example, this solution conducts dual monitoring based on short-term and long-term dynamic thresholds. The short-term threshold can promptly detect the instantaneous fluctuations in microbial content caused by accidental factors such as environmental changes and operation errors during the production process, while the long-term threshold can continuously evaluate the product quality to ensure that the product always meets strict quality standards throughout the production cycle. Through this dual guarantee mechanism, the economic losses and damage to the brand reputation caused by product quality problems are effectively avoided, the product quality is fundamentally improved, helping the enterprise to gain an advantage in the market competition and achieve steady growth in economic benefits.

[0026] Furthermore, the control module is also equipped with multiple preset combinations of equipment operating parameters according to different juice product formulas. Each juice product has its specific production process requirements, and different formulas require different equipment operating parameters to achieve the best production effect. The preset parameter combination modes are carefully debugged and verified, which can ensure that the products produced each time maintain a high degree of stability and consistency in quality, improving the trust and satisfaction of consumers with the products.

[0027] Furthermore, the sensors in the data acquisition module adopt a redundant design. At the acquisition node, at least two sensors of the same type are equipped. Multiple sensors collect the same parameter simultaneously, and the collected data can be cross-checked and compared with each other. If there are significant deviations in the data of different sensors, the problems can be promptly discovered and investigated. It may be that a certain sensor fails, or the measurement environment is abnormal. In this way, more accurate and reliable data can be screened out, reducing the impact of measurement errors on production decisions.

[0028] Furthermore, the data analysis and processing module also adds an artificial review interface. When the data is complex or there are anomalies, the artificial can review it based on professional knowledge and experience, promptly discover and correct the errors or omissions in the system analysis, ensuring the accuracy and reliability of the data.

[0029] Furthermore, the data analysis and processing module is provided with a data storage unit, which adopts a distributed storage architecture and classifies and stores data according to data type, collection time, and production line area. When specific data needs to be searched, it can be quickly located through the corresponding classification, without blindly searching in the data, saving time and resources. At the same time, the data is scattered and stored on multiple nodes, reducing the risk of data loss caused by a single point of failure. Even if a certain node has problems, the data on other nodes is still available, ensuring the integrity and availability of the data and ensuring that the monitoring and management of the production line will not be interrupted due to data storage problems.

[0030] Furthermore, the human-machine interaction module also displays the dynamic changes of the operating parameters of each link's equipment, raw material information, and product quality parameters in real time in the form of visual charts. Compared with complex data tables or text descriptions, intuitive charts can present a large amount of data in a simple and clear manner. Production management personnel can quickly grasp the key information of each link without having to laboriously interpret cumbersome data, thus greatly improving the efficiency and accuracy of decision-making.

[0031] Furthermore, the short-term time is 1 hour and the long-term time is 24 hours. By setting a 1-hour short-term time window and a 24-hour long-term time window, in terms of controlling the production rhythm, the 1-hour short-term window can track production in real time, promptly discover problems such as sudden equipment failures and short-term fluctuations in raw materials, quickly give early warnings and make adjustments to ensure smooth production. The 24-hour long-term window can analyze the production stability and trends as a whole, such as evaluating the operating efficiency of equipment within a day and the overall consumption of raw materials, providing a basis for production planning. The short-term window has a small amount of data, which is convenient for quick processing and can promptly detect abnormalities. The long-term window has a large amount of data, which can more comprehensively reflect the production rules, reduce the interference of abnormal data, and improve the accuracy of analysis. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Figure 1 is a flowchart of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0033] The embodiment is basically as shown in the Figure 1 accompanying drawings. The automated control and information management system for a new generation of fruit juice production line is characterized by including:

[0034] A control module for starting, stopping, and adjusting the operating parameters of the equipment in each link;

