A food data processing method, apparatus, and storage medium based on big data.

Abstract

This invention relates to the field of food data processing technology, and more particularly to a food data processing method, apparatus, and storage medium based on big data. The method includes: collecting environmental data, food data, and spatiotemporal data during the cold chain transportation of food; heterogeneously fusing the collected data to construct a spatiotemporal matrix and a food correlation matrix; dynamically extracting features and allocating environmental coupling weights to the spatiotemporal matrix to obtain temperature features, microbial features, and environmental coupling features; concatenating the temperature features, microbial features, and environmental coupling weights into a feature vector; determining the data status and issuing warnings based on the feature vector; collecting the food status after cold chain transportation; and updating the data status based on the food status, environmental data, and spatiotemporal data. This invention achieves precise monitoring and processing of food data during cold chain transportation.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of food data processing, and in particular to a food data processing method and device based on big data and a storage medium. BACKGROUND

[0002] Food data monitoring and processing is the use of big data technology to clean, integrate and analyze data at various stages of food production, circulation and consumption, to identify safety risks through data mining and machine learning algorithms, improve regulatory efficiency, ensure food safety, optimize production processes, provide scientific basis for policy making, and protect public health.

[0003] The food data processing in the prior art is mostly related to safety detection and abnormal analysis, and lacks in-depth analysis under specific environments or special conditions, real-time monitoring and prediction of the impact of food temperature fluctuations in cold chain logistics on quality, and real-time monitoring and risk warning of microbial growth dynamics in the food processing process. SUMMARY

[0004] The present application aims to provide a food data processing method and device based on big data and a storage medium to solve at least one of the problems in the prior art.

[0005] To achieve the above-mentioned purpose, the present application adopts the following technical solutions:

[0006] A food data processing method based on big data includes:

[0007] Collecting environmental data, food data and spatio-temporal data during food cold chain transportation;

[0008] Performing heterogeneous fusion on the collected data to construct a spatio-temporal matrix and a food association matrix;

[0009] Performing dynamic feature extraction and environmental coupling weight distribution on the spatio-temporal matrix to obtain temperature features, microbial features and environmental coupling features;

[0010] Concatenating the temperature features, microbial features and environmental coupling weights into a feature vector, judging the data state and issuing an early warning based on the feature vector;

[0011] Collecting the food state after food cold chain transportation;

[0012] Updating the data state analysis process based on the food state, environmental data and spatio-temporal data.

[0013] Further, the collection time of the environmental data and the spatio-temporal data is aligned, and the collection time of the environmental data and the spatio-temporal data is sequentially numbered in chronological order to obtain a collection number t, the environmental data and the spatio-temporal data at the same collection time are heterogeneously fused to construct a spatio-temporal matrix E(t), and E(t) is set as [T(t), H(t), C(t), L(t)], wherein T(t) represents temperature, H(t) represents humidity, C(t) represents carbon dioxide concentration, and L(t) represents altitude.

[0014] Further, the temperature in the spatio-temporal matrix within 1 hour is dynamically characterized based on the length of the preset analysis period to obtain a temperature feature, and the expression of the temperature feature is: , wherein F(t) represents the temperature feature, v represents a first extraction parameter, v ∈ N and v ≤ V, V represents a second extraction parameter, V = 3600 / t1, t1 represents the length of the preset analysis period, and λ represents a temperature decay factor.

[0015] Further, the temperature and the food-related matrix in the spatio-temporal matrix are dynamically characterized based on the initial microbial content to obtain a microbial feature, and the expression of the microbial feature is: G1(i)×[T(t)-G2(i)] / G2(i) , wherein G(t) represents the microbial feature, and G0 represents the initial microbial content.

[0016] Further, the carbon dioxide concentration and the altitude in the spatio-temporal matrix are subjected to environmental coupling weight distribution to obtain an environmental coupling feature, and the expression of the environmental coupling feature is W(t) = β1×(1-L(t) / Lmax) + β2×C(t) / Cmax, wherein β1 represents an altitude weight, β2 represents a carbon dioxide weight, β1 + β2 = 1, Lmax represents a preset maximum tolerance altitude, and Cmax represents a preset carbon dioxide critical concentration.

[0017] Further, the temperature feature, the microbial feature, and the environmental coupling weight are spliced to obtain a feature vector X(t), X(t) is set as [F(t), G(t), W(t)], and the dot product of the feature vectors adjacent in the collection number is taken as a risk feature value, and the expression of the risk feature value is R(t) = X(t)·X(t-1).

