Background monitoring system based on beverage dispenser

Abstract

The invention relates to the technical field of data monitoring analysis, in particular to a background monitoring system based on a beverage dispenser, which comprises a data acquisition module for calculating a dynamic deviation coefficient between a current ratio and a standard ratio; the compensation decision module is used for outputting flow compensation parameters of each channel based on the dynamic deviation coefficient; the regulation and control module adjusts the opening degree of the electromagnetic valve based on the flow compensation parameters; the uniformity extraction module is used for collecting the adjusted vibration frequency spectrum of the mixing bin and extracting a target frequency band energy value as a mixing uniformity index; and the risk prediction module is used for triggering a biological membrane risk prediction model when the mixing uniformity index is lower than a uniformity threshold value. According to the method, the problem of non-uniform mixing is found in time, early warning and optimization adjustment are triggered, and the matching problem and biological membrane generation caused by non-uniform mixing are effectively avoided.

Description

Technical Field

[0001] The present invention relates to the technical field of data monitoring and analysis, and particularly to a background monitoring system based on a beverage machine. Background Art

[0002] With the rapid development of the beverage production industry, automated and intelligent beverage production equipment has played an increasingly important role in improving production efficiency and ensuring product quality. In the process of beverage production, the precise proportioning of raw materials is one of the key factors to ensure the quality and taste stability of beverages. However, the proportioning control method of traditional beverage machines mainly relies on fixed quantitative control, and fails to fully consider the real-time changes in raw material flow and the influence of external factors on the proportioning, which is prone to inaccurate proportioning and thus affects product quality.

[0003] Meanwhile, in the process of beverage production, due to the complexity of the raw material mixing and storage environment, the formation of biofilms has become a major problem in equipment operation. A biofilm is a viscous film attached to the surface of the equipment and composed of microbial populations. It not only affects the hygienic quality of beverages, but may also block pipelines and reduce equipment efficiency. Currently, the prevention and control of biofilms mostly rely on regular cleaning and disinfection measures, but these methods have problems such as frequent manual intervention, low efficiency, energy waste, etc., and cannot adjust the cleaning strategy in real time according to different production states and environmental conditions. Summary of the Invention

[0004] The present invention provides a background monitoring system based on a beverage machine.

[0005] A background monitoring system based on a beverage machine includes:

[0006] A data acquisition module: to obtain the instantaneous flow values of each raw material channel of the beverage machine in real time, and calculate the dynamic deviation coefficient D between the current proportioning and the standard proportioning;

[0007] A compensation decision module: input the dynamic deviation coefficient into the compensation decision module, and output the flow compensation parameters of each channel;

[0008] A regulation module: adjust the solenoid valve opening based on the flow compensation parameters;

[0009] A uniformity extraction module: collect the vibration spectrum of the adjusted mixing bin, perform wavelet packet decomposition on the vibration spectrum, and extract the energy value of the target frequency band as the mixing uniformity index;

[0010] A risk prediction module: when the mixing uniformity index is lower than the uniformity threshold, trigger the biofilm risk prediction model, output the risk level of biofilm formation, and generate a real-time sterilization strategy.

[0011] Optionally, in the data acquisition module, ultrasonic flow meters are installed at the outlets of each raw material channel to synchronously collect the instantaneous flow rate data of the syrup, water, and gas channels at a sampling frequency of 100 Hz. The instantaneous flow rate data is filtered to eliminate pulse noise interference. Based on the measured flow rate value after filtering and the standard ratio flow rate value, the dynamic flow deviation coefficient D is calculated;

[0012] Calculate the dynamic deviation coefficient: where n is the number of raw material types, Q i实测 is the measured flow rate value after filtering, Q i标准 is the standard ratio flow rate value.

[0013] Optionally, the flow compensation parameter is calculated as:

[0014] δQ i =α i ×(Q i标准 -Q i实测 (, where δQ i is the flow compensation parameter of the i-th raw material channel, Q i标准 is the standard flow rate of the i-th raw material, Q i实测 is the actual flow rate of the i-th raw material, α i is the compensation coefficient;

[0015] In order to adjust the flow rate according to the dynamic deviation coefficient D, the compensation coefficient α i is calculated as follows:

[0016] where α i is the flow compensation coefficient of the i-th raw material channel, D i is the flow deviation coefficient of the i-th raw material channel, is the total deviation coefficient of all raw material channels.

