Health food production monitoring method and system based on big data

Abstract

The invention relates to the technical field of production data processing, in particular to a health food production monitoring method and system based on big data, and the method comprises the following steps: obtaining the temperature, humidity, air pressure and air velocity of a health food production workshop, collecting continuous sampling data, calculating a change rate, recording a rate time sequence, and generating a production environment trend change value. According to the method, the environment parameters are continuously collected, the change rate is calculated, the trend change value is constructed, based on the data offset condition, the over-threshold parameters are screened, the offset rate is calculated, and the environment calibration coefficient is adjusted in combination with the current period data, so that dynamic correction is achieved, hysteresis of single threshold regulation and control is avoided, and interaction among the parameters is analyzed; according to the scheme, the error offset condition is calculated, the adjustment amplitude is optimized, linkage errors caused by single adjustment are reduced, periodic fluctuation characteristics are extracted through long-term trend prediction, the perspectiveness of environment regulation and control is enhanced, the real-time response capacity to environment changes is improved, and dynamic balance among parameters is optimized.

Description

Technical Field

[0001] The present invention relates to the technical field of production data processing, and in particular, to a health food production monitoring method and system based on big data. Background Art

[0002] The technical field of production data processing includes the collection, transmission, storage, analysis, and application of various types of data in the production process to support production management, quality control, equipment maintenance, and supply chain optimization. In this technical field, the acquisition of production data usually relies on sensors, automation equipment, and industrial Internet technologies to achieve real-time monitoring of the production environment, material status, and process parameters. Subsequently, the data is transmitted to the data processing center through wired or wireless communication methods and is classified, screened, and converted after storage to adapt to different data analysis requirements. Production data processing also includes calculation methods based on rules or models to identify anomalies, optimize process parameters, or provide decision-making support. The development of this technical field has promoted the wide application of intelligent manufacturing, refined management, and automation control.

[0003] Among them, the health food production monitoring method based on big data refers to collecting, analyzing, and monitoring data such as raw material feeding, ratio control, processing technology, environmental parameters, and finished product inspection involved in the production process to ensure production quality and safety. This method usually uses data acquisition terminals to obtain key parameters such as equipment operating status, temperature, humidity, pressure, and time series, and uploads them to the calculation platform through a data transmission system. The calculation platform uses statistical calculations to perform trend analysis and anomaly identification on the collected data, and combines association analysis to provide adjustment suggestions for key process parameters. In addition, clustering calculations of historical data can be used to identify potential production risks, regression calculations can be used to predict the product quality fluctuation range, and logical reasoning can be used to determine production anomalies. The entire monitoring method covers links such as data acquisition, transmission, storage, calculation, analysis, and feedback to ensure that the production process meets the established standards.

[0004] The prior art relies on static threshold setting, lacks the calculation of the environmental change rate, resulting in a lag in response and being unable to accurately predict the parameter deviation trend. It lacks the analysis of the interaction between parameters, and adjusting a single parameter may cause chain fluctuations, affecting production stability. It does not establish a dynamic calibration mechanism based on trend data, and the regulation and adjustment rely on historical data, with insufficient adaptability and difficulty in accurately correcting sudden deviations. Long-term data analysis is only based on statistical calculations, fails to extract periodic fluctuation characteristics, is difficult to identify potential trend changes in advance, reduces the foresight of environmental regulation, and affects product quality stability and production consistency. Summary of the Invention

[0005] The object of the present invention is to solve the disadvantages existing in the prior art, and a health food production monitoring method and system based on big data are proposed.

[0006] To achieve the above object, the present invention adopts the following technical solutions: A health food production monitoring method based on big data, comprising the following steps:

[0007] S1: Obtain the temperature, humidity, air pressure, and air flow rate in the health food production workshop, collect continuous sampling data, calculate the change rate, record the rate time series, and generate a production environment trend change value;

[0008] S2: Based on the production environment trend change value, detect the change amplitude within a continuous target period of temperature, humidity, air pressure, and air flow rate, judge the data fluctuation before and after the mutation, calculate the deviation of the current data from the set environmental parameter offset threshold, and generate an environmental parameter offset prediction value;

[0009] S3: Based on the environmental parameter offset prediction value, screen the parameters with an offset amount exceeding the environmental parameter offset threshold, calculate the offset rate, call the environmental data of the current period, adjust the production environment calibration coefficient, and generate a dynamic calibration correction value;

[0010] S4: Based on the dynamic calibration correction value, analyze the trends of temperature, humidity, air pressure, and air flow rate, screen the parameter pairs with opposite directions, calculate the situation where the errors cancel each other out, and correct and generate an optimized adjustment value of environmental parameters;

[0011] S5: Based on the optimized adjustment value of environmental parameters, regulate the production workshop environment, collect long-term sampling environmental data, calculate the periodic fluctuations of temperature, humidity, air pressure, and air flow rate, extract long-term trend characteristics, and generate a long-term trend prediction value of the production environment.

[0012] As a further solution of the present invention, the production environment trend change value includes a temperature change rate, a humidity change rate, an air pressure change rate, and an air flow rate change rate. The environmental parameter offset prediction value includes a temperature offset prediction value, a humidity offset prediction value, an air pressure offset prediction value, and an air flow rate offset prediction value. The parameters with an offset amount exceeding the environmental parameter offset threshold are specifically parameters exceeding the temperature offset threshold, parameters exceeding the humidity offset threshold, parameters exceeding the air pressure offset threshold, and parameters exceeding the air flow rate offset threshold. The dynamic calibration correction value includes a temperature correction value, a humidity correction value, an air pressure correction value, and an air flow rate correction value. The optimized adjustment value of environmental parameters includes a temperature optimized adjustment value, a humidity optimized adjustment value, an air pressure optimized adjustment value, and an air flow rate optimized adjustment value. The long-term trend prediction value of the production environment includes a temperature long-term trend prediction value, a humidity long-term trend prediction value, an air pressure long-term trend prediction value, and an air flow rate long-term trend prediction value.

[0013] As a further solution of the present invention, the steps for obtaining the production environment trend change value are specifically as follows:

[0014] S101: Obtain the temperature, humidity, air pressure, and air flow velocity in the health food production workshop, collect environmental data within multiple consecutive sampling periods, record the data change values of adjacent sampling points, calculate the temperature change rate, humidity change rate, air pressure change rate, and air flow velocity change rate based on the continuous data, and obtain the environmental parameter change rate sequence;

[0015] S102: Call the environmental parameter change rate sequence, calculate the rate difference between adjacent multiple sampling points, and use the formula:

[0016]

[0017] Calculate the rate transition value sequence, and screen to obtain the key rate transition interval;

[0018] Among them, ΔS i1 represents the rate transition value sequence, V i1 represents the environmental parameter change rate of the i1-th sampling point, and T i1 represents the time stamp of the i1-th sampling point;

[0019] S103: Call the key rate transition interval, combine the rate change trends of multiple groups of environmental parameters, calculate the trend change rate, and obtain the production environment trend change value.