[0018] The risk feature value in the food cold chain transportation process is extracted, the average value and the standard deviation of the extracted risk feature value are calculated, and a feature value tolerance interval is formulated according to the average value and the standard deviation of the extracted risk feature value, the feature value tolerance interval is set as [μ(R(t))-3×σ(R(t)), μ(R(t))+3×σ(R(t))], if the risk feature value of the current analysis period is not in the feature value tolerance interval, it is determined that the data state is abnormal, otherwise, it is determined that the data state is normal.

[0019] According to the data state and the risk characteristic value, when the data state is abnormal, if the risk characteristic value is less than the first characteristic threshold, a severe early warning is performed, if the risk characteristic value is greater than or equal to the first characteristic threshold and less than the second characteristic threshold, a moderate early warning is performed, and if the risk characteristic value is greater than or equal to the second characteristic threshold, a mild early warning is performed; when the data state is normal, if the risk characteristic value is less than the first characteristic threshold, a moderate early warning is performed, and if the risk characteristic value is greater than or equal to the first characteristic threshold, no early warning is performed.

[0020] Further, the risk characteristic value is divided into normal characteristic values and deteriorated characteristic values according to the food state, the number of normal characteristic values not belonging to the characteristic value tolerance interval is counted as a first abnormality number, the number of deteriorated characteristic values not belonging to the characteristic value tolerance interval is counted as a second abnormality number, and a state correlation coefficient is analyzed, and an expression of the state correlation coefficient is S = σ (R1(t)) Nr1 / NR1 / σ (R2(t)) Nr2 / NR2 , wherein S represents the state correlation coefficient, R1(t) represents the normal characteristic value, Nr1 represents the first abnormality number, NR1 represents the number of normal characteristic values, R2(t) represents the deteriorated characteristic value, Nr2 represents the second abnormality number, and NR2 represents the number of deteriorated characteristic values.

[0021] The transportation time is taken as an independent variable, and the product of the humidity and the carbon dioxide concentration is taken as a dependent variable for regression analysis, so as to analyze a regression equation of the product of the humidity and the carbon dioxide concentration with respect to the transportation time, and calculate a determination coefficient of the regression equation, if the determination coefficient is greater than a determination threshold, the analysis process of the state correlation coefficient is not updated, otherwise, the expression of the state correlation coefficient is updated as S = r × σ (R1(t)) Nr1 / NR1 / σ (R2(t)) Nr2 / NR2 , wherein r represents the determination coefficient.

[0022] Further, the state correlation coefficient is compared with a correlation threshold, when the state correlation coefficient is greater than the correlation threshold, the characteristic tolerance interval is updated as [μ(R(t))-(3-S)×σ(R(t)), μ(R(t))+(3+S)×σ(R(t))], so as to update the analysis process of the data state, otherwise, the analysis process of the data state is not updated.

[0023] On the other hand, the application also provides a food data processing device based on big data, comprising:

[0024] The acquisition module is used to acquire environmental data, food data and space-time data in the food cold chain transportation process and food state after the food cold chain transportation;

[0025] The fusion module is used to perform heterogeneous fusion of the collected data to construct a spatiotemporal matrix and a food correlation matrix;

[0026] The analysis module is used to perform dynamic feature extraction and environmental coupling weight allocation on the spatiotemporal matrix to obtain temperature features, microbial features, and environmental coupling features.

[0027] The early warning module is used to concatenate temperature features, microbial features, and environmental coupling weights into a feature vector, and also to determine the data status and issue early warnings based on the feature vector;

[0028] The update module is used for the analysis process of updating the data status based on food status, environmental data, and spatiotemporal data.

[0029] On the other hand, the present invention also provides a storage medium storing instructions that, when run on a computer, cause the computer to perform the big data-based food data processing method as described in any of the preceding claims.

[0030] The beneficial effects of this invention are as follows: through multi-source data fusion, dynamic feature extraction, environmental coupling weight allocation and feedback-driven comprehensive analysis, it can achieve accurate prediction and hierarchical control of food spoilage risk in complex cold chain scenarios, solve the limitations of static threshold criteria in traditional methods, and realize proactive risk prediction and adaptive control. Attached Figure Description

[0031] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying 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.

[0032] Figure 1 This is a flowchart of the food data processing method based on big data in this embodiment.

[0033] Figure 2 This is a flowchart of the feature analysis method for the spatiotemporal matrix in this embodiment.

[0034] Figure 3 This is a flowchart of the data status update method in this embodiment.