[0035] The data acquisition module is set at each node of the fruit juice production line, including the raw material processing area, the juice extraction area, the blending area, the sterilization area, and the filling area, and is used to collect raw material information, equipment operation status information, and product quality parameter information. The raw material information includes the type of raw material, weight, and freshness. The equipment operation status information covers equipment rotation speed, temperature, pressure, and energy consumption. The product quality parameter information includes juice concentration, juice acidity, and microbial content. The data acquisition module is electrically connected to the control module and transmits the collected data to the control module in real time;

[0036] The data analysis and processing module is used to receive and analyze the data from the data acquisition module, and through data analysis, it monitors and warns the production process in real time. When it detects that the data deviates from the preset normal range, it sends an adjustment instruction to the central control unit in time;

[0037] The specific steps of the data analysis are as follows:

[0038] S01 The data analysis and processing module obtains the original data X from the data acquisition module. First, it cleans the original data X to remove the outliers generated during the acquisition process due to sensor failures, electromagnetic interference, etc. It uses the 3σ criterion based on statistics combined with the adjacent data smoothing comparison method. If the data at a certain moment fluctuates more than the set smoothing coefficient compared with the data at the previous and next moments, it is marked as abnormal and removed;

[0039] S02 For weight, freshness, rotation speed, temperature, pressure, energy consumption, juice concentration, juice acidity, and microbial content, the Z-score standardization method is used to convert the data into a standard normal distribution data with a mean of 0 and a standard deviation of 1. The formula is: where X is the original data processed in S01, Z is the data after standardization, is the mean, is the standard deviation; for the type of raw material, one-hot encoding is used to convert the type of raw material into a binary vector form.

[0040] S03 Based on the sliding window technology, a 1-hour time window is set, and the dynamic threshold is calculated for the standardized data within the window. For the equipment rotation speed, the mean of the rotation speed data within the window and the standard deviation are calculated. The upper threshold is set as and the lower threshold is set as where is the sensitivity coefficient preset according to the equipment stability and process requirements of the production line; similarly, corresponding short-term dynamic thresholds are established for weight, freshness, temperature, pressure, energy consumption, juice concentration, juice acidity, and microbial content;

[0041] S04 sets a 24-hour time window and calculates the standard deviation of the median M and interquartile range IQR of the data within the window. , for the acidity of the juice, set the upper threshold to , and the lower threshold to , The values are determined according to the strictness of the product quality standard and the historical quality fluctuation. Similarly, corresponding long-term dynamic thresholds are established for weight, freshness, rotational speed, temperature, pressure, energy consumption, juice concentration, and microbial content.

[0042] S05 compares the collected data with the corresponding short-term and long-term dynamic thresholds in real time. If the currently measured value is higher than the short-term upper threshold or lower than the short-term lower threshold, a first-level warning is immediately triggered. If it is simultaneously higher than the long-term upper threshold or lower than the long-term lower threshold, it is upgraded to a second-level warning. The warning information is sent to the control module to send an adjustment instruction, and the warning information is prominently displayed in the human-machine interaction module.

[0043] The human-machine interaction module is bidirectionally communicatively connected to the control module and is used for production management personnel to input control instructions, view real-time status information of the production line, historical production data reports, and receive warning information pushed by the system.

[0044] Through the control module, the control of equipment in each link of the juice production line is realized, and the first-level warning and the second-level warning are realized. The data acquisition module is distributed at each node of the production line, collects data information, and transmits these real-time data to the data analysis module. After receiving the data, the data analysis and processing module first cleans the original data, then standardizes and encodes the data to make it have a unified measurement and a form convenient for analysis. Based on the sliding window technology, time windows for short-term and long-term are respectively set to calculate short-term and long-term dynamic thresholds. By comparing the collected data with these thresholds in real time, abnormal situations in the production process can be detected in a timely manner. When the data deviates from the short-term threshold, a first-level warning is triggered, and when it deviates from the long-term threshold, it is upgraded to a second-level warning, and the warning information is sent to the control module. When the control module receives a first-level warning, since the parameter only deviates from the short-term threshold, it will make a small adjustment to the relevant equipment parameters, and at the same time increase the data acquisition frequency of the corresponding node to more closely monitor the data change, and will also record the warning information and notify the production management personnel to pay attention. When receiving a second-level warning, since the parameter deviates from the long-term threshold, the control module will immediately stop the production process, urgently adjust the equipment, start the product traceability and isolation mechanism to prevent problem products from flowing into the market, and continuously display the warning information in a prominent manner in the human-machine interaction module.

