A data analysis-based intelligent supervision method for food processing equipment

Abstract

The present application relates to the technical field of data processing, and more particularly to a food processing equipment intelligent supervision method based on data analysis, which, in the food processing process, when detecting that the equipment in the food processing process appears suspected abnormality, obtains a early warning time length basic value in combination with the recent equipment comprehensive temperature change characteristics and the processing time length; further analyzes from two core dimensions of the processing process stability and the food material itself thermodynamic characteristics, respectively calculates to obtain a process instability adjustment factor and a thermodynamic characteristic adjustment factor, optimizes and calibrates the early warning time length basic value by using the two adjustment factors, finally obtains an adaptive early warning time length, and performs abnormal supervision on the equipment, so that the early warning time length is more suitable for the actual production scene, so as to accurately adapt to the complex and changeable processing conditions, effectively improve the precision of the equipment early warning monitoring in the food processing process, and further ensure the food processing quality stability.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, and in particular to an intelligent monitoring method for food processing equipment based on data analysis. Background Technology

[0002] In the food processing industry, with the expansion of production scale and the increase in process complexity, the operational stability of food processing equipment has an increasingly significant impact on product quality, production safety, and production cost control. Food processing typically involves harsh conditions such as high temperature, high pressure, and high humidity, and the processing parameters vary greatly among different types of food. During food processing, important processing parameters such as temperature and pressure often experience short-term fluctuations. If temperature fluctuations exceed limits, abnormal pressure leaks, or transmission components jam, it will directly trigger a warning shutdown, causing the quality of the food being processed to be damaged due to process interruption. For example, insufficient or excessive pre-cooking of porridge can lead to poor taste, abnormal color, or even the scrapping of the entire batch, and may also cause safety accidents such as steam burns.

[0003] In existing technologies, early warnings for equipment anomalies during food processing are typically based on big data analytics. Specifically, a fixed warning duration is set, and when an abnormal trend is detected in the food processing process, the duration of this trend is recorded. If the duration exceeds the warning duration, an alarm is immediately triggered. While this method is easy to deploy and manage quickly and uniformly on-site, requiring no complex data analysis or algorithm support, the fixed warning duration still has significant drawbacks. Firstly, a single warning duration cannot adapt to the needs of cross-category processing. For example, the caramelization stage of roasted pears requires a response time of seconds, while the gelatinization stage of porridge can tolerate warnings of several minutes; a fixed value can easily lead to delayed or false alarms. Secondly, a fixed warning duration does not consider the stability fluctuations of the food processing process. When equipment ages, external interference occurs, or material batches change, the reliability of the warning decreases, and intervention time may be missed due to failure to adjust in time. Furthermore, a fixed warning duration lacks dynamic perception of thermodynamic critical points, such as failing to strengthen early warnings when approaching gelatinization / caramelization temperatures.

[0004] Therefore, how to dynamically adjust the warning duration, improve the efficiency of monitoring equipment malfunctions during food processing, and take timely and appropriate intervention measures has become an urgent problem to be solved. Summary of the Invention

[0005] In view of this, embodiments of the present invention provide a data analysis-based intelligent monitoring method for food processing equipment to address the problem of how to dynamically adjust the warning duration, improve the monitoring efficiency of equipment anomalies during food processing, and take timely and appropriate intervention measures.

[0006] This invention provides a data analysis-based intelligent monitoring method for food processing equipment, which includes the following steps: During food processing, if a suspected abnormal state is detected in the equipment at the current moment, the overall temperature of the equipment at each sampling moment is obtained by monitoring the temperature of the equipment at different preset locations. Based on the changes in the overall temperature of the equipment within the first preset historical period at the current moment, the processing time up to the current moment, and the standard processing time, the basic value of the warning duration at the current moment is obtained. Multi-source feature fusion analysis is performed on the overall temperature variation fluctuation of the equipment, external interference events, equipment temperature response degree, and temperature uniformity of the equipment at different preset locations within the second preset historical period including the current moment, to obtain the process instability adjustment factor at the current moment; Based on the equipment heat input and overall equipment temperature difference during the second preset historical period, as well as the current processing food material quality, processing food material temperature difference, and equipment temperature difference at different preset locations, the thermodynamic characteristics of the material at the current moment are comprehensively analyzed to obtain the thermodynamic characteristic adjustment factor at the current moment. The basic value of the warning duration is optimized and adjusted using the process instability adjustment factor and the thermodynamic property adjustment factor to obtain the adaptive warning duration at the current moment. Based on the adaptive warning duration, the equipment is monitored for abnormalities.

