Intelligent integrated management system of dairy beverage production line

Abstract

The invention discloses an intelligent integrated management system of a dairy beverage production line, and relates to the technical field of dairy beverages, and the system comprises the following steps: obtaining the operation environment and part data of a to-be-detected dairy beverage production line, and building a dairy equipment geometric model based on 3D modeling software; obtaining a plurality of different injection pressures to be screened; solvent leakage detection is carried out respectively, and leakage path data and solvent and equipment loss data during solvent leakage detection are obtained; constructing an injection pressure evaluation model; the method is applied to injection pressure management and control of intelligent solvent leakage detection of the dairy beverage production line, and solves the problems that in the prior art, injection pressure control is often not accurate and the detection efficiency is high in the intelligent solvent leakage detection of the dairy beverage production line, and the detection efficiency is high. And the intelligent control sensitivity of the production line of different equipment is insufficient or the equipment is damaged.

Description

Technical Field

[0001] The present invention relates to the technical field of dairy beverages, and more specifically, to an intelligent integrated management system for a dairy beverage production line. Background Art

[0002] The dairy beverage sector is a highly competitive and technology-intensive industry, covering a variety of products such as milk, yogurt, cheese, lactic acid beverages, etc. As consumers' health awareness increases, the types and functions of dairy beverages are also constantly diversified, and new products such as probiotic-rich, low-sugar, and low-fat products have emerged to meet the needs of different consumers. In order to ensure product quality and safety, the production of dairy beverages requires highly automated machinery and equipment, and all links from raw material processing to packaging rely on precise mechanical systems.

[0003] The intelligent integrated management system of the dairy beverage production line integrates advanced automation equipment, sensor technology, data analysis and artificial intelligence to achieve comprehensive monitoring and optimization of the production process. The system can collect production data in real time, automatically adjust production parameters, predict equipment failures and perform preventive maintenance, while improving production efficiency and flexibility through intelligent scheduling. Intelligent integrated management not only ensures product quality and safety, but also reduces costs, improves production transparency, and makes dairy beverage production more efficient, accurate and in line with consumer health needs.

[0004] Solvent leakage detection plays an important role in the intelligent management system of dairy beverage production lines. This method injects specific solvents (such as dyes or gases) into the sealing system of the equipment, and combines intelligent sensors and visual detection technology to monitor the sealing status of the equipment in real time. These sensors can accurately detect solvent leakage and promptly identify potential faults such as poor sealing, corrosion, and cracks in the equipment through data analysis. The intelligent detection system can automatically adjust the concentration, pressure and other parameters of solvent injection to ensure the sensitivity of the detection results without affecting the normal operation of the equipment. Through intelligent solvent leakage detection, the production line can achieve more accurate and efficient fault warning and maintenance, ensure the safe and stable operation of the dairy beverage production line, and improve production efficiency and product quality.

[0005] The above disclosed technical solution has at least the following technical problems: In the intelligent solvent leakage detection of the production line, the control of the injection pressure is crucial because it directly affects the effect of the solvent flowing in the equipment.

[0006] Too high an injection pressure may cause the solvent to flow too quickly, or even break the seal of the equipment, affecting the accuracy of the test results and possibly causing damage to the equipment;

[0007] If the injection pressure is too low, the solvent may not be able to effectively penetrate all potential leak points, reducing the detection sensitivity and missing subtle signs of failure.

[0008] Due to differences in structure and sealing characteristics of different dairy equipment, the optimal injection pressure of each device is also different. The existing technology often has inaccurate injection pressure control in intelligent solvent leakage detection of dairy beverage production lines, resulting in insufficient sensitivity of intelligent control of production lines of different equipment or equipment damage.

[0009] In view of the above problems, the present invention proposes a solution. Summary of the invention

[0010] In order to overcome the above-mentioned defects of the prior art, an embodiment of the present invention provides an intelligent integrated management system for a dairy beverage production line. By analyzing the injection pressure in the intelligent solvent leakage detection of the production line, it solves the problem that the prior art often has inaccurate injection pressure control in the intelligent solvent leakage detection of the dairy beverage production line, resulting in insufficient sensitivity of intelligent control of production lines of different equipment or equipment damage.

[0011] To achieve the above object, the present invention provides the following technical solutions:

[0012] The intelligent integrated management system of a dairy beverage production line includes the following steps: obtaining the operating environment and component data of a dairy beverage production line to be tested, and constructing a geometric model of dairy equipment based on 3D modeling software; assigning physical properties to the geometric model of dairy equipment and the solvent to be injected based on the material library corresponding to the 3D modeling software; obtaining the initial injection pressure of solvent leakage detection based on historical data, and performing random perturbations through a computer to obtain several different injection pressures to be screened; applying different injection pressures to be screened to perform solvent leakage detection on the geometric model of dairy equipment, and obtaining leakage path data and solvent and equipment loss data during solvent leakage detection; constructing an injection pressure evaluation model based on a convolutional neural network according to the leakage path data and the equipment and solvent loss data; constructing a pressure-detection effect display model according to the output of the injection pressure evaluation model and the corresponding injection pressure to be screened, and performing data analysis to obtain the screened injection pressure, which is applied to the injection pressure control of intelligent solvent leakage detection in the dairy beverage production line.