[0020] As a further solution of the present invention, the steps for obtaining the environmental parameter offset prediction value are specifically as follows:

[0021] S201: Based on the production environment trend change value, detect the change amplitude of the temperature, humidity, air pressure, and air flow velocity within consecutive n sampling periods, calculate the change rate of multiple parameters, and compare with the set threshold to determine whether a mutation occurs, and obtain the environmental parameter mutation determination result;

[0022] S202: Call the environmental parameter mutation determination result, calculate the data fluctuation amplitude of multiple sampling points before and after the mutation point, determine whether the inflection point belongs to the convergent or divergent type, and use the formula:

[0023]

[0024] Calculate the environmental parameter fluctuation amplitude, and screen in combination with the inflection point change trend to obtain the environmental parameter inflection point type;

[0025] Among them, D i1 represents the environmental parameter fluctuation amplitude, X j1 represents the environmental parameter value of the j1-th sampling point, represents the environmental parameter mean value of the interval, Vj1 represents the change rate of the environmental parameters at the j1-th sampling point, represents the average change rate of the environmental parameters in the interval, A j1 represents the acceleration of the environmental parameters at the j1-th sampling point, represents the average acceleration of the environmental parameters in the interval, k1 represents the length of the time window under investigation, and n1 represents the number of sampling points;

[0026] S203: Invoke the inflection point type of the environmental parameters, set the environmental parameter offset threshold, calculate the deviation degree of the current environmental parameters relative to the mean value, extract the temperature offset rate, humidity offset rate, air pressure offset rate, and air flow velocity offset rate within multiple consecutive sampling periods, combine the trend direction to determine the offset state, and obtain the environmental parameter offset prediction value.

[0027] As a further solution of the present invention, the steps for obtaining the dynamic calibration correction value are specifically as follows:

[0028] S301: Based on the environmental parameter offset prediction value, screen the parameters whose offset amount exceeds the environmental parameter offset threshold, invoke the environmental data to calculate the offset rate and duration of multiple parameters, calculate the change rate on the time series, determine whether the parameters exceed the set range and screen the qualified parameters, record the offset duration, invoke the offset rate and duration, and obtain the environmental parameter offset characteristic value;

[0029] S302: Invoke the environmental parameter offset characteristic value, calculate the environmental data of the current period and the previous period, adjust the production environment calibration coefficient, and use the formula:

[0030]

[0031] Calculate the dynamic calibration correction value and apply it to the adjustment of the production environment parameters;

[0032] where, C t represents the production environment calibration coefficient of the current period, C t-1 represents the production environment calibration coefficient of the previous period, P i1 represents the offset value of the i1-th environmental parameter, P th represents the environmental parameter offset threshold, T i1 represents the offset duration of the i1-th parameter, V i1 represents the offset rate of the i1-th parameter, A avg represents the average value of the environmental parameter acceleration, and n1 represents the number of screened environmental parameters.

[0033] As a further solution of the present invention, the steps for obtaining the environmental parameter optimization adjustment value are specifically as follows:

[0034] S401: Analyze the changing trends of temperature, humidity, air pressure, and air velocity based on the dynamic calibration correction value, calculate the change rates of multiple parameters over time series respectively, obtain the change directions, determine whether the change directions of multiple parameters are consistent, screen the parameter pairs with opposite change directions, and obtain the set of parameter pairs with opposite trends;

[0035] S402: Call the set of parameter pairs with opposite trends, calculate the error cancellation situation between multiple parameter pairs, calculate the complementary effect of parameter pairs using the absolute error difference, and count the error cancellation degree of all parameter pairs to obtain the error cancellation rate distribution value;

[0036] S403: Call the error cancellation rate distribution value, for the parameter pairs with incomplete error cancellation, calculate the correction coefficient of the production environment using the minimum error adjustment method, and the correction calculation formula is as follows:

[0037]

[0038] Obtain the optimized adjustment value of environmental parameters through calculation;

[0039] where, C adj represents the calibrated production environment calibration coefficient, C orig represents the original production environment calibration coefficient, E i1 represents the absolute value of the error of the i1-th parameter pair with incomplete error cancellation, V j1 represents the j1-th component of the change rate of the environmental parameters participating in the adjustment, S k1 represents the k1-th component of the standard deviation of the uncalibrated parameter, n1 is the number of parameter pairs with incomplete error cancellation, m1 is the number of change rate components of all environmental parameters, and p1 is the number of standard deviation components of the uncalibrated parameter.

[0040] As a further solution of the present invention, the steps for obtaining the long-term trend prediction value of the production environment are specifically as follows:

[0041] S501: Based on the optimized adjustment value of the environmental parameters, call the long-term production batch environmental monitoring data, calculate the fluctuation amplitude and change pattern of temperature, humidity, air pressure, and air velocity in each production cycle, calculate the periodic change trend using moving average, and obtain the environmental parameter periodic fluctuation pattern;

[0042] S502: Call the environmental parameter periodic fluctuation pattern, calculate the change rate of the long-term trend of each parameter, calculate the long-term trend change value using linear regression, and determine whether the long-term fluctuation trend is stable based on the fluctuation amplitude to obtain the long-term environmental trend characteristics;

[0043] S503: Call the long-term environmental trend characteristics, for the parameters with unstable trends, calculate the optimal long-term calibration reference value using the formula:

[0044]

[0045] Calculate the long-term calibration reference value and obtain the long-term trend prediction value of the production environment through operation;

[0046] Among them, X new represents the adjusted long-term calibration reference value, X init represents the initial long-term calibration reference value, Y r1 represents the environmental trend data of the r1th production cycle, Z represents the average long-term environmental trend, A r1 represents the adjustment factor associated with the data of the r1th production cycle, B s1 represents the s1th component in the parameter fluctuation range, C u1 represents the u1th data in the environmental stability evaluation, D represents the stability normalization constant, q1 represents the total number of production cycles, t1 represents the total number of fluctuation components, and v1 represents the total number of stability data components.

[0047] A health food production monitoring system based on big data, the health food production monitoring system based on big data is used to execute the above-mentioned health food production monitoring method based on big data, and the system includes:

[0048] The environmental data acquisition module obtains the temperature, humidity, air pressure, and air flow velocity data of the production workshop, records the continuous sampling data and calculates the change rate, constructs a time series, calls the change rate data to calculate the short-term trend change value of the production environment, and obtains the production environment trend change value;

[0049] The environmental change analysis module, based on the production environment trend change value, detects the change range of temperature, humidity, air pressure, and air flow velocity within a continuous target cycle, calculates the data fluctuation range, screens the fluctuation mutation data, and calculates the change rate before and after the mutation, calls the set environmental parameter offset threshold, and calculates the offset amount between the current data and the set environmental parameter to obtain the environmental parameter offset prediction value;

[0050] The offset prediction and calibration module, based on the environmental parameter offset prediction value, screens the parameters whose offset amount exceeds the set environmental parameter offset threshold, calculates the parameter offset rate, calls the environmental data of the current cycle, establishes a dynamic correction parameter, and adjusts the production environment calibration coefficient to obtain a dynamic calibration correction value;

[0051] The parameter optimization and adjustment module, based on the dynamic calibration correction value, analyzes the change trends of temperature, humidity, air pressure, and air flow velocity, screens the parameter pairs with opposite trend directions, calculates the error cancellation situation, and adjusts the deviation parameters to obtain the environmental parameter optimization and adjustment value;

[0052] The long-term trend monitoring module adjusts and controls the environment of the production workshop based on the optimized adjustment value of the environmental parameters, collects long-term sampled environmental data, calculates the periodic fluctuations of temperature, humidity, air pressure, and air velocity, extracts long-term trend features, and obtains the long-term trend prediction value of the production environment.