[0007] Preferably, the step of obtaining the basic value of the warning duration for the current moment based on the changes in the overall temperature of the equipment during a first preset historical period, the processing time up to the current moment, and the standard processing time includes: A linear fit is performed on the overall equipment temperature within the first preset historical period, including the current moment, to obtain the corresponding fitting slope, which is recorded as the temperature gradient at the current moment; the ratio between the processing time up to the current moment and the standard processing time is calculated, and the product between the absolute value of the temperature gradient and the ratio is normalized to obtain the warning sensitivity at the current moment; the difference between the constant 1 and the warning sensitivity is obtained, and the product between the difference and the preset maximum warning duration is used as the base value of the warning duration at the current moment.

[0008] Preferably, the method for obtaining the process instability adjustment factor at the current moment includes: The overall temperature of the equipment within a second preset historical period, including the current moment, is divided into multiple temperature subsequences according to a preset length. The temperature gradient corresponding to each temperature subsequence is obtained to form a temperature gradient sequence. Based on the numerical and directional differences of the temperature gradients in the temperature gradient sequence, a temperature change instability index is obtained. Based on the ambient temperature and humidity at each sampling moment within the second preset historical period, the frequency of external interference events is obtained; based on the overall equipment temperature at each sampling moment up to the current moment, the time required for the overall equipment temperature to reach the preset equipment temperature corresponding to the current food processing stage is obtained, and the time is normalized to obtain the equipment temperature response time. The temperature gradient of the device at each preset position at the current moment is obtained, the difference between the maximum temperature gradient and the minimum temperature gradient is obtained, and the difference is normalized to obtain the device temperature non-uniformity index. By combining the temperature change instability index, the frequency of external disturbance events, the equipment temperature response time, and the equipment temperature non-uniformity index, the process instability adjustment factor at the current moment is obtained.

[0009] Preferably, obtaining the temperature change instability index based on the numerical and directional differences of the temperature gradients in the temperature gradient sequence includes: Calculate the mean and standard deviation of the temperature gradient in the temperature gradient sequence, and normalize the product of the mean and standard deviation to obtain the degree of temperature change fluctuation. Obtain the maximum temperature gradient subsequence corresponding to the monotonic transformation of the temperature gradient in the temperature gradient sequence, calculate the length ratio between the maximum temperature gradient subsequence and the temperature gradient sequence, and normalize the reciprocal of the length ratio to obtain the degree of change in the temperature gradient direction. The product of the degree of temperature change fluctuation and the degree of change in the direction of the temperature gradient is used as an index of temperature change instability.

[0010] Preferably, the step of obtaining the frequency of external interference events based on the ambient temperature and humidity at each sampling time within the second preset historical period includes: The changes in ambient temperature and humidity at each sampling moment within the second preset historical period are obtained. If at least one of the changes in ambient temperature and humidity at any sampling moment is greater than the corresponding preset change threshold, then that sampling moment is recorded as a change moment. The number of change moments is counted within the second preset historical period, and the ratio between the number and the duration of the second preset historical period is normalized to obtain the frequency of external interference events.

[0011] Preferably, the step of combining the temperature change instability index, the frequency of external disturbance events, the equipment temperature response time, and the equipment temperature non-uniformity index to obtain the process instability adjustment factor at the current moment includes: The sum of the frequency of the external disturbance event, the temperature response time of the equipment, and the temperature non-uniformity index of the equipment is calculated. The product of the sum and the temperature change instability index is normalized to obtain the process instability adjustment factor at the current moment.

[0012] Preferably, the method for obtaining the thermodynamic property adjustment factor at the current moment includes: Based on the equipment heat input and overall temperature difference during the second preset historical period, and the current mass of the processed food material, calculate the specific heat capacity of the food material; based on the temperature of the equipment at different preset locations at the current moment, obtain the absolute value of the temperature difference between the equipment edge temperature and the equipment center temperature; calculate the temperature difference between the current overall equipment temperature and the preset critical temperature of the food material, and record it as the temperature difference degree of the processed food material. The specific heat capacity of the food material, the absolute value of the temperature difference, and the temperature difference of the processed food material are fused from multiple sources to obtain the thermodynamic property adjustment factor at the current moment.

[0013] Preferably, the step of fusing multi-source features of the specific heat capacity of the food material, the absolute value of the temperature difference, and the temperature difference of the processed food material to obtain the thermodynamic property adjustment factor at the current moment includes: The specific heat capacity of the food material, the absolute value of the temperature difference, and the temperature difference of the processed food material are normalized respectively. The normalized values ​​are accumulated to obtain the cumulative value. The cumulative value is then normalized to obtain the thermodynamic characteristic adjustment factor at the current moment.