[0013] In a preferred embodiment, the operating environment includes temperature, humidity, pressure conditions, ambient airflow, vibration and noise; the component data includes equipment structure parameters, seal information, material properties, joint and weld characteristics and fluid channel design; the construction of the dairy equipment geometric model based on 3D modeling software is specifically: using 3D modeling software to create basic geometric shapes according to the operating environment and component data; adding detailed structures to the basic geometric shapes, the detailed structures including seams and welds, seals, flanges and bolts; dividing the dairy equipment geometric model into small unit grids, and refining the small unit grid distribution in high stress areas.

[0014] In a preferred embodiment, the initial injection pressure for solvent leakage detection is obtained based on historical data, and random perturbations are performed by a computer to obtain several different injection pressures to be screened, specifically: relevant data for solvent leakage detection are extracted from historical detection records of the dairy beverage production line, including the injection pressure range, detection effect and fault characteristics when the leakage occurs; in combination with the equipment type, operating environment and solvent characteristics, commonly used injection pressure values ​​are counted and calculated as the initial injection pressure, and the average value is used as the initial injection pressure; the initial injection pressure value is randomly perturbed based on the Monte Carlo method to generate a set of candidate pressure values; the generated candidate pressure values ​​are screened according to the equipment and solvent characteristics, and extreme values ​​that may cause equipment damage or detection failure are eliminated to obtain several different injection pressures to be screened.

[0015] In a preferred embodiment, the leakage path data includes a detection area coverage influence coefficient and a leakage path influence coefficient; the specific method for obtaining the detection area coverage influence coefficient is as follows: the three-dimensional geometric model of the dairy beverage production line is unitized, the equipment is divided into multiple small volume units, and the unit sets are labeled and obtained; the number of equipment surface units that affect the solvent diffusion range is counted to obtain the volume of the affected area; the solvent diffusion path is identified and analyzed in the unitized equipment model, and the rate of the diffusion path is calculated; the detection area coverage influence coefficient is calculated by combining the rate of the diffusion path and the volume of the affected area.

[0016] In a preferred embodiment, the specific method for obtaining the leakage path influence coefficient is as follows: based on the solvent leakage detection process under different injection pressures, a set of paths that multiple injection pressures affect the detection results is obtained, wherein the set of paths that multiple injection pressures affect the detection results includes abnormal leakage paths caused by various injection pressures, and a weighted valuation is performed on the abnormal leakage paths based on historical data; a regression model is constructed based on the set of paths that multiple injection pressures affect the detection results and the corresponding weights to generate a leakage path influence coefficient.

[0017] In a preferred embodiment, the solvent and equipment loss data include a solvent and equipment loss cost coefficient; a specific method for obtaining the solvent and equipment loss cost coefficient is as follows: obtaining the ratio of solvent consumption to equipment loss for several different injection pressures; calculating the loss cost standard deviation and the loss cost average of the ratio of solvent consumption to equipment loss for several different injection pressures; calculating the coefficient of variation of the ratio of solvent consumption to equipment loss based on the loss cost standard deviation and the loss cost average; calculating the solvent and equipment loss cost coefficient based on the coefficient of variation of the ratio of solvent consumption to equipment loss based on a preset solvent and equipment loss cost coefficient calculation formula.

[0018] In a preferred embodiment, the pressure-detection effect display model is constructed based on the output of the injection pressure evaluation model and the corresponding injection pressure to be screened, specifically: the injection pressure evaluation coefficient of the output under different injection pressures to be screened is obtained from the injection pressure evaluation model; the X-axis is set to the injection pressure to be screened, and the Y-axis is represented as the injection pressure evaluation coefficient to establish a rectangular coordinate system; the injection pressure evaluation coefficient corresponding to each injection pressure to be screened is marked on the rectangular coordinate system to form marked points, and several marked points are smoothly connected with a smooth curve to construct a pressure-detection effect display model.

[0019] In a preferred embodiment, the data analysis is performed to obtain the screened injection pressure, which is applied to the injection pressure control of intelligent solvent leakage detection in the dairy beverage production line. Specifically, the injection pressures to be screened corresponding to each peak value of the pressure-detection effect display model are screened to obtain a first screening group; the secondary derivatives of each injection pressure to be screened in the first screening group on the pressure-detection effect display model are compared, and the injection pressure to be screened with the smallest comparison result is used as the injection pressure after screening, and applied to the injection pressure control of intelligent solvent leakage detection in the dairy beverage production line.