[0053] Compared with the prior art, the advantages and positive effects of the present invention are as follows:

[0054] In the present invention, by continuously collecting environmental parameters and calculating the change rate, a trend change value is constructed to improve the perception ability of environmental fluctuations. Based on the data deviation situation, parameters exceeding the threshold are screened, the deviation rate is calculated, and the environmental calibration coefficient is adjusted in combination with the current cycle data to achieve dynamic correction and avoid the lag of single-threshold regulation. The interaction between parameters is analyzed, the error cancellation situation is calculated, the adjustment amplitude is optimized, and the chain error caused by single-item adjustment is reduced. The long-term trend prediction extracts periodic fluctuation characteristics and enhances the forward-looking nature of environmental regulation. This solution improves the real-time response ability to environmental changes, optimizes the dynamic balance between parameters, enhances the stability of the production environment, reduces the quality deviation caused by fluctuations, and improves production consistency. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] Figure 1 is a schematic diagram of the working process of the present invention;

[0056] Figure 2 is a flowchart of the steps for obtaining the trend change value of the production environment of the present invention;

[0057] Figure 3 is a flowchart of the steps for obtaining the offset prediction value of the environmental parameters of the present invention;

[0058] Figure 4 is a flowchart of the steps for obtaining the dynamic calibration correction value of the present invention;

[0059] Figure 5 is a flowchart of the steps for obtaining the optimized adjustment value of the environmental parameters of the present invention;

[0060] Figure 6 is a flowchart of the steps for obtaining the long-term trend prediction value of the production environment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0061] In order to make the objectives, technical solutions, and advantages of the present invention clearer and more understandable, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0062] In the description of the present invention, it should be understood that the orientation or positional relationship indicated by terms such as "length", "width", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc. is based on the orientation or positional relationship shown in the drawings. It is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation. Therefore, it should not be construed as a limitation to the present invention. In addition, in the description of the present invention, the meaning of "a plurality of" is two or more, unless otherwise specifically defined.

[0063] Embodiment 1

[0064] Please refer to Figure 1 , the present invention provides a technical solution: a method for monitoring the production of health foods based on big data, including the following steps:

[0065] S1: Obtain the temperature, humidity, air pressure, and air flow rate in the health food production workshop, collect the environmental data of multiple consecutive sampling periods, calculate the temperature change rate, humidity change rate, air pressure change rate, and air flow rate change rate, record the time series of multiple rates, calculate the rate difference between adjacent multiple sampling points, determine whether there is a rate transition, and generate a production environment trend change value;

[0066] S2: Based on the production environment trend change value, detect the change amplitude of temperature, humidity, air pressure, and air flow rate within consecutive n sampling periods, determine whether a mutation occurs, calculate the data fluctuation amplitude of multiple sampling points before and after the mutation point, determine whether the inflection point belongs to the convergent or divergent type, call the production batch environmental data, set the environmental parameter offset threshold, calculate the deviation degree of the current environmental parameter relative to the mean value, extract the temperature offset rate, humidity offset rate, air pressure offset rate, and air flow rate offset rate within multiple consecutive sampling periods, combine the trend direction to determine the offset state, and generate an environmental parameter offset prediction value;

[0067] S3: Based on the environmental parameter offset prediction value, screen the parameters whose offset amount exceeds the environmental parameter offset threshold, calculate the offset rate and duration, call the environmental data of the current period and the previous period, adjust the production environment calibration coefficient, and generate a dynamic calibration correction value;

[0068] S4: Based on the dynamic calibration correction value, analyze the change trends of temperature, humidity, air pressure, and air flow rate, determine whether the change directions of multiple parameters are consistent, screen the parameter pairs with opposite trends, calculate the situation where the errors cancel each other out. If the errors are not completely canceled, use the minimum error adjustment method to correct the production environment calibration coefficient, and generate an environmental parameter optimization adjustment value;

[0069] S5: Based on the optimized adjustment value of the environmental parameters, call the long-term production batch environmental monitoring data, calculate the periodic fluctuation patterns of temperature, humidity, air pressure, and air velocity, extract the long-term environmental trend characteristics, determine whether the long-term fluctuation trend is stable, adjust the long-term calibration reference value of the production environment, and generate the long-term trend prediction value of the production environment.

[0070] The production environment trend change value includes the temperature change rate, humidity change rate, air pressure change rate, and air velocity change rate. The environmental parameter offset prediction value includes the temperature offset prediction value, humidity offset prediction value, air pressure offset prediction value, and air velocity offset prediction value. The offset amount exceeding the environmental parameter offset threshold parameter is specifically the temperature offset threshold parameter, humidity offset threshold parameter, air pressure offset threshold parameter, and air velocity offset threshold parameter. The dynamic calibration correction value includes the temperature correction value, humidity correction value, air pressure correction value, and air velocity correction value. The environmental parameter optimized adjustment value includes the temperature optimized adjustment value, humidity optimized adjustment value, air pressure optimized adjustment value, and air velocity optimized adjustment value. The long-term trend prediction value of the production environment includes the temperature long-term trend prediction value, humidity long-term trend prediction value, air pressure long-term trend prediction value, and air velocity long-term trend prediction value.

[0071] Please refer to Figure 2 , and the specific steps for obtaining the production environment trend change value are as follows:

[0072] S101: Obtain the temperature, humidity, air pressure, and air velocity in the health food production workshop, collect environmental data within multiple consecutive sampling periods, record the data change values at adjacent sampling points, and calculate the temperature change rate, humidity change rate, air pressure change rate, and air velocity change rate based on the continuous data to obtain the environmental parameter change rate sequence.

[0073] To obtain the temperature, humidity, air pressure, and air velocity in a health food production workshop, real-time monitoring needs to be carried out through a sensor network deployed at multiple key points inside the workshop. These sensors need to perform periodic data collection. Assuming the sampling period is set to 10 seconds, each sensor records data once within this period. For example, the temperature sensor records the current ambient temperature value as 24.5 °C, the humidity sensor records the current humidity as 60.2% RH, the air pressure sensor records the current air pressure as 101.2 kPa, and the air velocity sensor records the current air velocity as 0.6 m / s. These data will be stored and compared with the data at the previous sampling point to calculate the change values of environmental parameters between adjacent sampling points. Among them, the calculation method of the temperature change rate is (current temperature - previous temperature) / sampling period. If the previous temperature is 24.3 °C, the temperature change rate is calculated as follows: (24.5 - 24.3) / 10 = 0.02 °C / s. Similarly, the calculation method of the humidity change rate is (current humidity - previous humidity) / sampling period. Assuming the previous humidity is 59.8% RH, the humidity change rate is calculated as (60.2 - 59.8) / 10 = 0.04% RH / s. The calculation method of the air pressure change rate is (current air pressure - previous air pressure) / sampling period. Assuming the previous air pressure is 101.1 kPa, the change rate is calculated as follows: (101.2 - 101.1) / 10 = 0.01 kPa / s. The calculation method of the air velocity change rate is (current air velocity - previous air velocity) / sampling period. If the previous air velocity is 0.55 m / s, the air velocity change rate is calculated as follows: (0.6 - 0.55) / 10 = 0.005 m / s. The above calculates the environmental parameter change rate sequence and continuously records it.