[0014] Preferably, the step of optimizing the base value of the warning duration using the process instability adjustment factor and the thermodynamic property adjustment factor to obtain the adaptive warning duration at the current moment includes: The difference between constant 1 and the process instability adjustment factor is calculated as the first adjustment coefficient. The sum of constant 1 and the thermodynamic property adjustment factor is calculated as the second adjustment coefficient. The first multiplication value of the basic warning duration value and the first adjustment coefficient is calculated, and the second multiplication value of the basic warning duration value and the second adjustment coefficient is calculated. The average of the first multiplication value and the second multiplication value is used as the adaptive warning duration at the current moment.

[0015] Preferably, the step of monitoring device anomalies based on the adaptive warning duration includes: If the duration of a suspected abnormal state detected in the device at the current moment is greater than or equal to the adaptive warning duration, an alarm for device abnormality will be triggered immediately.

[0016] The beneficial effects of the embodiments of the present invention compared with the prior art are as follows: In this invention, when a suspected abnormality is detected in the equipment during food processing, a baseline value for the warning duration is first obtained by combining the recent comprehensive temperature change characteristics of the equipment with the processing time. This baseline value can effectively reduce abnormal response errors, avoid immediate false responses, and provide a reasonable benchmark for subsequent adjustments to the warning duration. Subsequently, further analysis is conducted from two core dimensions: the stability of the processing process and the thermodynamic properties of the food materials themselves. The process instability adjustment factor and the thermodynamic property adjustment factor are calculated respectively. The baseline value for the warning duration is optimized and calibrated using these two adjustment factors, ultimately resulting in an adaptive warning duration. This makes the warning duration more consistent with the actual production scenario, and this design can accurately adapt to complex and changing processing conditions, effectively improving the accuracy of equipment warning monitoring during food processing, thereby ensuring stable food processing quality. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is a flowchart of a data analysis-based intelligent monitoring method for food processing equipment provided in Embodiment 1 of the present invention. Detailed Implementation

[0019] Embodiments of this disclosure are described in detail below, with examples of these embodiments illustrated in the accompanying drawings. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain this disclosure, and should not be construed as limiting it.

[0020] It should be noted that the terms "first," "second," etc., used in this disclosure and the accompanying drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this disclosure described herein can be implemented in orders other than those illustrated or described herein. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this disclosure. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this disclosure.

[0021] To illustrate the technical solution of the present invention, specific embodiments are described below.

[0022] See Figure 1This is a flowchart of a data analysis-based intelligent monitoring method for food processing equipment provided in Embodiment 1 of the present invention. Figure 1 As shown, the method may include: Step S101: During food processing, if a suspected abnormal state is detected in the equipment at the current moment, the comprehensive temperature of the equipment at each sampling moment is obtained by monitoring the temperature of the equipment at different preset locations. Based on the changes in the comprehensive temperature of the equipment in the first preset historical period, the processing time up to the current moment, and the standard processing time, the basic value of the warning duration at the current moment is obtained.

[0023] In intelligent monitoring scenarios involving the processing of multiple food products, accurately setting the warning duration is crucial for ensuring product quality and production safety. Due to the complexity and variability of food processing, influenced by a variety of factors, a fixed warning duration is insufficient to meet actual needs. For example, the caramelization stage of roasted pears requires a response time of seconds, while the gelatinization stage of porridge can tolerate warnings of several minutes. Therefore, in this embodiment of the invention, multi-dimensional monitoring data during food processing is analyzed to dynamically obtain adaptive warning durations for abnormal monitoring of food processing equipment, enabling timely intervention and improving food processing quality.

[0024] Specifically, during food processing, the temperature of the processing equipment is collected at different preset locations. The average temperature at all preset locations at each sampling time is calculated and recorded as the overall equipment temperature at that sampling time. These preset locations are evenly distributed, including the equipment center, edge, upper part, and lower part. A temperature sensor is installed at each preset location, and each sensor's sampling frequency is set to 1 second. If the overall equipment temperature at the current moment is greater than or equal to a preset warning temperature, a suspected anomaly is considered to have occurred. By acquiring multi-dimensional historical monitoring data up to the current moment, the required warning duration for issuing a suspected anomaly warning is determined to avoid direct anomaly warnings causing processing shutdowns, which could affect food processing efficiency and food quality.