[0020] Technical effects and advantages of the intelligent integrated management system of the dairy beverage production line of the present invention:

[0021] 1. The present invention acquires the equipment operating environment and component data, establishes a geometric model and assigns corresponding physical properties, which can truly simulate the behavior of the equipment in the event of solvent leakage. Further, by accurately calculating the leakage path and the detection area coverage influence coefficient, the injection pressure is optimized to ensure that the solvent diffusion range is moderate, avoid false detection or missed detection, and reduce the risk of equipment damage. This method not only adapts to equipment differences and improves the detection consistency between different equipment, but also improves the efficiency of intelligent control of the production line by optimizing the analysis of solvent flow patterns, and solves the problems of insufficient sensitivity and equipment damage in the prior art, thereby achieving more accurate and reliable intelligent control of the production line. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 It is a structural schematic diagram of the intelligent integrated management system of the dairy beverage production line of the present invention. DETAILED DESCRIPTION

[0023] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0024] Embodiment 1, Figure 1 The invention provides an intelligent integrated management system for a dairy beverage production line, comprising the following steps:

[0025] S1, obtain the operating environment and component data of the dairy beverage production line to be tested, and build a geometric model of the dairy equipment based on 3D modeling software.

[0026] The operating environment includes temperature, humidity, pressure conditions, ambient air flow, vibration and noise;

[0027] The component data includes equipment structural parameters, seal information, material properties, joint and weld properties, and fluid channel design;

[0028] The geometric model of dairy equipment constructed based on 3D modeling software is specifically as follows:

[0029] Use 3D modeling software to create basic geometric shapes based on the operating environment and component data;

[0030] adding detailed structures to the basic geometry, including seams and welds, seals, and flanges and bolts;

[0031] The geometric model of dairy equipment is divided into small unit grids, and the distribution of small unit grids in high stress areas is refined.

[0032] It should be noted that refining the mesh distribution of leak points or high stress areas means that during the 3D modeling process, denser meshing is used for key areas where leaks may occur or where high stress is applied, so as to improve the calculation accuracy of these areas, while keeping coarser meshes in other secondary areas to reduce the amount of calculation. Through this targeted mesh optimization, it is possible to effectively balance computing resources and efficiency while ensuring the accuracy of simulation results, thus meeting the analytical needs of solvent leak detection.

[0033] It should be noted that for seams and welds: mark the seam positions on the model according to the actual structure, and use line or surface tools to simulate welds. Seals: create geometric models of rubber rings or gaskets at the sealing parts of the equipment. Flanges and bolts: add flanges, bolt holes, etc. at the connection points to enhance the realism of the model.

[0034] It should be noted that commonly used 3D modeling software include AutoCAD, SolidWorks, ANSYS SpaceClaim, Fusion 360, etc. The appropriate modeling software should be selected according to the actual equipment characteristics and complexity, which will not be elaborated here.

[0035] It should be noted that the operating environment data refers to information related to the working status of the equipment and the surrounding injection pressure, which is used to simulate the actual operating conditions of the equipment to ensure the accuracy of 3D modeling and intelligent control of the production line. Mainly including: Temperature: affects the physical properties of the equipment material (such as thermal expansion coefficient) and the fluid properties of the solvent (such as viscosity and diffusion rate). Humidity: High humidity may affect the diffusion path of the solvent and interfere with the signal strength of the test results. Pressure conditions: The equipment may exhibit different sealing properties under different operating pressures, which directly affects the degree of leakage and the difficulty of detection. Ambient airflow: External airflow will change the diffusion path of the solvent and affect the detectability of the leakage signal. Vibration and noise: A high vibration environment may cause additional stress on the sealing components and interfere with the sensor signal processing of the detection equipment.

[0036] Component data involves the structural information and material properties of the dairy beverage production line, and is the basis for building a 3D geometric model and assigning physical properties. Mainly includes: Equipment structural parameters: including the equipment's overall dimensions, internal pipeline layout, joint location, etc., which determine the possible distribution of leakage points and solvent diffusion paths. Seal information: including the material, thickness, aging degree, etc. of the seal, which affects the sealing performance and is a key component for leak detection. Material properties: physical properties such as elastic modulus, thermal conductivity, and corrosion resistance of each component, which are used to simulate real leakage conditions. Joint and weld characteristics: including joint type, weld location and width, etc. These are high-risk areas where solvents may leak. Fluid channel design: involves the flow path of internal liquids and gases, which has an important impact on the simulation of solvent diffusion and leakage behavior.