[0035] Figure 4 This is a schematic diagram of the food data processing device based on big data in this embodiment. Detailed Implementation

[0036] To make the objectives, technical solutions, and advantages of this invention clearer, the 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 merely illustrative and not intended to limit the invention.

[0037] It should be understood that in the description of the embodiments of the present invention, "multiple" (or "amounts") means two or more, "greater than," "less than," and "exceeding" are understood to exclude the stated number, while "above," "below," and "within" are understood to include the stated number. If "first," "second," etc., are used in the description, they are only for the purpose of distinguishing technical features and should not be construed as indicating or implying relative importance, or implicitly indicating the number of indicated technical features, or implicitly indicating the order of the indicated technical features.

[0038] Please see Figure 1 As shown, this is a food data processing method based on big data in this embodiment, including:

[0039] Step S1: Collect environmental data, food data, and spatiotemporal data during the cold chain transportation of food. The environmental data includes temperature, humidity, and carbon dioxide concentration. The temperature is in degrees Celsius, the humidity is relative humidity (a percentage value), and the carbon dioxide concentration is in ppm. The food data includes food type, initial microbial content, microbial growth rate, and critical spoilage temperature. The critical spoilage temperature is in degrees Celsius, and the microbial growth rate is the specific growth rate. The spatiotemporal data includes transportation time and altitude. The transportation time is in hours, and the altitude is in meters. The environmental and spatiotemporal data are collected once every preset analysis cycle. In this embodiment, the preset analysis cycle is set to 3 seconds. This embodiment does not specifically limit the setting of the preset analysis cycle; those skilled in the art can freely set it, such as 1 second, 5 seconds, etc. The environmental and spatiotemporal data are collected by sensors, and the food data is collected and uploaded through food microbiological testing methods.

[0040] Specifically, in step S1 of this embodiment, temperature, humidity, carbon dioxide concentration, altitude, transportation time and food microbial data are collected to cover the full dimensions of environment, time and space and biological attributes, so as to achieve multi-dimensional data fusion and avoid misjudgment caused by analyzing a single parameter.

[0041] Please continue reading. Figure 1 As shown, the big data-based food data processing method further includes:

[0042] Step S2 involves heterogeneous fusion of the collected data to construct a spatiotemporal matrix and a food association matrix.

[0043] Specifically, in step S2 of this embodiment, the collection times of environmental data and spatiotemporal data are aligned, and the collection times of environmental data and spatiotemporal data are sequentially numbered according to time order to obtain the collection number t. Environmental data and spatiotemporal data collected at the same time are heterogeneously fused to construct a spatiotemporal matrix E(t), which is set as E(t)=[T(t),H(t),C(t),L(t)], where T(t) represents temperature, H(t) represents humidity, C(t) represents carbon dioxide concentration, and L(t) represents altitude.

[0044] Specifically, in step S2 of this embodiment, the food type is associated with its corresponding microbial growth rate and the critical temperature of spoilage in the food association matrix to construct a food association matrix, E(i), which is set as E(i)=[G1(i),G2(i)], where i represents the food type, G1(i) represents the microbial growth rate, and G2(i) represents the critical temperature of spoilage.

[0045] Specifically, in step S2 of this embodiment, by aligning environmental data and spatiotemporal data according to timestamps, a spatiotemporal matrix and a food association matrix are constructed, thereby eliminating feature deviations caused by data time misalignment, supporting efficient feature extraction and weight allocation, and improving data processing efficiency.

[0046] Please continue reading. Figure 1 As shown, the big data-based food data processing method further includes:

[0047] Step S3 involves performing dynamic feature extraction and environmental coupling weight allocation on the spatiotemporal matrix to obtain temperature features, microbial features, and environmental coupling features.

[0048] Please see Figure 2 As shown, it is a feature analysis method for spatiotemporal matrices, including:

[0049] Step S31: Dynamic feature extraction is performed on the temperature in the spatiotemporal matrix to obtain temperature features.

[0050] Specifically, in step S31 of this embodiment, dynamic features of temperature are extracted from the spatiotemporal matrix within one hour based on a preset analysis period, to obtain temperature features. The expression for the temperature features is as follows: In the formula, F(t) represents the temperature characteristic, v represents the first extraction parameter, v∈N and v≤V, V represents the second extraction parameter, V=3600 / t1, t1 represents the duration of the preset analysis period, and λ represents the temperature decay factor, 0<λ≤0.5. It is understood that this embodiment does not specifically limit the value of the temperature decay factor; those skilled in the art can freely set it, as long as it satisfies the analysis of the temperature characteristic. The optimal value of the temperature decay factor is: λ=0.2.