[0025] The temperature change characteristics and processing stage during food processing have a significant impact on the warning duration. The rate of recent temperature change, i.e., the temperature gradient, directly reflects the temperature change trend. A larger temperature gradient means a more drastic temperature change, requiring a shorter warning time, i.e., a shorter warning duration. For example, during the caramelization stage of roasted pears, the temperature rises sharply, and without timely warning, the pears are prone to burning, so the corresponding warning duration should be shorter. The proportion of processing time reflects the urgency of warnings at different processing stages. In the initial processing stage, even if the temperature gradient is large, it may be normal heating, and a shorter warning duration is not required. However, near the end stage, the temperature approaches the critical value, requiring a more sensitive warning, i.e., a shorter warning duration. Therefore, in this embodiment of the invention, the basic value of the warning duration is calculated by combining the recent comprehensive temperature of the equipment and the proportion of processing time at the current moment, providing a reasonable benchmark for subsequent adjustments.

[0026] The method for calculating the basic value of the warning duration is as follows: First, obtain the comprehensive temperature of the equipment at each sampling moment within 1 minute, including the current moment, that is, the first preset historical period is 1 minute. Use the least squares method to perform linear fitting on the comprehensive temperature of the equipment within 1 minute to obtain the corresponding fitting slope, which is recorded as the temperature gradient at the current moment. Then, calculate the ratio between the processing time up to the current moment and the standard processing time. Normalize the product between the absolute value of the temperature gradient and the ratio to obtain the warning sensitivity at the current moment. Obtain the difference between the constant 1 and the warning sensitivity, and use the product between the difference and the preset maximum warning duration as the basic value of the warning duration at the current moment.

[0027] In one implementation, the formula for calculating the baseline value of the warning duration at the current moment is: in, This represents the base value for the current warning duration. This indicates the preset maximum warning duration, which is set to 60 seconds in this embodiment of the invention. It can be adjusted according to actual conditions, and 1 represents a constant. Represents the normalization function. This represents the temperature gradient at the current moment, where | represents the absolute value sign. This indicates the processing time up to the current moment. This indicates the standard processing time for the corresponding processed food.

[0028] It should be noted that the absolute value of the temperature gradient... The larger the value, the more drastic the temperature change, and the shorter the corresponding warning time. The larger the value, the closer the current time is to the processing end time, and the more sensitive it is to early warning; therefore, The larger the value, the greater the temperature gradient at the current moment, the larger the proportion of processing time, and the more sensitive the required early warning. In other words, the higher the early warning sensitivity at the current moment, the smaller the base value of the required early warning duration.

[0029] Step S102: Perform multi-source feature fusion analysis on the overall temperature change fluctuation of the equipment, external interference events, equipment temperature response degree, and temperature uniformity of the equipment at different preset locations within the second preset historical period including the current moment, to obtain the process instability adjustment factor at the current moment.

[0030] In food processing environments, basic warning durations are insufficient to fully account for complex and variable processing conditions. Therefore, in this embodiment of the invention, further analysis is conducted from two core dimensions: processing stability and the thermodynamic properties of the food materials themselves. Corresponding adjustment factors are obtained for each dimension, and these two adjustment factors are used to optimize and calibrate the basic value of the warning duration. Regarding processing stability, several characteristics reflecting process stability are considered, such as the fluctuation of the overall equipment temperature and the persistence of the fluctuation direction. (1) The fluctuation rate of the overall temperature of the equipment can intuitively show the degree of temperature gradient change. The persistence of the fluctuation direction can measure the consistency of the temperature change trend. The greater the fluctuation rate, the more unstable the processing process. The worse the persistence of the fluctuation direction, the more frequently the temperature direction changes, and the more unstable the processing process. Therefore, the overall temperature of the equipment within the second preset historical period including the current moment is taken as the analysis object. The second preset historical period is 10 minutes. Then, 30 seconds is set as the preset length. The shorter the preset length, the more data, the longer the temperature gradient sequence obtained later, and the higher the calculation accuracy. The overall temperature of the equipment within the second preset historical period including the current moment is divided into multiple temperature subsequences according to the preset length. That is, a temperature subsequence contains multiple overall temperatures of the equipment within 30 seconds. According to the above method of obtaining the temperature gradient, the temperature gradient corresponding to each temperature subsequence is obtained to form a temperature gradient sequence. Then, based on the numerical difference and directional difference of the temperature gradient in the temperature gradient sequence, the temperature change instability index is obtained.

[0031] Specifically, based on the numerical and directional differences in the temperature gradients within the temperature gradient sequence, an index of temperature change instability is obtained, including: Calculate the mean and standard deviation of the temperature gradient in the temperature gradient sequence, and normalize the product of the mean and standard deviation to obtain the degree of temperature change fluctuation. Obtain the maximum temperature gradient subsequence corresponding to the monotonic transformation of the temperature gradient in the temperature gradient sequence, calculate the length ratio between the maximum temperature gradient subsequence and the temperature gradient sequence, and normalize the reciprocal of the length ratio to obtain the degree of change in the temperature gradient direction. Use the product of the degree of temperature change fluctuation and the degree of change in the temperature gradient direction as an index of temperature change instability.