[0037] S2, assigning physical properties to the geometric model of the dairy equipment and the solvent to be injected based on the material library corresponding to the 3D modeling software.

[0038] The material library corresponding to the 3D modeling software assigns physical properties to the geometric model of the dairy equipment and the solvent to be injected, specifically:

[0039] Assign physical properties to the geometry model and the solvent to be injected based on the material library to simulate its real characteristics.

[0040] It should be noted that based on the material library corresponding to the 3D modeling software, by selecting or customizing the physical properties of the equipment and solvent, the geometric model of the dairy equipment is given material properties (such as density, elastic modulus, thermal conductivity, etc.), and the injected solvent is given fluid properties (such as density, viscosity, flow pressure, etc.). At the same time, for different parts of the equipment and their contact surfaces with the solvent, the distribution of material properties and the definition of contact conditions are refined to ensure that the model can truly reflect the mechanical and fluid behavior of the equipment in the intelligent solvent leakage detection of the production line, providing basic support for accurate fault simulation and analysis.

[0041] S3, obtaining the initial injection pressure for solvent leakage detection based on historical data, and performing random perturbations through a computer to obtain several different injection pressures to be screened.

[0042] The initial injection pressure for solvent leakage detection is obtained based on historical data, and random perturbations are performed by a computer to obtain several different injection pressures to be screened, specifically:

[0043] Extract relevant data of solvent leakage detection from the historical detection records of dairy beverage production lines, including the injection pressure range, detection effect and fault characteristics when leakage occurs;

[0044] Combined with the equipment type, operating environment and solvent characteristics, the commonly used injection pressure values ​​are counted and calculated as the initial injection pressure, and the average value is used as the initial injection pressure;

[0045] Perform random perturbations on the initial injection pressure value based on the Monte Carlo method to generate a set of candidate pressure values;

[0046] According to the characteristics of the equipment and solvent, the generated candidate pressure values ​​are screened, and extreme values ​​that may cause equipment damage or detection failure are eliminated, thereby obtaining several different injection pressures to be screened.

[0047] It should be noted that the process of screening the generated candidate pressure values ​​according to the characteristics of the equipment and solvent is an existing mature technology, which usually includes a comprehensive evaluation through material strength analysis, equipment pressure test results and solvent fluid mechanics properties to eliminate extreme pressure values ​​that may cause equipment damage or solvent leakage. For example, it can be determined through simulation calculation or experimental verification whether certain pressure values ​​exceed the safety range of the equipment or cause excessive detection errors. Since the relevant methods have been widely used in the field of industrial detection, their principles are clear and the technology is mature, they will not be repeated here.

[0048] S4, respectively applying different injection pressures to be screened to perform solvent leakage detection on the geometric model of the dairy equipment, and obtaining leakage path data and solvent and equipment loss data during the solvent leakage detection.

[0049] The leakage path data includes a detection area coverage influence coefficient and a leakage path influence coefficient;

[0050] The solvent and equipment loss data include solvent and equipment loss cost factors.

[0051] The detection area coverage influence coefficient is used to evaluate the effectiveness of solvent diffusion during solvent leak detection. It reflects the impact of injection pressure on the solvent diffusion range. Too high pressure may cause excessive solvent diffusion, covering unnecessary detection areas and increasing the risk of false detection; while too low pressure may cause insufficient solvent diffusion and fail to cover all potential leaks, thus affecting the comprehensiveness and accuracy of detection. This coefficient can effectively measure the coverage of the detection area, ensure the sensitivity and accuracy of intelligent control of the production line, and avoid insufficient detection results or equipment damage caused by improper injection pressure.

[0052] Analyzing the detection area coverage influence coefficient has the following advantages in solving the problem that the injection pressure control is often inaccurate in the intelligent solvent leakage detection of the dairy beverage production line in the existing technology, resulting in insufficient sensitivity of intelligent control of production lines of different equipment or equipment damage:

[0053] Improve detection accuracy: By reasonably evaluating and adjusting the injection pressure, the detection area coverage influence coefficient can ensure that the solvent diffusion range is moderate, avoiding false detection or missed detection due to excessively high or low pressure, thereby improving the accuracy of intelligent control of the production line.

[0054] Optimize detection sensitivity: This factor helps to accurately adjust the injection pressure according to the equipment type and operating environment, so that the sensitivity of intelligent control of production lines of different equipment can be optimized, avoiding insufficient sensitivity caused by inaccurate pressure and ensuring that equipment failures can be detected in time.

[0055] Reduced risk of equipment damage: By precisely controlling the injection pressure, the detection area coverage coefficient can avoid potential damage to the equipment due to excessive pressure, especially on high-pressure components, reducing the risk of equipment damage caused by excessive diffusion.