[0045] The system receives a production instruction. The control module will start the equipment of each link in sequence according to the preset production process. For example, it will first start the equipment in the raw material processing area. According to data such as the type and weight of the raw materials, appropriate operating parameters are calculated through precise algorithms to regulate the equipment speed and ensure that the raw materials can be processed evenly and efficiently. In the juicing area, based on information such as the freshness and weight of the raw materials, the equipment pressure and speed are adjusted to ensure the juice yield and juice quality. In the blending area, combined with data such as juice concentration and acidity, the start / stop and flow parameters of the ingredient addition equipment are controlled for precise blending. When the operating parameters of a certain link of equipment are abnormal, such as the temperature and pressure exceeding the set range, the control module will immediately take corresponding measures according to the warning level. If it is a first-level warning, the operating parameters of the equipment will be slightly adjusted. If it is a second-level warning, the operation of the equipment in this link will be stopped immediately. After the fault is detected and repaired, it will be restarted and the parameters will be adjusted again according to the established process to ensure the safe and efficient operation of the entire production process, making the production process more compact and smooth, thus greatly shortening the production cycle. The data collection module collects information such as the type, weight, and freshness of raw materials, the speed, temperature, pressure, and energy consumption of equipment, the concentration, acidity, and microbial content of juice. The determination of the raw material type provides a basic basis for the subsequent selection of production processes and formula adjustment. The accurate collection of weight helps to precisely control the input amount of raw materials, thus ensuring the consistency and quality stability of products. The monitoring of freshness can ensure the use of high-quality raw materials for production, avoid affecting the quality and taste of juice due to raw material deterioration, and also helps to optimize the procurement plan and inventory management, reduce costs and waste.

[0046] The collection of speed, temperature, pressure, and energy consumption in the equipment operating status information brings many benefits. The monitoring of speed can reflect the working efficiency and operating stability of the equipment. Timely detection of abnormal speed helps to prevent equipment failures and production interruptions. The collection of temperature and pressure is crucial for ensuring the process conditions during production, which directly affect the processing effect and quality of juice. The monitoring of energy consumption not only helps to evaluate the operating efficiency of the equipment but also provides data support for energy-saving optimization, thus reducing production costs and achieving sustainable production.

[0047] Taking the change of temperature parameters as an example, when the control module receives a first-level warning triggered by temperature, it means that the temperature has deviated from the short-term dynamic threshold. At this time, the control module will make a small adjustment to the power of the refrigeration or heating equipment according to the preset fine-tuning strategy. For example, in the juice sterilization area, if the sterilization temperature is slightly higher than the short-term upper limit threshold, the control module will increase the power of the refrigeration equipment by 5% to make the sterilization temperature drop slowly. At the same time, the control module will closely monitor the temperature change, collect temperature data every 30 seconds, and observe the adjustment effect.

[0048] When the temperature triggers a secondary warning, that is, when it exceeds the long-term dynamic threshold, the control module will take more radical measures. It will immediately stop the ongoing sterilization operation to avoid affecting the quality of the juice due to too high or too low temperature. Subsequently, the control module will recalculate the appropriate cooling or heating power based on historical temperature data and equipment performance parameters. For example, directly increase the cooling power by 20%, and after the equipment restarts, collect temperature data at high frequency at 15-second intervals to ensure that the temperature can quickly stabilize within a reasonable range. During the entire adjustment process, the control module will also interact with other relevant equipment, such as adjusting the material conveying speed, to prevent the material from staying in the sterilization area for too long or too short a time due to temperature adjustment.