[0017] Optionally, the regulation module is based on a PID controller to calculate the solenoid valve opening adjustment amount. The opening adjustment amount δθ i is calculated by the following formula: where K p is the proportional gain, which determines the influence degree of the current deviation, K i is the integral gain, which determines the cumulative effect of past errors, K d is the derivative gain, which determines the influence of the error change rate, e i is the flow deviation, is the integral part of the flow deviation, representing the past cumulative deviation, is the derivative part of the flow deviation, representing the deviation change rate.

[0018] Optionally, the uniformity extraction module specifically includes:

[0019] Vibration spectrum acquisition unit: It collects the vibration signals of the mixing bin in real time through vibration sensors installed on the mixing bin, and converts the vibration signals into digital data through a signal conditioning circuit. This data is filtered during the acquisition process to remove low-frequency noise and high-frequency interference. The sampling frequency is set to fs, and fs is taken as 10 kHz to ensure coverage of the frequency band of interest (2 - 5 kHz).

[0020] Wavelet packet decomposition unit: It performs wavelet packet transform decomposition on the collected vibration signals. The wavelet packet transform decomposition decomposes the signals into multiple frequency bands and extracts the signal features within the target frequency band. The target frequency band is set to 2 - 5 kHz. The specific steps are as follows:

[0021] Set the wavelet basis function ψ(t) and the decomposition level N, select the mother wavelet suitable for high-frequency and low-frequency signals, the Daubechies wavelet, and use the wavelet packet decomposition algorithm to decompose the vibration signal x(t):

[0022] Among them, c jk is the decomposition coefficient, and ψ jk (t) is the wavelet packet basis function of the j-th layer and the k-th frequency band; the wavelet packet decomposition process decomposes the vibration signal into multiple frequency bands, and each frequency band represents a frequency range.

[0023] Frequency band energy value extraction unit: After wavelet packet decomposition, select the wavelet packet coefficients of the target frequency band 2 - 5 kHz for energy calculation, set the target frequency band, and calculate the frequency band energy value E 2-5kHz for each frequency band.

[0024] Mixing uniformity index extraction unit: Extract the frequency band energy value E 2-5kHz of the target frequency band 2 - 5 kHz as the mixing uniformity index. The magnitude of the energy value reflects the uniformity of the materials in the mixing bin. A higher energy value usually indicates a more uniform mixing state, while a lower energy value may indicate uneven mixing.

[0025] Optionally, in the frequency band energy value extraction unit, set the frequency bands f1 = 2 kHz and f2 = 5 kHz, and calculate the frequency band energy value for each frequency band:

[0026] Among them, E 2-5kHz is the total energy value of the 2 - 5 kHz frequency band, c jk is the j-th layer and the k-th frequency band wavelet packet coefficient, and f j is the frequency band corresponding to the frequency j.

[0027] Optionally, the uniformity threshold is expressed as E min , which is set based on the uniformity reference value. E min = 0.7×Ebase , where E base is the reference value of the new machine's mixing uniformity under the equipment calibration state (i.e., the optimal energy value of the mixing bin). The experimental results show that when the energy value E 2-5kHz drops to 70% of the reference value, the mixing non-uniformity reaches 15%. Therefore, when the mixing uniformity is less than 0.7 times the reference value of the uniformity, that is, E 2-5kHz <0.7×E base at this time, E base is the reference value, it is determined that the mixing is non-uniform, and the biofilm risk prediction model is automatically triggered. The biofilm risk prediction model predicts the risk of biofilm formation based on the real-time mixing uniformity value, the raw material ratio, and historical data, and outputs the prediction result.

[0028] Optionally, the biofilm risk prediction model includes data input and preprocessing, risk assessment, and output. The biofilm risk prediction model predicts the risk of biofilm formation based on the real-time mixing uniformity value, the dynamic deviation coefficient, and historical data. The biofilm risk prediction model outputs the risk level of biofilm formation, including three levels: high, medium, and low.