[0032] In one embodiment, the formula for calculating the temperature change instability index is: in, Indicator representing temperature instability. Represents the normalization function. This represents the standard deviation of the temperature gradient in a temperature gradient sequence. This represents the mean temperature gradient of the temperature gradient sequence. Indicates the length of the temperature gradient sequence. This represents the length of the maximum temperature gradient subsequence corresponding to a monotonically changing (monotonically increasing or monotonically decreasing) temperature gradient in the temperature gradient sequence.

[0033] (2) Based on the ambient temperature and humidity sensors installed in the processing workshop, the ambient temperature and humidity at each sampling time are collected at a sampling frequency of one minute to characterize external factors. When events such as opening doors or starting and stopping the air conditioner interfere, the ambient temperature and humidity will change significantly, thereby affecting the stability of the processing process. For example, opening doors will cause the temperature in the processing workshop to drop, which will affect the temperature of the processing equipment and cause temperature fluctuations in the processing equipment. It can be seen that the higher the frequency of external interference, the worse the stability of the processing process. Therefore, in this embodiment of the invention, the frequency of external interference events is obtained based on the ambient temperature and humidity at each sampling time within the second preset historical period: The changes in ambient temperature and ambient humidity at each sampling moment within the second preset historical time period are obtained. The change refers to the absolute value of the difference between the collected data at the previous sampling moment and the next sampling moment. If at least one of the changes in ambient temperature and ambient humidity at any sampling moment is greater than the corresponding preset change threshold, then that sampling moment is recorded as the change moment. The ambient temperature change threshold is preferably set to 0.5 degrees Celsius and the ambient humidity change threshold is 3%. There are no restrictions here, and the settings can be adjusted according to the accuracy requirements of the implementation scenario.

[0034] Within the second preset historical period, the number of changing moments is counted, and the ratio between the number and the duration of the second preset historical period is normalized to obtain the frequency of external interference events. The formula for calculating the frequency of external interference events is as follows: in, Indicates the frequency of external interference events. Represents the normalization function. The number of change moments is used to characterize the number of times a disturbance event occurs. One change moment corresponds to one disturbance event. S represents the second preset historical time period.

[0035] (3) The equipment response level reflects the stability of the equipment status. The longer the response time, the more likely there is an abnormality in the equipment, which affects the stability of the processing process. Therefore, based on the comprehensive temperature of the equipment at each sampling time up to the current time, the time required for the comprehensive temperature of the equipment to reach the preset temperature of the equipment corresponding to the current food processing stage is obtained. The duration is normalized to obtain the device temperature response time. The preset temperature of the equipment corresponding to the current food processing stage is obtained from the equipment control system, and each processing stage corresponds to a preset temperature.

[0036] (4) The temperature difference monitored by the equipment at different preset positions can reflect the heating uniformity. The greater the difference, the more uneven the temperature distribution of the equipment and the less stable the processing. Therefore, according to the above method for obtaining the temperature gradient at the current moment, the temperature gradient of the equipment at each of the preset positions at the current moment is obtained to obtain the maximum temperature gradient. and minimum temperature gradient The difference between them is normalized to obtain the equipment temperature non-uniformity index. .

[0037] (5) Combining the temperature change instability index, the frequency of external disturbance events, the equipment temperature response time, and the equipment temperature non-uniformity index, the process instability adjustment factor at the current moment is obtained. The specific method is as follows: The sum of the frequency of the external disturbance event, the temperature response time of the equipment, and the temperature non-uniformity index of the equipment is calculated. The product of this sum and the temperature change instability index is then normalized to obtain the process instability adjustment factor at the current moment. The formula for calculating the process instability adjustment factor at the current moment is: in, This represents the process instability adjustment factor at the current moment.

[0038] It should be noted that the larger B1 is, the more drastic the temperature gradient change, the worse the continuity of the temperature gradient direction, and the worse the stability of the corresponding processing. The larger the value, the higher the frequency of recent external disturbances, the slower the equipment response speed, and the worse the heating uniformity of the equipment at the current moment, resulting in poorer process stability. Therefore, the process instability adjustment factor at the current moment... The larger the value, the better.

[0039] Step S103: Based on the equipment heat input and overall equipment temperature difference in the second preset historical period, as well as the current processing food material quality, processing food material temperature difference, and equipment temperature difference at different preset locations, a comprehensive analysis of the material thermodynamic characteristics at the current moment is performed to obtain the thermodynamic characteristic adjustment factor at the current moment.