[0049] The accurate control of the juice concentration determines the taste and flavor of the product, meeting the expectations of consumers and market demands. The monitoring of acidity helps to ensure the taste balance and quality stability of the product, avoiding adverse effects on the product quality caused by abnormal acidity. The detection of the microbial content is an important link in ensuring food safety, which can timely detect potential contamination and spoilage risks, ensuring the health and safety of consumers. Through these parameter characteristics, the key aspects of raw materials, equipment operation, and product quality are comprehensively covered, providing necessary data support for the optimization of the production process, quality control, cost management, equipment maintenance, and food safety guarantee, contributing to the realization of efficient, stable, high-quality, and safe juice production; through data cleaning and standardization methods, interference and error data are effectively removed, making the analysis results more reliable. Based on the dynamic threshold setting and hierarchical warning mechanism of the sliding window technology, the sliding window technology combines the settings of short-term and long-term dynamic thresholds to achieve the monitoring of the production process. The short-term dynamic threshold can timely capture short-term abnormal fluctuations and quickly trigger a warning; the long-term dynamic threshold evaluates the stability and quality trend of production from a more macroscopic perspective, further improving the accuracy and comprehensiveness of the warning. Compared with directly setting a fixed threshold, the dynamic threshold setting and hierarchical warning mechanism based on the sliding window technology have unique advantages. Fixed thresholds often cannot adapt to the complex and changeable actual situations in the production process, and in some scenarios where the working conditions change gradually, false alarms or missed alarms are likely to occur. The dynamic threshold setting can analyze and evaluate the data dynamically using the sliding window according to the real-time data changes in the production process and adaptively adjust the threshold. Once an abnormality occurs in the production, this mechanism can issue a precise alarm in the first time, enabling the management personnel to take measures quickly and effectively avoid the expansion of the problem, thus strongly guaranteeing the stability of the product quality, enabling the management personnel to take measures quickly and avoid the expansion of the problem, thus strongly guaranteeing the stability of the product quality.

[0050] The control module is also equipped with multiple preset operation parameter combination modes for different juice product formulas. Each juice product has its specific production process requirements, and different formulas require different equipment operation parameters to achieve the best production effect. The preset parameter combination modes are carefully debugged and verified, which can ensure that the products produced each time maintain a high degree of stability and consistency in quality, improving consumers' trust and satisfaction in the products.

[0051] The sensors in the data acquisition module adopt a redundant design. At the acquisition nodes, at least two sensors of the same type are equipped. Multiple sensors collect the same parameter simultaneously, and the collected data can be cross-checked and compared with each other. If there are significant deviations in the data from different sensors, problems can be discovered and investigated in a timely manner. It may be that a certain sensor fails, or there may be abnormalities in the measurement environment. In this way, more accurate and reliable data can be selected, reducing the impact of measurement errors on production decisions.

[0052] The data analysis and processing module is also equipped with a manual review interface. When the data is complex or abnormal, manual review can be carried out based on professional knowledge and experience to promptly discover and correct errors or omissions in the system analysis, ensuring the accuracy and reliability of the data.

[0053] The data analysis and processing module is provided with a data storage unit. This data storage unit adopts a distributed storage architecture, classifying and storing data in multiple dimensions such as data type, acquisition time, and production line area. When specific data needs to be searched, it can be quickly located through the corresponding classification, without blindly searching through the data, saving time and resources. At the same time, the data is scattered and stored on multiple nodes, reducing the risk of data loss caused by a single point of failure. Even if a certain node has problems, the data on other nodes is still available, ensuring the integrity and availability of the data and ensuring that the monitoring and management of the production line will not be interrupted due to data storage problems.

[0054] The human-machine interaction module also displays the dynamic changes of the operation parameters of each link's equipment, raw material information, and product quality parameters in real time in the form of visual charts. Compared with complex data tables or text descriptions, intuitive charts can present a large amount of data in a concise and clear manner. Production management personnel can quickly grasp the key information of each link without having to laboriously interpret cumbersome data, thus greatly improving the efficiency and accuracy of decision-making.