[0051] Specifically, in step S31 of this embodiment, the temperature data in the spatiotemporal matrix is ​​analyzed for features, and the temperature decay factor is used to give higher weight to recent temperature fluctuations, making it more sensitive to capturing sudden temperature anomalies.

[0052] Please continue reading. Figure 2 As shown, the feature analysis method for the spatiotemporal matrix further includes:

[0053] Step S32: Dynamic feature extraction is performed on the temperature and food correlation matrix in the spatiotemporal matrix to obtain microbial features.

[0054] Specifically, in step S32 of this embodiment, dynamic feature extraction is performed on the temperature and food correlation matrix in the spatiotemporal matrix based on the initial microbial content to obtain microbial features. The expression for the microbial features is: G(t) = G0 × e G1(i)×[T(t)-G2(i)] / G2(i) In the formula, G(t) represents the microbial characteristics, and G0 represents the initial microbial content.

[0055] Please continue reading. Figure 2 As shown, the feature analysis method for the spatiotemporal matrix further includes:

[0056] Step S33: Perform environmental coupling weight allocation on carbon dioxide concentration and altitude in the spatiotemporal matrix to obtain environmental coupling characteristics.

[0057] Specifically, in step S33 of this embodiment, environmental coupling weights are assigned to the carbon dioxide concentration and altitude in the spatiotemporal matrix to obtain environmental coupling characteristics. The expression of the environmental coupling characteristics is: W(t)=β1×(1-L(t) / Lmax)+β2×C(t) / Cmax, where β1 represents the altitude weight, β2 represents the carbon dioxide weight, β1+β2=1, Lmax represents the preset maximum tolerable altitude, and Cmax represents the preset critical carbon dioxide concentration.

[0058] Specifically, in this embodiment, the altitude weight is set to 0.6, the carbon dioxide weight is set to 0.4, the preset maximum tolerable altitude is set to 3000, and the preset carbon dioxide critical concentration is set to 1000. It is understood that this embodiment does not specifically limit the settings of altitude weight, carbon dioxide weight, preset maximum tolerable altitude, and preset carbon dioxide critical concentration. Those skilled in the art can freely set them, which are related to the types of microorganisms in food. The preset maximum tolerable altitude represents the upper limit of altitude at which microorganisms in food can survive. Exceeding this value indicates that the inhibitory effect of altitude on microorganisms is saturated. The preset carbon dioxide critical concentration is the concentration threshold at which carbon dioxide significantly inhibits microorganisms.

[0059] Specifically, in step S33 of this embodiment, environmental coupling weights are assigned to the carbon dioxide concentration and altitude in the spatiotemporal matrix to quantify the synergistic effect of low oxygen and modified atmosphere packaging in high-altitude transportation and improve risk sensitivity.

[0060] Please continue reading. Figure 1 As shown, the big data-based food data processing method further includes:

[0061] Step S4: The temperature features, microbial features, and environmental coupling weights are concatenated into a feature vector, and the data status is determined and an early warning is issued based on the feature vector.

[0062] Specifically, in step S4 of this embodiment, temperature features, microbial features, and environmental coupling weights are concatenated to obtain a feature vector X(t). X(t) is set to [F(t), G(t), W(t)], and the dot product of the feature vectors with adjacent collection numbers is used as the risk feature value. The expression for the risk feature value is R(t) = X(t)·X(t-1), where R(t) represents the risk feature value.

[0063] It is understood that this embodiment does not specifically limit the analysis method of risk feature values. Those skilled in the art can freely set it. For example, the feature vector can also be input into the LSTM network model to realize the prediction of food data, judge the similarity between the predicted data and the current data, and analyze the risk feature values.

[0064] Specifically, in step S4 of this embodiment, risk characteristic values ​​during the cold chain transportation of food are extracted, the average value and standard deviation of the extracted risk characteristic values ​​are calculated, and a characteristic value tolerance interval is determined based on the average value and standard deviation of the extracted risk characteristic values. The characteristic value tolerance interval is set as [μ(R(t))-3×σ(R(t)),μ(R(t))+3×σ(R(t))]. If the risk characteristic value of the current analysis period does not belong to the characteristic value tolerance interval, the data status is determined to be abnormal; otherwise, the data status is determined to be normal. Here, μ() represents the average value of the data in parentheses, and σ() represents the standard deviation of the data in parentheses.