[0040] Process instability adjustment factor Focusing on various factors affecting stability during processing, this invention analyzes characteristics such as temperature gradient, external disturbances, and equipment response to provide a basis for judging the stability of the process. However, to accurately set the warning duration, considerations of process stability alone are far from sufficient; the thermodynamic properties of the material itself are equally important. Therefore, this invention also utilizes multiple features that reflect the thermodynamic properties of food materials. First, specific heat capacity reflects a material's ability to absorb heat. Materials with high specific heat capacity experience slower temperature changes, allowing for a longer warning time. Therefore, a heating power sensor installed at the power supply of the heating equipment collects heating power data every minute as the equipment's heat input at each sampling moment. This data is then used to obtain the equipment's heat input at each sampling moment within a second preset historical period. Based on the equipment's heat input and overall temperature difference within the second preset historical period, as well as the current mass of the processed food material, the specific heat capacity of the food material is calculated. ,in, This represents the total heat input of all equipment within the second preset historical time period, and M represents the mass of the processed food material at the current moment, which can be collected by a weighing sensor installed at the material feeding point. This represents the absolute value of the difference between the current overall temperature of the equipment and the overall temperature of the equipment at the first sampling time within the second preset historical period. c represents a constant with a value of 0.01. It should be noted that the calculation of specific heat capacity is an existing technique and will not be elaborated upon here.

[0041] Then, thermal diffusion reflects the rate at which heat is transferred from the surface to the center. For large materials, thermal diffusion is slow, and the temperature change at the center lags, affecting the judgment of thermodynamic properties. Therefore, based on the temperature of the equipment at different preset locations at the current moment, the absolute value of the temperature difference between the equipment edge temperature and the equipment center temperature is obtained. Specifically, the average temperature between the equipment edge, the upper part of the equipment, and the lower part of the equipment at the current moment is calculated and recorded as the equipment edge temperature. The temperature at the center of the equipment at the current moment is recorded as the equipment center temperature. The absolute value of the temperature difference between the edge temperature and the center temperature of the equipment is then obtained. R2 is used to characterize the thermal diffusion rate of materials at the current moment. The larger the value of R2, the slower the diffusion rate. The warning time can be appropriately extended to prevent false alarms.

[0042] Secondly, the critical temperature of the material is obtained from literature or historical data, and then the overall temperature of the equipment at the current moment is calculated. With respect to the preset critical temperature of food materials The temperature difference between them is denoted as the temperature difference degree of the processed food materials. The greater the temperature difference in the processed food materials, the less likely they are currently in the critical temperature zone, and the warning period can be appropriately extended. It should be noted that the overall equipment temperature can also be considered as the food material temperature.

[0043] Finally, the specific heat capacity, absolute value of temperature difference, and temperature difference of processed food materials are fused using multi-source features to obtain the thermodynamic property adjustment factor at the current moment: the specific heat capacity, absolute value of temperature difference, and temperature difference of processed food materials are normalized respectively, the normalized values ​​are accumulated to obtain the accumulated value, and the accumulated value is normalized to obtain the thermodynamic property adjustment factor at the current moment.

[0044] In one embodiment, the formula for calculating the thermodynamic property adjustment factor at the current moment is: in, This represents the adjustment factor for thermodynamic properties at the current moment.

[0045] Step S104: Optimize and adjust the basic value of the warning duration using the process instability adjustment factor and the thermodynamic property adjustment factor to obtain the adaptive warning duration at the current moment. Based on the adaptive warning duration, monitor the equipment for abnormalities.

[0046] The process instability adjustment factor mentioned above takes into account the dynamic changes in the processing process, while the thermodynamic property adjustment factor takes into account the thermodynamic properties of the food materials themselves. The more unstable the processing process, the less reliable the detection result of the equipment being in a suspected abnormal state at the current moment. The warning time should be shortened to reduce the risk of false alarms. Food materials with high thermal inertia and slow reaction require a longer warning time. Therefore, this embodiment of the invention approaches the issue from two different dimensions, using the process instability adjustment factor and the thermodynamic property adjustment factor to optimize and adjust the basic value of the warning time. By integrating multiple factors, the warning time is made more in line with the actual food processing situation.

[0047] The adaptive warning duration is obtained by optimizing the basic value of the warning duration using process instability adjustment factors and thermodynamic property adjustment factors. This includes: calculating the difference between constant 1 and the process instability adjustment factor as a first adjustment coefficient; calculating the sum of constant 1 and the thermodynamic property adjustment factor as a second adjustment coefficient; calculating the first multiplication value of the basic warning duration value and the first adjustment coefficient; calculating the second multiplication value of the basic warning duration value and the second adjustment coefficient; and taking the average of the first multiplication value and the second multiplication value as the adaptive warning duration at the current moment.