[0055] The short-term time is 1 hour, and the long-term time is 24 hours. A 1-hour short-term time window and a 24-hour long-term time window are set. In terms of production rhythm control, the 1-hour short-term window can track production in real time, promptly detect problems such as sudden equipment failures and short-term raw material fluctuations, quickly give early warnings and make adjustments to ensure smooth production. The 24-hour long-term window can analyze production stability and trends as a whole, such as evaluating the operating efficiency of equipment within a day and the overall consumption of raw materials, etc., providing a basis for production planning. The data volume of the short-term window is small, which is convenient for rapid processing and can promptly detect abnormalities. The data volume of the long-term window is large, which can more comprehensively reflect production rules, reduce interference from abnormal data, and improve the accuracy of analysis. Specific Embodiment

[0057] Data acquisition modules are installed at each key node in the raw material processing area, juicing area, blending area, sterilization area, and filling area of the juice production line. Each node is equipped with sensors as needed for collecting the type, weight, freshness, equipment rotation speed, temperature, pressure, energy consumption, juice concentration, juice acidity, and microbial content of the raw materials. These sensors are electrically connected to the control module to ensure real-time data transmission.

[0058] The control module is installed in the central control room of the production line, connected to each device to achieve the start, stop, and adjustment of operating parameters of the device. At the same time, the data analysis and processing module and the human-machine interaction module are also installed and configured in the control room.

[0059] During the production process, in the raw material processing area, when a batch of fresh fruits arrives, the sensors collect information such as the type, weight, and freshness of the fruits, and transmit this data to the control module and the data analysis and processing module in real time.

[0060] In the juicing area, sensors such as equipment rotation speed, temperature, and pressure continuously monitor the operating status of the equipment. In the blending area, quality parameters such as juice concentration and juice acidity are accurately collected. Similarly, relevant data in the sterilization area and filling area are also promptly obtained and transmitted.

[0061] After receiving the data transmitted by the data acquisition module, the data analysis and processing module processes it according to the following steps:

[0062] First, data cleaning is performed. Using the 3σ criterion based on statistics combined with the adjacent data smoothing and comparison method, abnormal values caused by sensor failures, electromagnetic interference, etc. are removed. For example, if the temperature data at a certain moment fluctuates significantly from the data at the previous and subsequent moments beyond the set smoothing coefficient, this data is marked as abnormal and removed.

[0063] Next, perform Z-score standardization on data such as weight, freshness, rotation speed, temperature, pressure, energy consumption, juice concentration, juice acidity, and microbial content. At the same time, convert the raw material types into binary vector form through one-hot encoding, so that the data has a unified measurement and a form convenient for analysis.

[0064] Then, based on the sliding window technique, set a short-term time window of 1 hour and a long-term time window of 24 hours. For acidity, calculate the mean and standard deviation of the rotation speed data within the 1-hour window, and set the upper threshold and lower threshold. Similarly, establish corresponding short-term dynamic thresholds for other parameters. For juice acidity, by calculating the median M and interquartile range IQR of the data within the 24-hour window, set the corresponding long-term threshold, for determining the long-term dynamic threshold of juice acidity , collect the juice acidity data of the production line in the past two years, a total of 15,000 samples. Through preliminary analysis, obtain the median (pH value) and the standard deviation of the interquartile range IQR = 0.3 (pH value). Starting from the strictness of the product quality standard, the target acidity range of the juice is set to 3.4 - 3.8 (pH value). Through text mining and analysis of the online evaluation and offline research data of consumers, it is found that when the juice acidity deviates from the target range by 0.12 (pH value), the negative evaluation of the taste by consumers increases significantly. After cost accounting and production efficiency evaluation, determine the balance point between the acceptable cost increase due to adjustment and quality improvement, and the corresponding acidity fluctuation range is 0.13 (pH value). Combining the above market feedback and cost-benefit analysis, take the average value of the allowable fluctuation ranges of the two as a reference

[0065] = 0.125, interquartile range IQR = 0.3, then , so, for the long-term dynamic threshold of juice acidity, the upper threshold is , and the lower threshold is . When the real-time monitored juice acidity exceeds this range, the system can give an early warning and make adjustments in a timely manner.