[0074] S102: Call the environmental parameter change rate sequence, calculate the rate difference between adjacent multiple sampling points, and use the formula:

[0075]

[0076] Calculate the rate jump value sequence and screen to obtain the key rate jump interval;

[0077] Among them, ΔS i1 represents the rate jump value sequence, V i1 represents the environmental parameter change rate of the i1-th sampling point, and T i1 represents the timestamp of the i1-th sampling point;

[0078] Call the previously obtained environmental parameter change rate sequence, calculate the rate change between consecutive multiple sampling points, calculate the rate jump value sequence, and use the formula to calculate the rate change. Assume the data of a certain consecutive three sampling points is as follows:

[0079] Table 1 Rate Change Table of Environmental Parameters at Sampling Points

[0080]

[0081] As shown in Table 1, the rate of change of environmental parameters at different sampling points is used to calculate the rate transition value.

[0082] Calculate the adjacent rate changes. For example, the calculation of the temperature rate transition value is as follows:

[0083]

[0084] Similarly, the calculation of the humidity rate transition value is as follows:

[0085]

[0086] The calculation of the air pressure rate transition value is as follows:

[0087]

[0088] The calculation of the air flow rate transition value is as follows:

[0089]

[0090] The calculated rate transition value sequence is used to screen the key intervals of rate transition. If the rate transition value of a certain parameter exceeds the set threshold (for example, 0.002 °C / s² for temperature, 0.002 %RH / s² for humidity, 0.001 kPa / s² for air pressure, 0.0003 m / s² for air flow rate), then it is determined that this time period is a key interval. At this time, the air pressure rate transition value within the 20 - 30 s time period is less than the set threshold, so it is not included in the key interval, while the temperature rate transition value is close to the threshold and needs further confirmation.

[0091] S103: Call the key interval of the rate transition, and combine with the rate change trends of multiple groups of environmental parameters to calculate the trend change rate, and obtain the production environment trend change value.

[0092] Call the key interval of the rate transition obtained by screening, and combine with the rate change trends of environmental parameters to calculate the trend change rate. Extract the significant change intervals from the rate transition values of multiple environmental parameters, and calculate the environmental parameter trend change value. The calculation method is that the trend change rate λ is calculated as follows:

[0093]

[0094] Among them, λ represents the trend change rate, and ΔS j represents the rate transition value of each environmental parameter, and n is the number of sampling points in the selected key interval. Assume that the data within the selected key interval is as follows:

[0095] Table 2 Data Table of Key Intervals for Rate Transitions

[0096]

[0097] As shown in Table 2, the rate transition values of different environmental parameters within the key intervals can be used to calculate the trend change rate.

[0098] Calculate the trend change rate:

[0099]

[0100] According to the calculation results, if the trend change rate exceeds the set threshold, it is determined that there is a significant environmental trend change in this key interval. In this example, the humidity trend change rate λ 湿度 = 0.0021%RH / s² approaches the set threshold, indicating that the humidity parameter in this key interval has changed.

[0101] Please refer to Figure 3 for the specific steps to obtain the predicted value of the environmental parameter offset:

[0102] S201: Based on the trend change value of the production environment, detect the change amplitude of temperature, humidity, air pressure, and air velocity within n consecutive sampling periods, calculate the change rate of multiple parameters, and compare with the set threshold to determine whether a mutation has occurred, obtaining the mutation determination result of the environmental parameter;

[0103] Based on the trend change value of the production environment, to obtain the change amplitude of temperature, humidity, air pressure, and air velocity within n consecutive sampling periods, historical sampling data needs to be retrieved, and the change rate between adjacent sampling points is calculated. The sampling period is set to 10 seconds. Assume that the temperature sensor records the temperature value as 25.3°C at the current moment t = 100 seconds and 25.8°C at t = 110 seconds. Then the temperature change rate is calculated as follows: (25.8 - 25.3) / (110 - 100) = 0.05°C / s. Similarly, the change rates of humidity, air pressure, and air velocity are calculated in the same way. If the change rate of a certain environmental parameter exceeds the set threshold within multiple consecutive sampling periods, for example, the temperature rate threshold is set to 0.06°C / s, the humidity rate threshold is set to 0.08%RH / s, the air pressure rate threshold is set to 0.005 kPa / s, and the air velocity rate threshold is set to 0.003 m / s, then it is determined whether a mutation has occurred. For example, at a certain moment, the temperature rate reaches 0.07°C / s, exceeding the threshold, then record this time point and mark it as a mutation point. The same method is used to determine mutations for humidity, air pressure, and air velocity, and finally, the mutation determination result of the environmental parameter is obtained.

[0104] S202: Invoke the determination result of the environmental parameter mutation, calculate the data fluctuation amplitude of multiple sampling points before and after the mutation point, and determine whether the inflection point belongs to the convergence or divergence type, using the formula:

[0105]

[0106] Calculate the environmental parameter fluctuation amplitude, and perform screening in combination with the inflection point change trend to obtain the environmental parameter inflection point type;

[0107] Among them, D i1 represents the environmental parameter fluctuation amplitude, X j1 represents the environmental parameter value of the j1-th sampling point, represents the average value of the environmental parameters in the interval, V j1 represents the environmental parameter change rate of the j1-th sampling point, represents the average value of the environmental parameter change rates in the interval, A j1 represents the environmental parameter acceleration of the j1-th sampling point, represents the average value of the environmental parameter accelerations in the interval, k1 represents the length of the time window under investigation, and n1 represents the number of sampling points;

[0108] Invoke the determination result of the environmental parameter mutation, calculate the data fluctuation amplitude of multiple sampling points before and after the mutation point to judge the convergence or divergence type of the inflection point, and the calculation formula is:

[0109]

[0110] Among them, X j1 is the environmental parameter value of the j1-th sampling point, is the average value of the environmental parameters in this interval, V j1 is the environmental parameter change rate of the j1-th sampling point, is the average value of the change rates in this interval, A j1 is the environmental parameter acceleration of the j1-th sampling point, is the average value of the accelerations in this interval, k1 represents the length of the time window under investigation, and n1 represents the number of sampling points in this time period.

[0111] Table 3 Calculation data of environmental parameter fluctuation amplitude

[0112]

[0113]

[0114] As shown in Table 3, calculate the fluctuation amplitudes of temperature, humidity, air pressure, and air velocity, and screen the data in combination with the inflection point change trend. If the fluctuation amplitude meets the set threshold and the change trend converges to the mean value, then determine that this inflection point is of the convergence type; otherwise, determine it as the divergence type.

[0115] S203: Invoke the inflection point type of the environmental parameters, set the environmental parameter offset threshold, calculate the deviation degree of the current environmental parameters relative to the mean value, extract the temperature offset rate, humidity offset rate, air pressure offset rate, and air flow rate offset rate within multiple consecutive sampling periods, and determine the offset status in combination with the trend direction to obtain the environmental parameter offset prediction value.