[0029] Optionally, the data input and preprocessing include the mixing uniformity index E 2-5kHz , the dynamic deviation coefficient D, the environmental temperature and humidity data, and historical data. The historical data includes the historical dynamic deviation coefficient, the historical mixing uniformity, and the historical environmental temperature and humidity data;

[0030] Normalized mixing uniformity value E′:

[0031] Define the historical environmental impact factor η env : where and are the reference values of temperature and humidity in the historical data respectively;

[0032] The risk assessment includes setting a comprehensive evaluation function R risk , the comprehensive evaluation function R risk outputs the risk value of the biofilm. The risk value includes the following factors:

[0033] Mixing uniformity factor: The greater the non-uniformity, the higher the biofilm risk;

[0034] Historical dynamic deviation factor: The greater the ratio deviation, the higher the biofilm risk;

[0035] Environmental temperature and humidity factor: The more suitable the temperature and humidity are for biofilm growth, the higher the risk;

[0036] The comprehensive evaluation function is expressed as: R risk =α1·E′+α2·Dmix +α3·η env , where α1, α2, and α3 are weight coefficients obtained from historical data;

[0037] According to R risk value, set a threshold and divide it into three risk levels:

[0038] R risk < 0.3 Low risk;

[0039] 0.3 ≤ R risk < 0.6 Medium risk;

[0040] R risk ≥ 0.6 High risk.

[0041] Optionally, the real-time sterilization strategy includes:

[0042] Low risk: Do not perform sterilization operations, but maintain regular inspections;

[0043] Medium risk: Initiate mild sterilization and increase equipment monitoring, reminding the operator to pay attention to adjustments;

[0044] High risk: Immediately initiate high-intensity sterilization and perform equipment cleaning, and adjust the raw material ratio.

[0045] Advantages of the present invention:

[0046] In the present invention, by real-time monitoring the instantaneous flow rates of the raw material channels of the beverage machine, combined with dynamic deviation coefficient calculation and compensation decision-making, the accuracy of the raw material ratio of the beverage machine is improved. Based on this technology, the opening degree of the solenoid valve can be adjusted in real time to ensure that the raw material flow rate remains in the optimal state, thereby avoiding the risk of biofilm formation caused by ratio errors. This technology significantly improves the ratio stability of the beverage machine, extends the service life of the equipment, and reduces the maintenance cost brought by biofilms.

[0047] In the present invention, based on wavelet packet decomposition technology for real-time monitoring of mixing uniformity, the vibration spectrum of the mixing bin is analyzed using wavelet packet decomposition technology, and the energy value in the 2 - 5 kHz frequency band is extracted as the mixing uniformity index, which can reflect the mixing effect in real time. Through the correlation analysis between the spectrum characteristics and the mixing uniformity, accurate mixing state data can be provided for the operator. In practical applications, this technology can timely detect problems of uneven mixing, trigger warnings and optimization adjustments, and effectively avoid ratio problems and biofilm generation caused by uneven mixing.

[0048] The present invention proposes a prediction of biofilm formation and an intelligent sterilization strategy based on risk levels by combining real-time mixing uniformity, raw material ratio deviation, and environmental temperature and humidity data. When the biofilm formation risk reaches a set threshold, the system automatically adjusts the raw material ratio and triggers corresponding sterilization measures, thereby preventing the formation and accumulation of biofilms. Different from traditional regular cleaning methods, the strategy of the present invention is based on real-time data and risk prediction, and intelligently triggers sterilization operations, greatly improving the sterilization efficiency and reducing the waste of energy and resources. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are only those of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0050] Figure 1 Schematic diagram of the system function module of the embodiment of the present invention;

[0051] Figure 2 Schematic diagram of each factor of risk assessment of the embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0052] The present invention will be described in detail below in conjunction with the drawings and specific embodiments. At the same time, it should be noted here that in order to make the embodiments more detailed, the following embodiments are the best and preferred embodiments. For some well-known technologies, those skilled in the art can also adopt other alternative methods for implementation; and the drawings are only for more specific description of the embodiments, and are not intended to specifically limit the present invention.

[0053] It should be pointed out that in the specification, when referring to "an embodiment", "embodiment", "exemplary embodiment", "some embodiments", etc., it indicates that the described embodiment may include specific features, structures or characteristics, but not necessarily every embodiment includes the specific feature, structure or characteristic. In addition, when combining an embodiment to describe a specific feature, structure or characteristic, implementing such a feature, structure or characteristic in combination with other embodiments (whether explicitly described or not) should be within the knowledge of those skilled in the relevant art.