[0066] Compare the collected data with the short-term and long-term dynamic thresholds in real time. Once it is found that the currently measured value is higher than the short-term upper threshold or lower than the short-term lower threshold, immediately trigger a first-level warning; if it is simultaneously higher than the long-term upper threshold or lower than the long-term lower threshold, upgrade it to a second-level warning. The warning information is quickly sent to the control module and prominently displayed in the human-machine interaction module.

[0067] The control module precisely controls the equipment at each link of the production line according to the preset parameters and the analysis results of the received data.

[0068] When producing apple juice, the control module automatically switches to the preset operating parameter combination mode of the apple juice production equipment. During the production process, if the data analysis and processing module issues a warning of excessive equipment speed, the control module immediately reduces the speed to ensure the stability of production and product quality.

[0069] In the pressure monitoring of the filling area, two pressure sensors of the same type are equipped to work simultaneously. During a certain production, the data collected by the two sensors showed a large deviation. The system quickly alarmed. After investigation by the technical staff, it was found that one of the sensors had a measurement error due to loose installation. It was repaired in time to ensure the accuracy of the data.

[0070] When processing the production data of a batch of new raw materials, the system analysis results were slightly abnormal. Through the additional manual review interface, experienced engineers reviewed the data and found that the characteristics of the new raw materials caused the original analysis algorithm of the system to be inaccurate. The engineers promptly corrected the algorithm to ensure the normal progress of subsequent production.

[0071] At the same time, the data storage unit adopts a distributed storage architecture, classifying and storing data in multiple dimensions such as raw material data, equipment operation data, product quality data, collection time, and production line area. When querying the equipment energy consumption data in the filling area last month, the staff quickly located and obtained accurate data through classification, providing a strong basis for optimizing energy consumption management.

[0072] The human-machine interaction module also visually displays the dynamic changes of the equipment operation parameters, raw material information, and product quality parameters in each link in the form of visual charts. Production managers can intuitively see the real-time data of each area in front of the monitoring large screen.

[0073] The juice concentration in the blending area shows a downward trend in the visual chart of the human-machine interaction module. The manager immediately inputs an adjustment instruction through the human-machine interaction module to increase the addition amount of concentrated juice, correcting the concentration deviation in time to ensure the consistency of product quality.

[0074] The above are only embodiments of the present invention. Specific structures and common knowledge such as characteristics that are well-known in the art are not described in detail herein. Those of ordinary skill in the art know all the common general technical knowledge in the technical field to which the invention pertains before the filing date or the priority date, are able to obtain all the prior art in this field, and have the ability to apply conventional experimental means before this date. Those of ordinary skill in the art can, under the inspiration given in this application, combine their own abilities to complete and implement this solution. Some typical well-known structures or well-known methods should not become an obstacle for those of ordinary skill in the art to implement this application. It should be noted that for those skilled in the art, without departing from the structure of the present invention, several deformations and improvements can also be made, and these should also be regarded as the protection scope of the present invention, and these will not affect the implementation effect of the present invention and the practicality of the patent. The protection scope required by this application should be based on the content of its claims, and the specific implementation manners and the like recorded in the specification can be used to interpret the content of the claims.