[0116] Invoke the inflection point type of the environmental parameters, and set the environmental parameter offset threshold. Calculate the deviation degree of the current environmental parameters relative to the mean value, and extract the offset rate within multiple consecutive sampling periods. Assume that the set temperature offset threshold is ±0.5°C, the humidity offset threshold is ±0.8%RH, the air pressure offset threshold is ±0.02 kPa, and the air flow rate offset threshold is ±0.05 m / s. In the current measurement data, the average temperature is 25.6°C. If the measured temperature at a certain moment is 26.2°C, the temperature offset is calculated as follows: (26.2 - 25.6) = 0.6°C, exceeding the threshold of 0.5°C. Therefore, it is determined that there is an offset phenomenon for this temperature. Similarly, calculate the humidity offset rate (61.4 - 60.6) = 0.8%RH, which does not exceed the threshold, the air pressure offset rate (101.45 - 101.4) = 0.05 kPa, exceeding the threshold, and the air flow rate offset rate (0.65 - 0.6) = 0.05 m / s, exactly equal to the threshold. Then, determine the offset status of the current environmental parameters in combination with the trend direction.

[0117] Table 4 Calculation data of environmental parameter offset rate

[0118]

[0119] As shown in Table 4, compare the calculated offset rate with the set threshold, and determine the offset status in combination with the trend direction. If the offset value continues to rise and exceeds the threshold, it is determined as continuous offset. If it oscillates within the threshold, it is determined as stable. Finally, obtain the environmental parameter offset prediction value.

[0120] Please refer to Figure 4 , and the specific steps for obtaining the dynamic calibration correction value are as follows:

[0121] S301: Based on the environmental parameter offset prediction value, screen the parameters whose offset amount exceeds the environmental parameter offset threshold, invoke the environmental data to calculate the offset rate and duration of multiple parameters, calculate the change rate in the time series, determine whether the parameters exceed the set range and screen the qualified parameters, record the offset duration, invoke the offset rate and duration, and obtain the environmental parameter offset characteristic value;

[0122] Based on the predicted value of environmental parameter offset, the parameters whose offset exceeds the environmental parameter offset threshold are screened, and the environmental data is called to calculate the offset rate and duration of multiple parameters. First, the offset threshold of each environmental parameter is set, such as the temperature offset threshold is ±0.5°C, the humidity offset threshold is ±0.8%RH, the air pressure offset threshold is ±0.02kPa, and the air flow rate offset threshold is ±0.05m / s. In the current measurement data, the temperature reaches 26.2°C at a certain moment, which is 0.6°C offset relative to the mean value of 25.6°C, exceeding the 0.5°C threshold. Therefore, the parameter is judged to be offset, and then its offset rate is calculated. The offset rate calculation method is (current offset - previous offset) / sampling For example, if the temperature offset at the previous moment is 0.4°C, the current temperature offset is 0.6°C, and the sampling period is 10 seconds, the temperature offset rate is calculated as follows: (0.6-0.4) / 10=0.02°C / s. The offset rates of humidity, air pressure, and air flow rate are calculated in the same way, and the offset duration is recorded at the same time, that is, the timing starts from the time point when the offset exceeds the threshold for the first time and continues until the offset returns to within the threshold. If the temperature offset reaches 0.6°C at t=100s and returns to 0.4°C at t=200s, the offset duration is 200s-100s=100s. All calculation results are summarized to finally obtain the environmental parameter offset characteristic value.

[0123] Table 5 Calculation data of environmental parameter offset characteristic values

[0124]

[0125] As shown in Table 5, the calculated environmental parameter offset rate and offset duration can be used to further adjust the production environment. The results show that the current environmental parameters have a large offset within a certain period of time and a long offset duration, which may affect production stability. Therefore, subsequent calibration steps are required to dynamically adjust them to ensure that the production environment is in a stable state.

[0126] S302: Call the environmental parameter offset characteristic value, calculate the environmental data of the current cycle and the previous cycle, and adjust the production environment calibration coefficient using the formula:

[0127]

[0128] Calculate dynamic calibration correction values ​​and apply them to production environment parameter adjustments;

[0129] Among them, C t Represents the production environment calibration coefficient of the current cycle, C t-1 Represents the production environment calibration coefficient of the previous cycle, P i1 represents the offset value of the i1th environmental parameter, P th represents the environmental parameter deviation threshold, T i1Represents the offset duration of the i1-th parameter, V i1 Represents the offset rate of the i1-th parameter, A avg Represents the mean value of the environmental parameter acceleration, and n1 represents the number of selected environmental parameters.

[0130] Call the environmental parameter offset eigenvalue, calculate the environmental data of the current cycle and the previous cycle, adjust the production environment calibration coefficient, and set the production environment calibration coefficient C of the previous cycle t-1 Is 1.0. During the current cycle, the temperature offset P i1 Is 0.6 °C, the offset threshold P th Is set to 0.5 °C, the offset duration T i1 Is 100 s, the offset rate V i1 Is 0.02 °C / s. When calculating the correction value, the mean value A of the environmental parameter acceleration also needs to be introduced avg , Assuming that the calculated A avg Is 0.002 °C / s2, then substitute it into the formula:

[0131]

[0132] Adjust some parameter values to fine-tune the calibration coefficient, and set T i1 = 100 s, V i1 = 0.02 °C / s, the calculation is as follows:

[0133]

[0134] Adjust the value to make the calibration adjustment amplitude more in line with the actual situation, and adjust T i1 = 10 s, V i1 = 0.02 °C / s:

[0135]

[0136] C t = 46.45;

[0137] Finally, calculate the production environment calibration coefficient C of the current cycle t = 46.45.

[0138] The calculated calibration coefficient can be used to adjust the production environment parameters. This result indicates that the production environment of the current cycle needs to be fine-tuned due to the offset of the temperature parameter, from 1.0 to 46.45, indicating that the impact of temperature offset on the production environment requires a certain degree of compensation, but it has not reached the extreme correction situation. Further optimization may be required in combination with other environmental parameters in the future to ensure the stability of the production environment.

[0139] Please refer to Figure 5, the specific steps for obtaining the optimized adjustment value of the environmental parameters are as follows:

[0140] S401: Based on the dynamic calibration correction value, analyze the change trends of temperature, humidity, air pressure, and air flow velocity, calculate the change rates of multiple parameters in the time series respectively, obtain the change directions, judge whether the change directions of multiple parameters are consistent, screen out the parameter pairs with opposite change directions, and obtain the set of parameter pairs with opposite trends;

[0141] Based on the dynamic calibration correction value, analyze the change trends of temperature, humidity, air pressure, and air flow velocity, calculate the change rates of each parameter in the time series respectively, and obtain the change directions. Assume that within a certain time period, the temperature change rate is 0.03 °C / s, the humidity change rate is -0.02%RH / s, the air pressure change rate is 0.005 kPa / s, and the air flow velocity change rate is -0.003 m / s. Then the temperature and air pressure change rates are positive, indicating that their values are increasing, while the humidity and air flow velocity change rates are negative, indicating that their values are decreasing. Subsequently, screen out the parameter pairs with opposite change directions, judge whether the change directions of multiple parameters are consistent. If the change rates in the same parameter group are one positive and one negative, then this parameter pair is selected into the set of parameter pairs with opposite trends. For example, {temperature, humidity} and {air pressure, air flow velocity} both meet this condition, then screen out the set of parameter pairs with opposite trends {(temperature, humidity), (air pressure, air flow velocity)}, and finally obtain the set of parameter pairs with opposite trends.

[0142] Table 6 Calculation data of parameter pairs with opposite trends

[0143]

[0144] As shown in Table 6, the change directions of temperature-humidity and air pressure-air flow velocity are opposite and are screened as parameter pairs with opposite trends. This result indicates that some parameters in the current production environment have a tendency to cancel each other out, which can be used for the subsequent calculation of the error cancellation rate.