[0054] Generally, the terms can be understood at least in part from their use in the context. For example, at least in part depending on the context, the term "one or more" used herein can be used to describe any feature, structure or characteristic in a singular sense, or can be used to describe a combination of features, structures or characteristics in a plural sense. In addition, the term "based on" can be understood as not necessarily intended to convey a set of exclusive factors, but rather, at least in part depending on the context, allowing for the existence of other factors that may not be explicitly described.

[0055] As Figure 1 - Figure 2 shown, a background monitoring system based on a beverage machine includes:

[0056] Data acquisition module: It acquires the instantaneous flow rate values of each raw material channel of the beverage machine in real time, and calculates the dynamic deviation coefficient D between the current ratio and the standard ratio;

[0057] Compensation decision module: It inputs the dynamic deviation coefficient into the compensation decision module and outputs the flow compensation parameters of each channel;

[0058] Regulation module: It adjusts the solenoid valve opening based on the flow compensation parameters;

[0059] Uniformity extraction module: It collects the vibration spectrum of the adjusted mixing bin, performs wavelet packet decomposition on the vibration spectrum, and extracts the energy value of the target frequency band as the mixing uniformity index;

[0060] Risk prediction module: When the mixing uniformity index is lower than the uniformity threshold, it triggers the biofilm risk prediction model, outputs the risk level of biofilm formation, and generates a real-time sterilization strategy.

[0061] In the data acquisition module; an ultrasonic flowmeter is installed at the outlet of each raw material channel to synchronously collect the instantaneous flow rate data of the syrup, water, and gas channels at a sampling frequency of 100 Hz, filter the instantaneous flow rate data to eliminate pulse noise interference, and calculate the dynamic flow deviation coefficient D based on the filtered measured flow rate value and the standard ratio flow rate value;

[0062] Calculating the dynamic deviation coefficient: where n is the number of raw material types, Q i实测 is the filtered measured flow rate value, and Q i标准 is the standard ratio flow rate value.

[0063] The flow compensation parameter is calculated as:

[0064] δQ i =α i ×(Q i标准 -Q i实测 ), where δQ i is the flow compensation parameter of the i-th raw material channel, Q i标准 is the standard flow rate of the i-th raw material, Q i实测 is the actual flow rate of the i-th raw material, and α i is the compensation coefficient;

[0065] In order to adjust the flow according to the dynamic deviation coefficient D, the compensation coefficient α i is calculated as follows:

[0066] where α iis the flow compensation coefficient of the i-th raw material channel, D i is the flow deviation coefficient of the i-th raw material channel, is the total deviation coefficient of all raw material channels.

[0067] The regulation module calculates the solenoid valve opening adjustment amount based on the PID controller. The opening adjustment amount δθ i is calculated by the following formula: where K p is the proportional gain, which determines the influence degree of the current deviation, K i is the integral gain, which determines the cumulative effect of past errors, K d is the derivative gain, which determines the influence of the error change rate, e i is the flow deviation, is the integral part of the flow deviation, representing the past cumulative deviation, is the derivative part of the flow deviation, representing the deviation change rate.

[0068] When beverage raw materials (such as syrup, water, gas) are mixed, a large number of tiny bubbles will be generated. During the bubble collapse process, energy will be released, exciting the high-frequency vibration of the mixing chamber wall (usually in the range of 2 - 5 kHz); the more uniform the mixing, the more regular the bubble distribution, the more concentrated the energy release during bubble collapse, and the more stable the energy value in the specific frequency band of the vibration spectrum; when the mixing is uneven (such as local aggregation of syrup), the bubble distribution is irregular and the energy release during bubble collapse is dispersed; the energy value in the 2 - 5 kHz frequency band is extracted through wavelet packet decomposition as a quantization index of the mixing uniformity; the higher and more concentrated the energy value, the more uniform the mixing; the dispersed energy value or the increase in high-frequency components indicate uneven mixing or local aggregation. By real-time monitoring the mixing uniformity, the raw material ratio or mixing parameters can be adjusted in a timely manner to ensure the consistency of beverage quality. Therefore, the uniformity extraction module specifically includes:

[0069] Vibration spectrum acquisition unit: The vibration signal of the mixing chamber is collected in real time through a vibration sensor installed on the mixing chamber, and the vibration signal is converted into digital data through a signal conditioning circuit. This data is filtered during the acquisition process to remove low-frequency noise and high-frequency interference. The sampling frequency is set to fs, and fs is taken as 10 kHz to ensure coverage of the frequency band of interest (2 - 5 kHz);