[0048] In one implementation, the formula for calculating the adaptive warning duration at the current moment is: in, This indicates the adaptive warning duration at the current moment. This represents the base value for the current warning duration. This represents the process instability adjustment factor at the current moment. This represents the thermodynamic property adjustment factor at the current moment, where 1 represents a constant.

[0049] It should be noted that, This indicates that the base value of the warning duration is adjusted according to the process instability adjustment factor. The larger the process instability adjustment factor, the more unstable the processing process, and the shorter the warning duration should be. The corresponding adaptive warning duration should be shorter. This indicates that the base value of the warning duration is adjusted based on the thermodynamic characteristic adjustment factor. The larger the thermodynamic characteristic adjustment factor, the more relaxed the warning duration can be. The corresponding adaptive warning duration should be longer to prevent false alarms.

[0050] Thus, the adaptive warning duration for the current moment is obtained. Based on the adaptive warning duration, the device is monitored. Specifically, after detecting a suspected abnormality in the device at the current moment and obtaining the adaptive warning duration, the device continues to detect suspected abnormalities. If the duration of the suspected abnormal state is greater than or equal to the adaptive warning duration, an alarm is immediately triggered. Conversely, if the duration of the suspected abnormal state is less than the adaptive warning duration, it indicates that the device is not abnormal, and the device is monitored. When a suspected abnormality is detected again, the adaptive warning duration is re-obtained for device anomaly monitoring.

[0051] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included within the protection scope of the present invention.

Claims

1. A data-driven intelligent monitoring method for food processing equipment, characterized in that, The method includes: During food processing, if a suspected abnormal state is detected in the equipment at the current moment, the overall temperature of the equipment at each sampling moment is obtained by monitoring the temperature of the equipment at different preset locations. Based on the changes in the overall temperature of the equipment within the first preset historical period at the current moment, the processing time up to the current moment, and the standard processing time, the basic value of the warning duration at the current moment is obtained. Multi-source feature fusion analysis is performed on the overall temperature variation fluctuation of the equipment, external interference events, equipment temperature response degree, and temperature uniformity of the equipment at different preset locations within the second preset historical period including the current moment, to obtain the process instability adjustment factor at the current moment; Based on the equipment heat input and overall equipment temperature difference during the second preset historical period, as well as the current processing food material quality, processing food material temperature difference, and equipment temperature difference at different preset locations, the thermodynamic characteristics of the material at the current moment are comprehensively analyzed to obtain the thermodynamic characteristic adjustment factor at the current moment. The basic value of the warning duration is optimized and adjusted using the process instability adjustment factor and the thermodynamic property adjustment factor to obtain the adaptive warning duration at the current moment. Based on the adaptive warning duration, the equipment is monitored for abnormalities.

2. The intelligent monitoring method for food processing equipment based on data analysis according to claim 1, characterized in that, The step of obtaining the baseline value of the warning duration for the current moment based on the changes in the overall temperature of the equipment within a first preset historical period, the processing time up to the current moment, and the standard processing time includes: A linear fit is performed on the overall equipment temperature within the first preset historical period, including the current moment, to obtain the corresponding fitting slope, which is recorded as the temperature gradient at the current moment; the ratio between the processing time up to the current moment and the standard processing time is calculated, and the product between the absolute value of the temperature gradient and the ratio is normalized to obtain the warning sensitivity at the current moment; the difference between the constant 1 and the warning sensitivity is obtained, and the product between the difference and the preset maximum warning duration is used as the base value of the warning duration at the current moment.

3. The intelligent monitoring method for food processing equipment based on data analysis according to claim 1, characterized in that, The method for obtaining the process instability adjustment factor at the current moment includes: The overall temperature of the equipment within a second preset historical period, including the current moment, is divided into multiple temperature subsequences according to a preset length. The temperature gradient corresponding to each temperature subsequence is obtained to form a temperature gradient sequence. Based on the numerical and directional differences of the temperature gradients in the temperature gradient sequence, a temperature change instability index is obtained. Based on the ambient temperature and humidity at each sampling moment within the second preset historical period, the frequency of external interference events is obtained; based on the overall equipment temperature at each sampling moment up to the current moment, the time required for the overall equipment temperature to reach the preset equipment temperature corresponding to the current food processing stage is obtained, and the time is normalized to obtain the equipment temperature response time. The temperature gradient of the device at each preset position at the current moment is obtained, the difference between the maximum temperature gradient and the minimum temperature gradient is obtained, and the difference is normalized to obtain the device temperature non-uniformity index. By combining the temperature change instability index, the frequency of external disturbance events, the equipment temperature response time, and the equipment temperature non-uniformity index, the process instability adjustment factor at the current moment is obtained.