[0145] S402: Call the set of parameter pairs with opposite trends, calculate the error cancellation situation between multiple parameter pairs, calculate the complementary effect of parameter pairs using the absolute error difference, and count the error cancellation degree of all parameter pairs to obtain the error cancellation rate distribution value;

[0146] Call the set of parameter pairs with opposite trends, calculate the error cancellation situation between multiple parameter pairs, calculate the complementary effect of parameter pairs using the absolute error difference, set temperature-humidity and air pressure-air flow velocity as parameter pairs, calculate the error cancellation rate of each parameter pair, and the error calculation method is:

[0147] E i1 =|V 1 -V 2 |;

[0148] Among them, V1 and V 2 are the change rates of the two parameters in the parameter pair respectively, and are calculated as follows:

[0149] E 温度-湿度 = |0.03 - (-0.02)| = 0.05;

[0150] E 气压-空气流速 = |0.005 - (-0.003)| = 0.008;

[0151] Subsequently, the error cancellation degree of all parameter pairs is counted, and the error cancellation rate distribution value is calculated. The calculation method of the error cancellation rate is as follows:

[0152]

[0153] It is calculated that:

[0154]

[0155] The calculated error cancellation rate is relatively low, indicating that the parameters in the current production environment cannot fully cancel each other out and still need to be further adjusted. This result shows that the current parameter adjustment is insufficient and requires minimum error adjustment to obtain a more optimal production environment correction coefficient.

[0156] S403: Call the error cancellation rate distribution value, and for the parameter pairs with incomplete error cancellation, use the minimum error adjustment method to calculate the correction coefficient of the production environment. The correction calculation formula is as follows:

[0157]

[0158] Calculate to obtain the optimized adjustment value of the environmental parameters;

[0159] Among them, C adj represents the calibrated coefficient of the corrected production environment, C orig represents the calibrated coefficient of the original production environment, E i1 represents the absolute value of the error of the i1-th parameter pair with incomplete error cancellation, V j1 represents the j1-th component of the change rate of the environmental parameters participating in the adjustment, S k1 represents the k1-th component of the standard deviation of the uncalibrated parameters, n1 is the number of parameter pairs with incomplete error cancellation, m1 is the number of change rate components of all environmental parameters, and p1 is the number of standard deviation components of the uncalibrated parameters.

[0160] Call the error cancellation rate distribution value, and for the parameter pairs with incomplete error cancellation, use the minimum error adjustment method to calculate the correction coefficient of the production environment. Set the original production environment calibration coefficient C orig = 46.45, and the absolute value of the error E of the parameter pair with incomplete error cancellation i1They are 0.05 and 0.008 respectively, and the environmental parameter change rate component V j1 They are 0.03 and 0.005 respectively, and the standard deviation S of the uncalibrated parameter k1 Set it to 0.002, then calculate the calibrated coefficient of the production environment after correction:

[0161]

[0162] The calculation is as follows:

[0163]

[0164] C adj = 46.45 + 1.79 = 48.24;

[0165] Finally, the calibrated coefficient C of the production environment after correction is calculated adj = 48.24.

[0166] This result indicates that the error compensation effect of the current production environment has been optimized. By adjusting the correction coefficient, the parameter pairs with a lower error cancellation rate have obtained additional compensation, ultimately improving the accuracy of parameter adjustment and making the production environment more stable.

[0167] Please refer to Figure 6 , and the specific steps for obtaining the long-term trend prediction value of the production environment are as follows:

[0168] S501: Based on the optimized adjustment value of the environmental parameter, call the long-term production batch environmental monitoring data, calculate the fluctuation amplitude and change pattern of temperature, humidity, air pressure, and air velocity in each production cycle, and use moving average to calculate the periodic change trend to obtain the periodic fluctuation pattern of the environmental parameter;

[0169] Based on the optimized adjustment value of the environmental parameter, call the environmental monitoring data of the long-term production batch, calculate the fluctuation amplitude and change pattern of temperature, humidity, air pressure, and air velocity in each production cycle. First, collect the environmental data of multiple past production cycles, set the duration of each production cycle to 24 hours, and calculate the fluctuation range of each environmental parameter respectively. For example, if the highest temperature value in a production cycle is 26.5 °C and the lowest value is 24.8 °C, then the temperature fluctuation amplitude is calculated as (26.5 - 24.8) = 1.7 °C. Similarly, the humidity fluctuation amplitude is calculated as (62.1 - 59.5) = 2.6% RH, the air pressure fluctuation amplitude is calculated as (101.6 - 101.3) = 0.3 kPa, and the air velocity fluctuation amplitude is calculated as (0.68 - 0.52) = 0.16 m / s. Subsequently, use moving average to calculate the periodic change trend, set the moving window size to 5 production cycles, calculate the moving average value of each parameter, and obtain the periodic fluctuation pattern of the environmental parameter through smoothing processing.

[0170] Table 7 Environmental parameter fluctuation data during the production cycle

[0171]

[0172] As shown in Table 7, the calculated periodic environmental fluctuation pattern shows a certain variation law. This result indicates that there is a stable fluctuation trend of environmental parameters during the production cycle, which can be used for further calculation of long-term environmental trend characteristics.

[0173] S502: Invoke the periodic fluctuation pattern of the environmental parameters, calculate the change rate of the long-term trend of each parameter, use linear regression to calculate the long-term trend change value, and determine whether the long-term fluctuation trend is stable based on the change amplitude to obtain the long-term environmental trend characteristics;

[0174] Invoke the periodic fluctuation pattern of the environmental parameters, calculate the change rate of the long-term trend of each parameter, use linear regression to calculate the long-term trend change value, set the time axis as the production cycle sequence, and use the fluctuation amplitudes of temperature, humidity, air pressure, and air velocity as the dependent variables to construct a regression equation. For example, the linear regression model of temperature is as follows:

[0175] T 波动 = a × production cycle + b;

[0176] Use the least squares method to fit the data points, calculate the regression slope a and intercept b. Set the calculated temperature change slope to 0.02 °C / cycle, the humidity change slope to 0.03 %RH / cycle, the air pressure change slope to 0.005 kPa / cycle, and the air velocity change slope to 0.002 m / s / cycle. Then calculate the change amplitude and determine whether the long-term fluctuation trend is stable. Set a threshold. If the change slope is less than the set threshold (for example, the temperature change rate threshold is 0.05 °C / cycle), the trend is considered stable; otherwise, it is considered unstable. Finally, obtain the long-term environmental trend characteristics.

[0177] Table 8 Long-term trend change data of the production environment

[0178]

[0179] As shown in Table 8, the long-term change rates of all parameters are within the threshold range. This result indicates that the long-term fluctuation trend of the current environmental parameters is stable, which can be used for further calculation of the long-term calibration reference value.

[0180] S503: Invoke the long-term environmental trend characteristics, calculate the optimal long-term calibration reference value for the parameters with unstable trends, using the formula:

[0181]

[0182] Calculate the long-term calibration reference value and obtain the long-term trend prediction value of the production environment through calculation

[0183] Among them, X new represents the adjusted long-term calibration reference value, and X init represents the initial long-term calibration reference value, Y r1 represents the environmental trend data of the r1-th production cycle, Z represents the long-term environmental trend average value, A r1 represents the adjustment factor associated with the data of the r1-th production cycle, B s1 represents the s1-th component in the parameter fluctuation range, C u1 represents the u1-th data in the environmental stability evaluation, D represents the stability normalization constant, q1 represents the total number of production cycles, t1 represents the total number of fluctuation components, and v1 represents the total number of stability data components.