[0070] Wavelet packet decomposition unit: The collected vibration signal is decomposed by wavelet packet transform. The wavelet packet transform decomposes the signal into multiple frequency bands and extracts the signal characteristics within the target frequency band. The target frequency band is set to 2 - 5 kHz. The specific steps are as follows:

[0071] Set the wavelet basis function ψ(t) and the decomposition level N. Select the mother wavelet applicable to high-frequency and low-frequency signals, the Daubechies wavelet, and decompose the vibration signal x(t) using the wavelet packet decomposition algorithm:

[0072] Among them, c jk is the decomposition coefficient, and ψ jk (t) is the wavelet packet basis function of the j-th layer and the k-th frequency band; the wavelet packet decomposition process decomposes the vibration signal into multiple frequency bands, and each frequency band represents a frequency range;

[0073] Frequency band energy value extraction unit: After wavelet packet decomposition, select the wavelet packet coefficients of the target frequency band 2 - 5 kHz for energy calculation, set the target frequency band, and calculate the frequency band energy value E 2-5kHz for each frequency band;

[0074] Mixing uniformity index extraction unit: Extract the frequency band energy value E 2-5kHz of the target frequency band 2 - 5 kHz as the mixing uniformity index. The magnitude of the energy value reflects the uniformity of the materials in the mixing bin. A higher energy value usually indicates a more uniform mixing state, while a lower energy value may indicate uneven mixing.

[0075] In the frequency band energy value extraction unit, set the frequency bands f1 = 2 kHz and f2 = 5 kHz, and calculate the frequency band energy value for each frequency band:

[0076] Among them, E 2-5kHz is the total energy value of the 2 - 5 kHz frequency band, c jk is the j-th layer and the k-th frequency band of the wavelet packet coefficient, and f j is the frequency band corresponding to the frequency j.

[0077] The uniformity threshold is expressed as E min , which is set based on the uniformity reference value. E min = 0.7×E base , where E base is the new machine mixing uniformity reference value (i.e., the optimal energy value of the mixing bin) under the equipment calibration state. Experimental results show that when the energy value E 2-5kHz drops to 70% of the reference value, the mixing non-uniformity reaches 15%. Therefore, when the mixing uniformity is less than 0.7×the uniformity reference value, that is, E 2-5kHz <0.7×E base , with E base being the reference value, it is determined that the mixing is uneven, and the biofilm risk prediction model is automatically triggered. The biofilm risk prediction model predicts the risk of biofilm formation based on the real-time mixing uniformity value, the raw material ratio, and historical data, and outputs the prediction result.

[0078] When the mixing is uneven, high-viscosity raw materials (such as syrup) tend to form local aggregation in low-velocity areas (dead corners, elbows) of pipelines or mixing bins. These aggregation areas will become nutrient-rich areas, providing sufficient energy sources for microbial growth. Local aggregation of raw materials will change the flow state in the pipeline and form a flow separation zone (i.e., an area with a flow velocity significantly lower than that of the mainstream area). In the flow separation zone, the liquid retention time is prolonged and the shear force is reduced, creating conditions for microbial attachment and colonization. Microorganisms tend to form biofilms in an environment with low shear force and high nutrient concentration. The flow separation zone just meets these two conditions:

[0079] Low shear force: microorganisms are not easily washed away by the fluid and are easily attached to the tube wall;

[0080] High nutrient concentration: Locally concentrated raw materials provide a continuous supply of nutrients to microorganisms.

[0081] Mixing unevenness (quantified by vibration spectrum energy dispersion) is positively correlated with biofilm risk. Therefore, the biofilm risk prediction model includes data input and preprocessing, risk assessment and output. The biofilm risk prediction model predicts the risk of biofilm formation based on real-time mixing uniformity value, dynamic deviation coefficient and historical data. The biofilm risk prediction model outputs the risk level of biofilm formation, including high, medium and low levels.