[0056] Adapt to equipment differences: This coefficient can flexibly adjust the pressure value according to the characteristics of different dairy equipment to ensure the best detection effect of each equipment and avoid the detection inconsistency caused by mismatched injection pressure between equipment.

[0057] Improve the efficiency of intelligent control of production lines: Through refined pressure control, the detection area coverage influence coefficient can ensure that all potential leakage points are covered, reduce missed detections, and improve the efficiency and timeliness of intelligent control of production lines.

[0058] In summary, the detection area coverage influence coefficient effectively solves the problems of low precision, insufficient sensitivity and equipment damage in the existing technology by optimizing the control of injection pressure, making the intelligent control of the dairy beverage production line more accurate and reliable.

[0059] The specific method for obtaining the detection area coverage influence coefficient is as follows:

[0060] Unitize the three-dimensional geometric model of the dairy beverage production line, divide the equipment into multiple small volume units, label them and obtain a unit set;

[0061] Count the number of equipment surface units that affect the solvent diffusion range and obtain the volume of the affected area;

[0062] Identify and analyze the solvent diffusion path in the unitized equipment model, and calculate the diffusion path rate; combine the diffusion path rate and the volume of the affected area to calculate the detection area coverage influence coefficient;

[0063] The volume of the affected area is calculated as follows:

[0064]

[0065] The calculation formula for the rate of the diffusion path is:

[0066]

[0067] The calculation formula of the detection area coverage influence coefficient is:

[0068]

[0069] Where α is the coverage influence coefficient of the detection area, N is the total number of units in the unit set, and i is the unit number; V i is the volume of the ith unit, Dld i is the leakage path resistance length of the i-th diffusion path.

[0070] The leak path influence coefficient is used to evaluate the effect of injection pressure on the flow pattern of the solvent in the leak path, which in turn affects the accuracy of the leak point location. Different injection pressures will change the flow characteristics of the solvent. For example, high pressure may cause turbulence, while low pressure may cause laminar flow. These changes in flow patterns will affect the propagation of the solvent in the leak path, which in turn affects the accuracy of detection. By analyzing the flow pattern of the leak path, the leak path influence coefficient can help optimize the leak point location algorithm, improve the sensitivity and accuracy of the intelligent control of the production line, and effectively solve the problems of insufficient detection sensitivity and equipment damage caused by inaccurate injection pressure control in the existing technology.

[0071] Analyzing the leakage path influence coefficient has the following advantages in solving the problem that the injection pressure control is often inaccurate in the intelligent solvent leakage detection of the dairy beverage production line, which leads to insufficient sensitivity of intelligent control of production lines of different equipment or equipment damage:

[0072] Optimize flow pattern judgment: By evaluating the flow pattern in the leak path, the leak path influence coefficient can accurately reflect the flow characteristics of the solvent (such as turbulence and laminar flow) under different injection pressures, thereby improving the accuracy of leak point location judgment and avoiding missed detection or false detection problems caused by changes in flow patterns.

[0073] Improve the sensitivity of intelligent control of production lines: This coefficient can help analyze the impact of flow patterns in the leakage path on detection sensitivity, thereby optimizing the injection pressure according to the equipment type and operating environment, thereby improving the detection sensitivity and avoiding insufficient sensitivity of intelligent control of production lines due to inaccurate injection pressure.

[0074] Reduce the risk of equipment damage: By optimizing the flow pattern of the leak path, the leak path influence coefficient can ensure the stability of the solvent flow in the leak path, avoid turbulence caused by high pressure or insufficient flow caused by too low pressure, thereby reducing the risk of equipment damage and ensuring equipment safety.

[0075] Adapting to differences in different equipment: This coefficient can adjust the solvent flow pattern according to the characteristics of different dairy equipment, ensuring that the intelligent control effect of the equipment's production line matches the design and function of the equipment itself, avoiding injection pressure mismatch and inconsistent detection results caused by equipment differences.

[0076] Improve fault location accuracy: The leakage path influence coefficient can optimize the fault location algorithm, improve the accuracy of fault location based on flow pattern and leakage path data, ensure that potential leak points are accurately identified, and thus improve the reliability and accuracy of the detection process.

[0077] In summary, the leakage path influence coefficient can effectively solve the problems of insufficient sensitivity of production line intelligent control, equipment damage and inaccurate detection caused by inaccurate injection pressure control in the existing technology by optimizing the flow characteristics of the solvent in the leakage path, thereby realizing efficient, accurate and safe intelligent control of the dairy beverage production line.