Claims

1. A new generation of juice production line automation control and information management system, characterized by: include: Control module, used for starting, stopping and adjusting operating parameters of equipment in each link; The data acquisition module is arranged at each node of the juice production line, including the raw material processing area, the juicing area, the mixing area, the sterilization area and the filling area, and is used to collect raw material information, equipment operation status information and product quality parameter information. The raw material information includes raw material type, weight and freshness; the equipment operation status information includes equipment speed, temperature, pressure and energy consumption; the product quality parameter information includes juice concentration, juice acidity and microbial content; the data acquisition module is electrically connected to the data analysis module, and transmits the collected data to the data analysis module in real time; The data analysis and processing module is used to receive and analyze the data from the data acquisition module, and after data analysis, when the monitoring parameters deviate from the preset normal range, send adjustment instructions to the central control unit; The specific steps of the data analysis are: The S01 data analysis and processing module obtains the original data X from the data acquisition module, cleans the original data X, removes abnormal values ​​caused by sensor failure and electromagnetic interference during the acquisition process, and uses the 3σ criterion based on statistics combined with the adjacent data smoothing comparison method to process the data. If the fluctuation of the data at a certain moment and the data before and after exceeds the set smoothing coefficient, it is marked as abnormal and removed; S02 uses the Z-score standardization method to convert the weight, freshness, rotation speed, temperature, pressure, energy consumption, juice concentration, juice acidity, and microbial content into standard normal distribution data with a mean of 0 and a standard deviation of 1. The formula is: Where X is the original data processed in S01, and Z is the standardized data. is the mean, is the standard deviation; the raw material types are converted into binary vector form using one-hot encoding; S03 is based on sliding window technology. It sets a short-term time window and calculates the dynamic threshold for the standardized data in the window. For the equipment speed, it calculates the mean of the speed data in the window. and standard deviation , set the upper threshold to , the lower threshold is ,in The sensitivity coefficient is pre-set according to the stability of production line equipment and process requirements; and short-term dynamic thresholds are established for weight, freshness, temperature, pressure, energy consumption, juice concentration, juice acidity, and microbial content; S04 sets a long-term time window and calculates the median M and standard deviation of the interquartile range IQR of the data in the window , for juice acidity, the upper threshold is set to , the lower threshold is , The values ​​are determined by the strictness of product quality standards and historical quality fluctuations, and long-term dynamic thresholds are established for weight, freshness, rotation speed, temperature, pressure, energy consumption, juice concentration, and microbial content; S05 compares the collected parameters with the corresponding short-term and long-term dynamic thresholds in real time. If the current measured value is higher than the short-term upper threshold or lower than the short-term lower threshold, a first-level warning is triggered; if the current measured value is higher than the long-term upper threshold and the short-term upper threshold or lower than the long-term lower threshold and the short-term lower threshold, it is upgraded to a second-level warning, and the warning information is sent to the control module and an adjustment instruction is generated, and the warning information is highlighted in the human-computer interaction module; The human-computer interaction module is connected to the control module in a two-way communication manner and is used by production managers to input control instructions, view real-time status information of the production line, historical production data reports, and receive warning information pushed by the system.

2. The new generation juice production line automation control and information management system according to claim 1 is characterized in that: The control module is also capable of presetting a plurality of equipment operation parameter combination modes according to different juice product formulas.

3. The new generation juice production line automation control and information management system according to claim 1 is characterized in that: The sensors in the data acquisition module adopt a redundant design, and at least two sensors of the same type are equipped at the acquisition node.

4. The new generation juice production line automation control and information management system according to claim 1 is characterized in that: The data analysis and processing module is also provided with a manual review interface.

5. The new generation juice production line automation control and information management system according to claim 1 is characterized in that: The data analysis and processing module is provided with a data storage unit, which adopts a distributed storage architecture and classifies and stores data according to data type, collection time, and production line area.

6. The new generation juice production line automation control and information management system according to claim 1 is characterized in that: The human-computer interaction module also displays the dynamic changes of equipment operating parameters, raw material information, and product quality parameters in each link in real time in the form of visual charts.

7. The new generation juice production line automation control and information management system according to claim 1 is characterized in that: The short-term time is 1 hour, and the long-term time is 24 hours.

Images