[0082] Data input and preprocessing including mixing uniformity index E 2-5kHz , dynamic deviation coefficient D, ambient temperature and humidity data and historical data, the historical data includes historical dynamic deviation coefficient, mixing uniformity history and historical ambient temperature and humidity data;

[0083] Standardized mixing uniformity value E′:

[0084] Define the historical environmental impact factor η based on environmental temperature and humidity data env : in and They are the temperature and humidity benchmark values ​​in historical data respectively;

[0085] Risk assessment includes setting a comprehensive assessment function R risk , comprehensive evaluation function R risk Output the biofilm risk value, which includes the following factors:

[0086] Mixing uniformity factor: the greater the unevenness, the higher the biofilm risk;

[0087] Historical dynamic deviation factor: The greater the ratio deviation, the higher the biofilm risk;

[0088] Environmental temperature and humidity factor: The more suitable the temperature and humidity are for the growth of the biofilm, the higher the risk;

[0089] The comprehensive evaluation function is expressed as: R risk = α1·E′ + α2·D mix + α3·η env , where α1, α2, and α3 are weight coefficients obtained from historical data;

[0090] According to the value of R risk set the threshold, and divide it into three risk levels:

[0091] R risk < 0.3 Low risk;

[0092] 0.3 ≤ R risk < 0.6 Medium risk;

[0093] R risk ≥ 0.6 High risk.

[0094] Low risk: This interval indicates that the system is operating well, with high mixing uniformity, the ratio is close to the standard, and the environmental conditions have little impact on the formation of the biofilm;

[0095] Medium risk: In this interval, there may be some uneven mixing or slight ratio errors, but it has not reached the critical point of biofilm formation. It is necessary to pay attention to monitoring and take timely measures;

[0096] High risk: This interval indicates that the system has obvious ratio errors, uneven mixing or adverse environmental conditions, resulting in a significant increase in the risk of biofilm formation. Immediate measures need to be taken, such as cleaning or sterilization operations.

[0097] The ways to obtain the weight coefficients α1, α2, and α3 are as follows:

[0098] Collect actual data under different environments, ratio deviations, and mixing uniformity conditions through multiple experiments, monitor the formation of the biofilm (formation time, thickness, density), conduct regression analysis on the experimental data, find out the specific contributions of each factor to the formation of the biofilm, and fit the model through the least squares statistical method to determine the optimal values of α1, α2, and α3.

[0099] For the influence of mixing uniformity, set experimental groups, control different mixing uniformities, and observe the speed and amount of biofilm formation, so as to fit the value of α1.

[0100] Use parameter sensitivity analysis to identify and quantify the influence of each factor (mixing uniformity, ratio deviation, environmental conditions) on the formation of the biofilm.

[0101] By changing the values of α1, α2, and α3, observe how the model output (i.e., the risk of biofilm formation) changes, thereby determining the sensitivity of these parameters.

[0102] Real-time sterilization strategies include:

[0103] Corresponding to different risk levels, the system automatically or manually selects and executes the corresponding sterilization strategies:

[0104] Low risk: Do not perform sterilization operations, but maintain regular inspections;

[0105] Medium risk: Initiate mild sterilization and increase equipment monitoring, reminding the operator to pay attention to adjustments;

[0106] High risk: Immediately initiate high-intensity sterilization and conduct equipment cleaning, and adjust the raw material ratio;

[0107] Specifically:

[0108] 1. Low-risk sterilization strategy: At this risk level, the system detects good mixing uniformity, the raw material ratio is close to the standard, the environmental conditions are suitable, and the risk of biofilm formation is low. Therefore, there is no need to immediately perform sterilization operations;

[0109] Strategy:

[0110] Regular inspection: Maintain the regular operating state of the equipment, conduct regular inspections and maintenance to ensure the long-term stable operation of the system;

[0111] Extended cleaning cycle: Extend the cleaning cycle of the equipment, but still maintain the hygiene inspection of the equipment, such as performing low-intensity cleaning once a week or once a month.

[0112] Trigger condition: At this stage, the sterilization strategy will not be activated, but cleaning operations are performed regularly.