[0078] The specific method for obtaining the leakage path influence coefficient is as follows:

[0079] Based on the solvent leakage detection process under different injection pressures, a set of paths of influence of various injection pressures on the detection results is obtained, wherein the set of paths of influence of various injection pressures on the detection results includes abnormal leakage paths caused by various injection pressures, and a weighted valuation is performed on the abnormal leakage paths based on historical data;

[0080] A regression model is constructed based on the path set and corresponding weights of the impact of various injection pressures on the detection results to generate the leakage path influence coefficient.

[0081] The specific calculation formula of the leakage path influence coefficient is as follows:

[0082]

[0083] Where β is the leakage path influence coefficient, μ1…μ n is the weight estimation corresponding to the path of the solvent leakage detection process under different injection pressures, ω1…ω n YC is the proportion of the paths of the solvent leakage detection process under different injection pressures in the paths of all solvent leakage detection processes. n is the corresponding detection accuracy of the path of the solvent leakage detection process under different injection pressures, n is the number of the path of the solvent leakage detection process under different injection pressures, and e is a natural number.

[0084] The solvent and equipment loss cost coefficient is used to evaluate the solvent consumption during the solvent leak detection process and its impact on the equipment. High injection pressure will lead to increased solvent consumption, which may cause unnecessary waste; low injection pressure may not be able to fully inject the solvent, affecting the continuity of the detection effect. This coefficient also focuses on equipment loss. Excessive injection pressure may impose additional burdens on the equipment and even cause secondary damage to sealing components; low pressure may lead to insufficient detection results and delay the discovery of problems. By evaluating solvent consumption and equipment status data, the solvent and equipment loss cost coefficient can help determine the economy, safety and sustainability of intelligent control of production lines, and avoid the problems of insufficient detection sensitivity and equipment damage caused by inaccurate injection pressure control in existing technologies.

[0085] Analyzing the solvent and equipment loss cost coefficients has the following advantages in solving the problem that the injection pressure control is often inaccurate in the intelligent solvent leakage detection of the dairy beverage production line, which leads to insufficient sensitivity of intelligent control of production lines of different equipment or equipment damage:

[0086] The specific method for obtaining the solvent and equipment loss cost coefficient is as follows:

[0087] Obtain the ratio of solvent consumption to equipment loss at several different injection pressures;

[0088] Calculate the loss cost standard deviation and the loss cost average of the ratio of solvent consumption to equipment loss at different injection pressures;

[0089] According to the standard deviation of loss cost and the average value of loss cost, calculate the coefficient of variation of the ratio of solvent consumption to equipment loss;

[0090] The solvent and equipment loss cost coefficients are calculated based on the preset solvent and equipment loss cost coefficient calculation formula using the coefficient of variation of the ratio of solvent consumption to equipment loss.

[0091] The specific calculation formula of the coefficient of variation of the ratio of solvent consumption to equipment loss is as follows:

[0092]

[0093] The specific calculation formula of the solvent and equipment loss cost coefficient is as follows:

[0094]

[0095] Where μ is the coefficient of variation of the ratio of solvent consumption to equipment loss, wl j is the ratio of the j-th solvent consumption to the equipment loss, N is the total number of collected data, j is the data number; χ is the solvent and equipment loss cost coefficient.

[0096] This embodiment acquires the equipment operating environment and component data, establishes a geometric model and assigns corresponding physical properties, which can truly simulate the behavior of the equipment in the event of a solvent leak. Further, by accurately calculating the leakage path and the detection area coverage influence coefficient, the injection pressure is optimized to ensure that the solvent diffusion range is moderate, avoid false detection or missed detection, and reduce the risk of equipment damage. This method not only adapts to equipment differences and improves the detection consistency between different equipment, but also improves the efficiency of intelligent control of the production line by optimizing the analysis of solvent flow patterns, and solves the problems of insufficient sensitivity and equipment damage in the prior art, thereby achieving more accurate and reliable intelligent control of the production line.

[0097] Example 2, S5, constructs an injection pressure assessment model based on a convolutional neural network according to leakage path data and equipment and solvent loss data.

[0098] The injection pressure assessment model is constructed based on the convolutional neural network according to the defect location data and the detection robustness data, specifically:

[0099] The obtained detection area coverage influence coefficient, leakage path influence coefficient, and solvent and equipment loss cost coefficient are used to establish an injection pressure assessment model to generate an injection pressure assessment coefficient;

[0100] The specific calculation formula of the injection pressure assessment coefficient is as follows:

[0101]

[0102] Wherein, α is the detection area coverage influence coefficient, β is the leakage path influence coefficient, χ is the solvent and equipment loss cost coefficient, YL is the injection pressure evaluation coefficient, λ1 is the preset proportional coefficient of the detection area coverage influence coefficient, λ2 is the preset proportional coefficient of the leakage path influence coefficient, and λ3 is the preset proportional coefficient of the solvent and equipment loss cost coefficient.