[0184] Call the long-term environmental trend characteristics, calculate the optimal long-term calibration reference value for the parameters with unstable trends, and set the initial long-term calibration reference value X init = 25.5 °C, calculate the long-term trend average value Z as 25.6 °C, and calculate the environmental trend data Y of all production cycles r1 of the deviation degree, and use the weight adjustment factor A r1 to correct the parameter fluctuation. For example, the temperature trend data Y of a certain production cycle 1 = 25.8 °C, calculate its deviation value |Y 1 - Z| = 0.2 °C. Assume that the relevant adjustment factor A 1 = 0.5, calculate the weighted error correction value of all production cycles, set the total number of fluctuation components t1 = 5, and calculate the sum of squares of all fluctuation components Set the total number of environmental stability data v1 = 3, calculate the sum of all environmental stability data ΣC u1 = 0.03, set the stability normalization constant D = 10, and substitute it into the formula:

[0185]

[0186] Calculate as follows:

[0187]

[0188] X new = 25.5 - 0.44 = 25.06;

[0189] The calculated long-term calibration reference value X new = 25.06 °C. This result indicates that there is a slight change in the current long-term environmental trend, and the long-term reference value needs to be fine-tuned to make it more in line with the actual long-term trend, so as to ensure that the production environment is in an optimized state for a long time.

[0190] A health food production monitoring system based on big data, the health food production monitoring system based on big data is used to execute the above-mentioned health food production monitoring method based on big data, and the system includes:

[0191] The environmental data acquisition module obtains the temperature, humidity, air pressure, and air velocity data of the production workshop, records the continuous sampling data and calculates the change rate, constructs a time series, calls the change rate data to calculate the trend change value of the production environment in the short term, and obtains the production environment trend change value;

[0192] The environmental change analysis module is based on the production environment trend change value, detects the change range of temperature, humidity, air pressure, and air velocity within a continuous target period, calculates the data fluctuation range, screens the fluctuation mutation data, and calculates the change rate before and after the mutation. Calls the set environmental parameter offset threshold to calculate the offset amount between the current data and the set environmental parameter, and obtains the environmental parameter offset prediction value;

[0193] The offset prediction and calibration module is based on the environmental parameter offset prediction value, screens the parameters whose offset amount exceeds the set environmental parameter offset threshold, calculates the parameter offset rate, calls the environmental data of the current period, establishes a dynamic correction parameter, and adjusts the production environment calibration coefficient to obtain a dynamic calibration correction value;

[0194] The parameter optimization and adjustment module is based on the dynamic calibration correction value, analyzes the change trends of temperature, humidity, air pressure, and air velocity, screens the parameter pairs with opposite trend directions, calculates the error cancellation situation, and adjusts the deviation parameters to obtain the environmental parameter optimization and adjustment value;

[0195] The long-term trend monitoring module is based on the environmental parameter optimization and adjustment value, regulates the production workshop environment, collects long-term sampling environmental data, calculates the periodic fluctuations of temperature, humidity, air pressure, and air velocity, extracts long-term trend characteristics, and obtains the long-term trend prediction value of the production environment.

[0196] The above is only the preferred embodiment of the present invention, and it does not limit the present invention in other forms. Any person skilled in the art may use the disclosed technical content to make changes or modifications into equivalent embodiments with equivalent changes and apply them to other fields. However, as long as it does not depart from the technical solution content of the present invention, any simple modification, equivalent change, and modification made to the above embodiments based on the technical essence of the present invention still belong to the protection scope of the technical solution of the present invention.

Claims

1. A health food production monitoring method based on big data, characterized in that: The following steps are involved: S1: Obtain the temperature, humidity, air pressure, and air flow rate of the health food production workshop, collect continuous sampling data, calculate the change rate, record the rate time series, and generate the production environment trend change value; S2: Based on the production environment trend change value, detect the change range of temperature, humidity, air pressure, and air flow rate within the continuous target period, judge the data fluctuation before and after the mutation, calculate the deviation of the current data from the set environmental parameter offset threshold, and generate the environmental parameter offset prediction value; S3: Based on the predicted value of the environmental parameter offset, filter the offset exceeding the environmental parameter offset threshold parameter, calculate the offset rate, call the current cycle environmental data, adjust the production environment calibration coefficient, and generate a dynamic calibration correction value; S4: Based on the dynamic calibration correction value, analyze the trends of temperature, humidity, air pressure, and air velocity, select parameter pairs with opposite directions, calculate the mutual offset of errors, and correct and generate the optimized adjustment value of the environmental parameters; S5: Based on the optimized adjustment value of the environmental parameters, the production workshop environment is regulated, long-term sampling environmental data is collected, periodic fluctuations of temperature, humidity, air pressure, and air flow rate are calculated, long-term trend characteristics are extracted, and long-term trend prediction values ​​of the production environment are generated.

2. The method for monitoring the production of health food based on big data according to claim 1, characterized in that: The production environment trend change value includes the temperature change rate, the humidity change rate, the air pressure change rate, and the air flow rate change rate; the environmental parameter offset prediction value includes the temperature offset prediction value, the humidity offset prediction value, the air pressure offset prediction value, and the air flow rate offset prediction value; the offset exceeding the environmental parameter offset threshold parameter is specifically an exceeding temperature offset threshold parameter, an exceeding humidity offset threshold parameter, an exceeding air pressure offset threshold parameter, and an exceeding air flow rate offset threshold parameter; the dynamic calibration correction value includes a temperature correction value, a humidity correction value, an air pressure correction value, and an air flow rate correction value; the environmental parameter optimization adjustment value includes a temperature optimization adjustment value, a humidity optimization adjustment value, an air pressure optimization adjustment value, and an air flow rate optimization adjustment value; the production environment long-term trend prediction value includes a temperature long-term trend prediction value, a humidity long-term trend prediction value, an air pressure long-term trend prediction value, and an air flow rate long-term trend prediction value.

3. The method for monitoring the production of health food based on big data according to claim 2, characterized in that: The steps for obtaining the production environment trend change value are specifically as follows: S101: Acquire the temperature, humidity, air pressure, and air flow rate of the health food production workshop, collect environmental data in multiple continuous sampling cycles, record the data change values ​​of adjacent sampling points, calculate the temperature change rate, humidity change rate, air pressure change rate, and air flow rate change rate based on the continuous data, and obtain the environmental parameter change rate sequence; S102: Call the environmental parameter change rate sequence to calculate the rate difference between multiple adjacent sampling points using the formula: Calculate the rate transition value sequence and screen out the rate transition key interval; Among them, ΔS i1 represents the rate transition value sequence, V i1 represents the rate of change of environmental parameters at the i1th sampling point, T i1 Represents the timestamp of the i1th sampling point; S103: calling the rate transition key interval, combining the rate change trends of multiple groups of environmental parameters, calculating the trend change rate, and obtaining the production environment trend change value.