4. The intelligent monitoring method for food processing equipment based on data analysis according to claim 3, characterized in that, The step of obtaining a temperature change instability index based on the numerical and directional differences of the temperature gradients in the temperature gradient sequence includes: Calculate the mean and standard deviation of the temperature gradient in the temperature gradient sequence, and normalize the product of the mean and standard deviation to obtain the degree of temperature change fluctuation. Obtain the maximum temperature gradient subsequence corresponding to the monotonic transformation of the temperature gradient in the temperature gradient sequence, calculate the length ratio between the maximum temperature gradient subsequence and the temperature gradient sequence, and normalize the reciprocal of the length ratio to obtain the degree of change in the temperature gradient direction. The product of the degree of temperature change fluctuation and the degree of change in the direction of the temperature gradient is used as an index of temperature change instability.

5. The intelligent monitoring method for food processing equipment based on data analysis according to claim 3, characterized in that, The step of obtaining the frequency of external interference events based on the ambient temperature and humidity at each sampling time within the second preset historical period includes: The changes in ambient temperature and humidity at each sampling moment within the second preset historical period are obtained. If at least one of the changes in ambient temperature and humidity at any sampling moment is greater than the corresponding preset change threshold, then that sampling moment is recorded as a change moment. The number of change moments is counted within the second preset historical period, and the ratio between the number and the duration of the second preset historical period is normalized to obtain the frequency of external interference events.

6. The intelligent monitoring method for food processing equipment based on data analysis according to claim 3, characterized in that, The process instability adjustment factor for the current moment is obtained by combining the temperature change instability index, the frequency of external disturbance events, the equipment temperature response time, and the equipment temperature non-uniformity index, including: The sum of the frequency of the external disturbance event, the temperature response time of the equipment, and the temperature non-uniformity index of the equipment is calculated. The product of the sum and the temperature change instability index is normalized to obtain the process instability adjustment factor at the current moment.

7. The intelligent monitoring method for food processing equipment based on data analysis according to claim 1, characterized in that, The method for obtaining the thermodynamic property adjustment factor at the current moment includes: Based on the equipment heat input and overall temperature difference during the second preset historical period, and the current mass of the processed food material, calculate the specific heat capacity of the food material; based on the temperature of the equipment at different preset locations at the current moment, obtain the absolute value of the temperature difference between the equipment edge temperature and the equipment center temperature; calculate the temperature difference between the current overall equipment temperature and the preset critical temperature of the food material, and record it as the temperature difference degree of the processed food material. The specific heat capacity of the food material, the absolute value of the temperature difference, and the temperature difference of the processed food material are fused from multiple sources to obtain the thermodynamic property adjustment factor at the current moment.

8. The intelligent monitoring method for food processing equipment based on data analysis according to claim 7, characterized in that, The process of fusing multiple features—specific heat capacity of the food material, absolute value of the temperature difference, and temperature difference of the processed food material—to obtain the thermodynamic property adjustment factor for the current moment includes: The specific heat capacity of the food material, the absolute value of the temperature difference, and the temperature difference of the processed food material are normalized respectively. The normalized values ​​are accumulated to obtain the cumulative value. The cumulative value is then normalized to obtain the thermodynamic characteristic adjustment factor at the current moment.

9. The intelligent monitoring method for food processing equipment based on data analysis according to claim 1, characterized in that, The optimization adjustment of the basic value of the warning duration using the process instability adjustment factor and the thermodynamic property adjustment factor to obtain the adaptive warning duration at the current moment includes: The difference between constant 1 and the process instability adjustment factor is calculated as the first adjustment coefficient. The sum of constant 1 and the thermodynamic property adjustment factor is calculated as the second adjustment coefficient. The first multiplication value of the basic warning duration value and the first adjustment coefficient is calculated, and the second multiplication value of the basic warning duration value and the second adjustment coefficient is calculated. The average of the first multiplication value and the second multiplication value is used as the adaptive warning duration at the current moment.

10. The intelligent monitoring method for food processing equipment based on data analysis according to claim 1, characterized in that, The step of monitoring device anomalies based on the adaptive warning duration includes: If the duration of a suspected abnormal state detected in the device at the current moment is greater than or equal to the adaptive warning duration, an alarm for device abnormality will be triggered immediately.

Images