[0103] S6, based on the output of the injection pressure evaluation model and the corresponding injection pressure to be screened, construct a pressure-detection effect display model, and perform data analysis to obtain the screened injection pressure, which is applied to the injection pressure control of intelligent solvent leakage detection in the dairy beverage production line.

[0104] According to the output of the injection pressure evaluation model and the corresponding injection pressure to be screened, a pressure-detection effect display model is constructed, specifically:

[0105] Obtaining injection pressure assessment coefficients of outputs under different injection pressures to be screened from the injection pressure assessment model;

[0106] The X-axis is set as the injection pressure to be screened, and the Y-axis is represented as the injection pressure evaluation coefficient, and a rectangular coordinate system is established;

[0107] The injection pressure assessment coefficient corresponding to each injection pressure to be screened is marked on the rectangular coordinate system to form marking points, and a smooth curve is used to smoothly connect several marking points to construct a pressure-detection effect display model.

[0108] The above data is analyzed to obtain the selected injection pressure, which is applied to the injection pressure control of intelligent solvent leakage detection in the dairy beverage production line, specifically:

[0109] The injection pressures to be screened corresponding to the peak values ​​of the pressure-detection effect display model are screened to obtain a first screening group;

[0110] The quadratic derivatives of each injection pressure to be screened in the first screening group on the pressure-detection effect display model are compared, and the injection pressure to be screened with the smallest comparison result is used as the injection pressure after screening, and is applied to the injection pressure control of intelligent solvent leakage detection in the dairy beverage production line.

[0111] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters in the formula are set by technicians in this field according to actual conditions.

[0112] The above embodiments may be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented by software, the above embodiments may be implemented in whole or in part in the form of a computer program product.

[0113] Those of ordinary skill in the art will appreciate that the modules and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0114] In addition, each functional module in each embodiment of the present application may be integrated into one processing module, or each module may exist physically separately, or two or more modules may be integrated into one module.

[0115] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

[0116] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. Intelligent integrated management system for dairy beverage production line, characterized by: It includes modeling module, refinement module, injection pressure acquisition module, data acquisition module, injection pressure evaluation module and injection pressure screening module: Modeling module, used to obtain the operating environment and component data of the dairy beverage production line to be tested, and build a geometric model of the dairy equipment based on 3D modeling software; A refinement module is used to assign physical properties to the geometric model of the dairy equipment and the solvent to be injected based on the material library corresponding to the 3D modeling software; An injection pressure acquisition module is used to obtain the initial injection pressure for solvent leakage detection based on historical data, and to perform random perturbations through a computer to obtain several different injection pressures to be screened; A data acquisition module, used for respectively applying different injection pressures to be screened to perform solvent leakage detection on the geometric model of dairy equipment, and obtaining leakage path data and solvent and equipment loss data during solvent leakage detection; An injection pressure assessment module, which is used to build an injection pressure assessment model based on a convolutional neural network according to the leakage path data and the equipment and solvent loss data; The injection pressure screening module is used to build a pressure-detection effect display model based on the output of the injection pressure evaluation model and the corresponding injection pressure to be screened, and perform data analysis to obtain the screened injection pressure, which is applied to the injection pressure control of intelligent solvent leakage detection in the dairy beverage production line.

2. The intelligent integrated management system for the dairy beverage production line according to claim 1, characterized in that: The operating environment includes temperature, humidity, pressure conditions, ambient air flow, vibration and noise; The component data includes equipment structural parameters, seal information, material properties, joint and weld properties, and fluid channel design; The geometric model of dairy equipment constructed based on 3D modeling software is specifically as follows: Use 3D modeling software to create basic geometric shapes based on the operating environment and component data; adding detailed structures to the basic geometry, including seams and welds, seals, and flanges and bolts; The geometric model of dairy equipment is divided into small unit grids, and the distribution of small unit grids in high stress areas is refined.

3. The intelligent integrated management system for the dairy beverage production line according to claim 2, characterized in that: The initial injection pressure for solvent leakage detection is obtained based on historical data, and random perturbations are performed by a computer to obtain several different injection pressures to be screened, specifically: Extract relevant data of solvent leakage detection from the historical detection records of dairy beverage production lines, including the injection pressure range, detection effect and fault characteristics when leakage occurs; Combined with the equipment type, operating environment and solvent characteristics, the commonly used injection pressure values ​​are counted and calculated as the initial injection pressure, and the average value is used as the initial injection pressure; Perform random perturbations on the initial injection pressure value based on the Monte Carlo method to generate a set of candidate pressure values; According to the characteristics of the equipment and solvent, the generated candidate pressure values ​​are screened, and extreme values ​​that may cause equipment damage or detection failure are eliminated, thereby obtaining several different injection pressures to be screened.