[0113] 2. Medium-risk sterilization strategy: At this risk level, the system detects some slight mixing unevenness or raw material ratio errors, but the formation of biofilm has not reached the critical value. However, as the risk increases, the system still needs to actively respond to possible biofilm problems;

[0114] Strategy:

[0115] Early sterilization warning: The system issues a warning to remind the operator that the risk of biofilm formation is increasing and initiates a mild sterilization procedure;

[0116] Regular sterilization: Strengthen the cleaning frequency of the equipment, perform more frequent low-intensity sterilization, such as setting the automated sterilization cycle to once a day or once every two days;

[0117] Adjust operating parameters: Optimize operating conditions, such as slightly reducing the raw material flow rate and adjusting the raw material ratio to improve the mixing effect and reduce the risk of biofilm formation;

[0118] Trigger condition: Trigger based on a mild sterilization mechanism, and decide whether to take more aggressive sterilization measures according to the operator's feedback.

[0119] 3. High-risk sterilization strategy: At this risk level, the system detects that the mixing uniformity is extremely poor, the raw material ratio is unqualified, and conditions such as environmental temperature and humidity are very conducive to biofilm formation. At this time, more urgent and comprehensive sterilization measures need to be taken to prevent the rapid accumulation of biofilms and equipment damage;

[0120] Strategy:

[0121] High-intensity sterilization: Immediately initiate a high-intensity sterilization program (high temperature, ultraviolet sterilization, chemical sterilization) to ensure that biofilms are completely removed in the shortest time;

[0122] Reset the raw material ratio: Manually adjust the raw material ratio according to the risk of biofilm formation to ensure that it returns to the standard ratio state. It is also necessary to adjust conditions such as the storage temperature and humidity of the raw materials to reduce future biofilm formation;

[0123] Equipment suspension and thorough cleaning: Consider suspending equipment operation and performing thorough cleaning and disinfection according to the severity of the risk to ensure that there is no biofilm accumulation;

[0124] Comprehensive monitoring and real-time tracking: Strengthen the real-time monitoring of the equipment, collect more data and dynamically adjust the sterilization strategy to prevent the problem from deteriorating further.

[0125] Trigger condition: Automatically detect the high-risk level, trigger the emergency sterilization process, and perform operations such as ratio adjustment, equipment cleaning, and disinfection according to the specific situation.

[0126] The present invention covers any alternatives, modifications, equivalent methods, and solutions made within the spirit and scope of the present invention. For the public to have a thorough understanding of the present invention, specific details are described in detail in the following preferred embodiments of the present invention, and those skilled in the art can fully understand the present invention without these detailed descriptions. In addition, well-known methods, processes, procedures, components, and circuits are not described in detail to avoid unnecessary confusion to the essence of the present invention.

[0127] The above are only the preferred embodiments of the present invention. It should be noted that for those of ordinary skill in the art, without departing from the principle of the present invention, several improvements and refinements can be made, and these improvements and refinements should also be regarded as the protection scope of the present invention.

Claims

1. A background monitoring system based on a beverage machine, characterized in that: include: Data acquisition module: real-time acquisition of instantaneous flow values ​​of each raw material channel of the beverage machine, and calculation of the dynamic deviation coefficient D between the current ratio and the standard ratio; Compensation decision module: input the dynamic deviation coefficient into the compensation decision module, and output the flow compensation parameters of each channel; Control module: adjust the opening of the solenoid valve based on the flow compensation parameters; Uniformity extraction module: collects the adjusted vibration spectrum of the mixing bin, performs wavelet packet decomposition on the vibration spectrum, and extracts the energy value of the target frequency band as a mixing uniformity index; Risk prediction module: When the mixing uniformity index is lower than the uniformity threshold, the biofilm risk prediction model is triggered, the risk level of biofilm formation is output, and a real-time sterilization strategy is generated.

2. A background monitoring system based on a beverage machine according to claim 1, characterized in that: In the data acquisition module, an ultrasonic flow meter is installed at the outlet of each raw material channel to synchronously collect instantaneous flow data of syrup, water and gas channels, filter the instantaneous flow data to eliminate pulse noise interference, and calculate the dynamic deviation coefficient D based on the filtered measured flow value and the standard ratio flow value.

3. A background monitoring system based on a beverage machine according to claim 1, characterized in that: The flow compensation parameter is calculated as: δQ i =α i ×(Q i标准 -Q i实测 ), where δQ i is the flow compensation parameter of the i-th raw material channel, Q i标准 is the standard flow rate of the i-th raw material, Q i实测 is the actual flow rate of the i-th raw material, α i is the compensation coefficient.