4. The method for monitoring the production of health food based on big data according to claim 3, characterized in that: The steps for obtaining the predicted value of the environmental parameter offset are specifically as follows: S201: Based on the production environment trend change value, detect the change range of temperature, humidity, air pressure, and air flow rate within n consecutive sampling periods, calculate the change rate of multiple parameters, and compare them with the set threshold to determine whether a mutation occurs, and obtain the environmental parameter mutation determination result; S202: calling the environmental parameter mutation determination result, calculating the data fluctuation amplitude of multiple sampling points before and after the mutation point, and judging whether the inflection point belongs to the convergence or divergence type, using the formula: Calculate the fluctuation range of environmental parameters, and screen them based on the inflection point change trend to obtain the inflection point type of environmental parameters; Among them, D i1 Represents the fluctuation range of environmental parameters, X j1 represents the environmental parameter value of the j1th sampling point, represents the mean value of the environmental parameter in the interval, V j1 represents the rate of change of environmental parameters at the j1th sampling point, represents the mean rate of change of environmental parameters in the interval, A j1 represents the environmental parameter acceleration of the j1th sampling point, represents the mean value of the environmental parameter acceleration in the interval, k1 represents the length of the time window for investigation, and n1 represents the number of sampling points; S203: Call the environmental parameter inflection point type, set the environmental parameter offset threshold, calculate the deviation degree of the current environmental parameter relative to the mean, extract the temperature offset rate, humidity offset rate, air pressure offset rate, and air flow rate offset rate within multiple continuous sampling periods, determine the offset state based on the trend direction, and obtain the environmental parameter offset prediction value.

5. The method for monitoring the production of health food based on big data according to claim 4, characterized in that: The steps for obtaining the dynamic calibration correction value are specifically as follows: S301: Based on the predicted value of the environmental parameter offset, filter out parameters whose offset exceeds the environmental parameter offset threshold, call environmental data to calculate the offset rate and duration of multiple parameters, calculate the rate of change in the time series, determine whether the parameter exceeds the set range and filter out parameters that meet the conditions, record the offset duration, call the offset rate and duration, and obtain the environmental parameter offset characteristic value; S302: Call the environmental parameter offset characteristic value, calculate the environmental data of the current cycle and the previous cycle, and adjust the production environment calibration coefficient using the formula: Calculate dynamic calibration correction values ​​and apply them to production environment parameter adjustments; Among them, C t Represents the production environment calibration coefficient of the current cycle, C t-1 Represents the production environment calibration coefficient of the previous cycle, P i1 represents the offset value of the i1th environmental parameter, P th represents the environmental parameter deviation threshold, T i1 represents the offset duration of the i1th parameter, V i1 represents the offset rate of the i1th parameter, A avg represents the mean value of the environmental parameter acceleration, and n1 represents the number of screened environmental parameters.

6. The method for monitoring the production of health food based on big data according to claim 5, characterized in that: The steps for obtaining the optimized adjustment value of the environmental parameters are specifically as follows: S401: Based on the dynamic calibration correction value, analyzing the change trends of temperature, humidity, air pressure, and air flow rate, respectively calculating the change rates of multiple parameters in the time series, and obtaining the change directions, determining whether the change directions of multiple parameters are consistent, and screening parameter pairs with opposite change directions to obtain a set of parameter pairs with opposite trends; S402: calling the set of parameter pairs with opposite trends, calculating the error cancellation between multiple parameter pairs, using absolute error differences to calculate the complementary effects of parameter pairs, and counting the error cancellation degrees of all parameter pairs to obtain a distribution value of the error cancellation rate; S403: calling the error compensation rate distribution value, and using the minimum error adjustment method to calculate the correction coefficient of the production environment for the parameter pairs whose errors are not completely compensated. The correction calculation formula is as follows: Calculate and obtain the optimized adjustment value of environmental parameters; Among them, C adj Represents the corrected production environment calibration factor, C orig Represents the original production environment calibration factor, E i1 Represents the absolute value of the error of the i1th parameter pair that does not completely cancel the error value, V j1 represents the j1th component of the rate of change of the environmental parameter involved in the adjustment, S k1 represents the k1th component of the standard deviation of the uncalibrated parameter, n1 is the number of parameter logarithms whose errors are not completely offset, m1 is the number of rate of change components of all environmental parameters, and p1 is the number of standard deviation components of the uncalibrated parameter.

7. The method for monitoring the production of health food based on big data according to claim 6, characterized in that: The steps for obtaining the long-term trend forecast value of the production environment are specifically as follows: S501: Based on the optimized adjustment value of the environmental parameter, long-term production batch environmental monitoring data is called to calculate the fluctuation amplitude and change pattern of temperature, humidity, air pressure, and air flow rate in each production cycle, and the periodic change trend is calculated by moving average to obtain the periodic fluctuation pattern of the environmental parameter; S502: calling the environmental parameter periodic fluctuation mode, calculating the change rate of the long-term trend of each parameter, using linear regression to calculate the long-term trend change value, and determining whether the long-term fluctuation trend is stable according to the change amplitude, and obtaining the long-term environmental trend characteristics; S503: Call the long-term environmental trend feature, and calculate the optimal long-term calibration reference value for the trend instability parameter, using the formula: Calculate long-term calibration benchmark values ​​and calculate long-term trend forecast values ​​for the production environment; Among them, X new represents the adjusted long-term calibration reference value, X init represents the initial long-term calibration reference value, Y r1 represents the environmental trend data of the r1th production cycle, Z represents the long-term environmental trend average, A r1 represents the adjustment factor associated with the r1th production cycle data, B s1 represents the s1th component of the parameter fluctuation amplitude, C u1 represents the u1th data in the environmental stability evaluation, D represents the stability normalization constant, q1 represents the total number of production cycles, t1 represents the total number of fluctuation components, and v1 represents the total number of stability data components.

8. A health food production monitoring system based on big data, characterized in that: According to any one of claims 1 to 7, the method for monitoring the production of health food based on big data comprises: The environmental data acquisition module obtains the temperature, humidity, air pressure, and air flow rate data of the production workshop, records the continuous sampling data and calculates the change rate, constructs a time series, and calls the change rate data to calculate the trend change value of the production environment in a short period of time to obtain the trend change value of the production environment; The environmental change analysis module detects the change range of temperature, humidity, air pressure, and air flow rate within the continuous target period based on the production environment trend change value, calculates the data fluctuation range, filters the fluctuation mutation data, and calculates the change rate before and after the mutation, calls the set environmental parameter offset threshold, calculates the offset between the current data and the set environmental parameter, and obtains the environmental parameter offset prediction value; The offset prediction and calibration module screens parameters whose offset exceeds the set environmental parameter offset threshold based on the environmental parameter offset prediction value, calculates the parameter offset rate, calls the current cycle environmental data, establishes dynamic correction parameters, adjusts the production environment calibration coefficient, and obtains the dynamic calibration correction value; The parameter optimization and adjustment module analyzes the change trends of temperature, humidity, air pressure, and air velocity based on the dynamic calibration correction value, selects parameter pairs with opposite trend directions, calculates the error offset, adjusts the deviation parameters, and obtains the environmental parameter optimization adjustment value; The long-term trend monitoring module optimizes the adjustment value based on the environmental parameters, regulates the production workshop environment, collects long-term sampled environmental data, calculates the periodic fluctuations of temperature, humidity, air pressure, and air flow rate, extracts long-term trend characteristics, and obtains the long-term trend prediction value of the production environment.

Images