4. The intelligent integrated management system for the dairy beverage production line according to claim 3, characterized in that: The leakage path data includes the detection area coverage influence coefficient and the leakage path influence coefficient; the specific method for obtaining the detection area coverage influence coefficient is as follows: Unitize the three-dimensional geometric model of the dairy beverage production line, divide the equipment into multiple small volume units, label them and obtain a unit set; Count the number of equipment surface units that affect the solvent diffusion range and obtain the volume of the affected area; The solvent diffusion path is identified and analyzed in the unitized equipment model, and the rate of the diffusion path is calculated; the detection area coverage influence coefficient is calculated by combining the rate of the diffusion path and the volume analysis of the affected area.

5. The intelligent integrated management system for the dairy beverage production line according to claim 4, characterized in that: The specific method for obtaining the leakage path influence coefficient is as follows: Based on the solvent leakage detection process under different injection pressures, a set of paths of influence of various injection pressures on the detection results is obtained, wherein the set of paths of influence of various injection pressures on the detection results includes abnormal leakage paths caused by various injection pressures, and a weighted valuation is performed on the abnormal leakage paths based on historical data; A regression model is constructed based on the path set and corresponding weights of the impact of various injection pressures on the detection results to generate the leakage path influence coefficient.

6. The intelligent integrated management system for the dairy beverage production line according to claim 5, characterized in that: The solvent and equipment loss data include solvent and equipment loss cost coefficients; the specific method for obtaining the solvent and equipment loss cost coefficients is as follows: Obtain the ratio of solvent consumption to equipment loss at several different injection pressures; Calculate the loss cost standard deviation and the loss cost average of the ratio of solvent consumption to equipment loss at different injection pressures; According to the standard deviation of loss cost and the average value of loss cost, calculate the coefficient of variation of the ratio of solvent consumption to equipment loss; The solvent and equipment loss cost coefficients are calculated based on the preset solvent and equipment loss cost coefficient calculation formula using the coefficient of variation of the ratio of solvent consumption to equipment loss.

7. The intelligent integrated management system for the dairy beverage production line according to claim 6, characterized in that: According to the output of the injection pressure evaluation model and the corresponding injection pressure to be screened, a pressure-detection effect display model is constructed, specifically: Obtaining injection pressure assessment coefficients of outputs under different injection pressures to be screened from the injection pressure assessment model; The X-axis is set as the injection pressure to be screened, and the Y-axis is represented as the injection pressure evaluation coefficient, and a rectangular coordinate system is established; The injection pressure assessment coefficient corresponding to each injection pressure to be screened is marked on the rectangular coordinate system to form marking points, and a smooth curve is used to smoothly connect several marking points to construct a pressure-detection effect display model.

8. The intelligent integrated management system for the dairy beverage production line according to claim 7, characterized in that: The above data is analyzed to obtain the selected injection pressure, which is applied to the injection pressure control of intelligent solvent leakage detection in the dairy beverage production line, specifically: The injection pressures to be screened corresponding to the peak values ​​of the pressure-detection effect display model are screened to obtain a first screening group; The quadratic derivatives of each injection pressure to be screened in the first screening group on the pressure-detection effect display model are compared, and the injection pressure to be screened with the smallest comparison result is used as the injection pressure after screening, and is applied to the injection pressure control of intelligent solvent leakage detection in the dairy beverage production line.

9. The intelligent integrated management system for the dairy beverage production line according to claim 8, characterized in that: The injection pressure assessment model is constructed based on the convolutional neural network according to the defect location data and the detection robustness data, specifically: The obtained detection area coverage influence coefficient, leakage path influence coefficient, and solvent and equipment loss cost coefficient are used to establish an injection pressure assessment model to generate an injection pressure assessment coefficient; The specific calculation formula of the injection pressure assessment coefficient is as follows: Wherein, α is the detection area coverage influence coefficient, β is the leakage path influence coefficient, χ is the solvent and equipment loss cost coefficient, YL is the injection pressure evaluation coefficient, λ1 is the preset proportional coefficient of the detection area coverage influence coefficient, λ2 is the preset proportional coefficient of the leakage path influence coefficient, and λ3 is the preset proportional coefficient of the solvent and equipment loss cost coefficient.

10. The intelligent integrated management system for the dairy beverage production line according to claim 9, characterized in that: The volume of the affected area is calculated as follows: The calculation formula for the rate of the diffusion path is: The calculation formula of the detection area coverage influence coefficient is: Where α is the coverage influence coefficient of the detection area, N is the total number of units in the unit set, and i is the unit number; V i is the volume of the ith unit, Dld i is the leakage path resistance length of the i-th diffusion path.

Images