4. A beverage machine-based background monitoring system according to claim 1, characterized in that: The control module calculates the solenoid valve opening adjustment value based on the PID controller. i Calculated by the following formula: Among them, K p is the proportional gain, which determines the influence of the current deviation, K i is the integral gain, which determines the cumulative effect of past errors, K d is the differential gain, which determines the influence of the error change rate, e i is the flow deviation, is the integral part of the flow deviation, representing the accumulated deviation in the past. It is the differential part of the flow deviation and represents the rate of change of the deviation.

5. A beverage machine-based background monitoring system according to claim 1, characterized in that: The uniformity extraction module specifically includes: Vibration spectrum acquisition unit: The vibration signal of the mixing bin is collected in real time through the vibration sensor installed on the mixing bin, and the vibration signal is converted into digital data through the signal conditioning circuit. The sampling frequency is set to fs, and fs is 10kHz; Wavelet packet decomposition unit: The collected vibration signal is decomposed by wavelet packet transform. The wavelet packet transform decomposition decomposes the signal into multiple frequency bands and extracts the signal features in the target frequency band. The target frequency band is set to 2-5kHz. The wavelet packet decomposition process decomposes the vibration signal into multiple frequency bands, each of which represents a frequency range. Frequency band energy value extraction unit: After wavelet packet decomposition, the wavelet packet coefficients of the target frequency band 2-5kHz are selected for energy calculation, the target frequency band is set, and the frequency band energy value E of each frequency band is calculated. 2-5kHz Perform calculations; Mixed uniformity index extraction unit: extract the frequency band energy value E of the target frequency band 2-5kHz 2-5kHz It is used as an indicator of mixing uniformity, and the energy value reflects the uniformity of the material in the mixing bin.

6. A beverage machine-based background monitoring system according to claim 5, characterized in that: In the frequency band energy value extraction unit, the frequency bands f1=2kHz and f2=5kHz are set, and the frequency band energy value of each frequency band is calculated: Among them, E 2-5kHz is the total energy value in the 2-5kHz frequency band, c jk is the wavelet packet coefficient of the jth layer and the kth frequency band, f j is the frequency band corresponding to frequency j.

7. A beverage machine-based background monitoring system according to claim 1, characterized in that: The uniformity threshold is set based on the uniformity reference value. When the mixing uniformity is less than 0.7× the uniformity reference value, it is determined as uneven mixing and the biofilm risk prediction model is automatically triggered.

8. A beverage machine-based background monitoring system according to claim 5, characterized in that: The biofilm risk prediction model includes data input and preprocessing, risk assessment and output. The biofilm risk prediction model predicts the risk of biofilm formation based on real-time mixing uniformity value, dynamic deviation coefficient and historical data. The biofilm risk prediction model outputs the risk level of biofilm formation, including three levels: high, medium and low.

9. A background monitoring system based on a beverage machine according to claim 8, characterized in that: The data input and preprocessing include the mixing uniformity index E 2-5kHz , dynamic deviation coefficient D, ambient temperature and humidity data and historical data, wherein the historical data includes historical dynamic deviation coefficient, mixing uniformity history and historical ambient temperature and humidity data; Standardized mixing uniformity value E′: Define the historical environmental impact factor η based on environmental temperature and humidity data env : in and They are the temperature and humidity benchmark values ​​in historical data respectively; The risk assessment includes setting a comprehensive assessment function R risk , comprehensive evaluation function R risk Output the biofilm risk value, which includes the following factors: Mixing uniformity factor: the greater the unevenness, the higher the biofilm risk; Historical dynamic deviation factor: The greater the ratio deviation, the higher the biofilm risk; Environmental temperature and humidity factors: The more suitable the temperature and humidity are for biofilm growth, the higher the risk; The comprehensive evaluation function is expressed as: R risk =α1·E′+α2·D mix +α3·η env , where α1, α2, α3 are weight coefficients; According to R risk The values ​​are divided into three risk levels: R risk <0.3 low risk; 0.3≤R risk <0.6 medium risk; R risk ≥0.6High risk.

10. A beverage machine-based background monitoring system according to claim 9, characterized in that: The real-time sterilization strategy includes: Low risk: No sterilization operation is performed, but regular inspection is performed; Medium risk: Start mild sterilization and increase equipment monitoring to remind operators to make adjustments; High risk: Immediately start high-intensity sterilization and clean the equipment, and adjust the raw material ratio.

Images