[0065] Specifically, in step S4 of this embodiment, an early warning is issued based on the data status and risk characteristic value. When the data status is abnormal, if the risk characteristic value is less than the first characteristic threshold, a severe warning is issued; if the risk characteristic value is greater than or equal to the first characteristic threshold and less than the second characteristic threshold, a moderate warning is issued; if the risk characteristic value is greater than or equal to the second characteristic threshold, a mild warning is issued. When the data status is normal, if the risk characteristic value is less than the first characteristic threshold, a moderate warning is issued; if the risk characteristic value is greater than or equal to the first characteristic threshold, no warning is issued.

[0066] Specifically, in this embodiment, the first feature threshold is set to 0.2 and the second feature threshold is set to 0.9. In this embodiment, the values ​​of the first feature threshold and the second feature threshold are not specifically limited. Those skilled in the art can set them freely, as long as the warning is met. The value range of the first feature threshold should be (0, 0.3) and the value range of the second feature threshold should be [0.8, 1].

[0067] Specifically, in step S4 of this embodiment, risk feature values ​​are generated by constructing feature vectors and dynamically setting tolerance intervals in combination with the 3σ principle. The dynamic tolerance interval is used to adapt to changes in data distribution and reduce misjudgments caused by data drift.

[0068] Please continue reading. Figure 1 As shown, the big data-based food data processing method further includes:

[0069] Step S5: Collect the food status after cold chain transportation. The food status includes normal and spoiled. The food status is collected through user interaction input.

[0070] Please continue reading. Figure 1 As shown, the big data-based food data processing method further includes:

[0071] Step S6 is the analysis process of updating the data status based on food status, environmental data, and spatiotemporal data.

[0072] Please see Figure 3 As shown, this is a method for updating data status, including:

[0073] Step S61: Analyze the correlation coefficient between risk characteristic values ​​and food status.

[0074] Specifically, in step S61 of this embodiment, the risk characteristic values ​​are divided into two categories according to the food state: normal characteristic values ​​and spoiled characteristic values. The number of normal characteristic values ​​that do not belong to the characteristic value tolerance range is counted as the first type of abnormal quantity, and the number of spoiled characteristic values ​​that do not belong to the characteristic value tolerance range is counted as the second type of abnormal quantity. The state correlation coefficient is analyzed, and the expression of the state correlation coefficient is S=σ(R1(t)). Nr1 / NR1 / σ(R2(t)) Nr2 / NR2 In the formula, S represents the state correlation coefficient, R1(t) represents the normal characteristic value, Nr1 represents the number of Class I anomalies, NR1 represents the number of normal characteristic values, R2(t) represents the deteriorated characteristic value, Nr2 represents the number of Class II anomalies, and NR2 represents the number of deteriorated characteristic values.

[0075] Specifically, in step S61 of this embodiment, the reliability of the model prediction is quantified by statistically correlating abnormal feature values ​​with the actual food state, thus distinguishing between occasional noise and actual spoilage.

[0076] Please continue reading. Figure 3 As shown, the data status update method further includes:

[0077] Step S62 is the analysis process of updating the data state based on the state correlation coefficient.

[0078] Specifically, in step S62 of this embodiment, the state correlation coefficient is compared with the correlation threshold. When the state correlation coefficient is greater than the correlation threshold, the feature tolerance interval is updated to [μ(R(t))-(3-S)×σ(R(t)),μ(R(t))+(3+S)×σ(R(t))] to update the data state analysis process. Otherwise, the data state analysis process is not updated.

[0079] Specifically, in this embodiment, the association threshold is set to 0.7. This embodiment does not impose specific limitations on the value of the association threshold, and those skilled in the art can set it freely, as long as it satisfies the updating of the data state analysis process. The value of the association threshold should satisfy the condition of being within [0.6, 1).

[0080] Specifically, in step S62 of this embodiment, the feature tolerance interval is dynamically updated by comparing and analyzing the state correlation coefficients, so that the feature tolerance interval is related to the correlation features of food data between different food states.

[0081] Please continue reading. Figure 3 As shown, the data status update method further includes:

[0082] Step S63 is the analysis process of updating the state correlation coefficient based on environmental data and transportation time.

[0083] Specifically, in step S63 of this embodiment, the transportation time is used as the independent variable, and the product of humidity and carbon dioxide concentration is used as the dependent variable for regression analysis to analyze the regression equation of the product of humidity and carbon dioxide concentration with respect to the transportation time, and the coefficient of determination of the regression equation is calculated. If the coefficient of determination is greater than the determination threshold, the analysis process of the state correlation coefficient is not updated; otherwise, the expression of the state correlation coefficient is updated to S=r×σ(R1(t)). Nr1 / NR1 / σ(R2(t)) Nr2 / NR2 In the formula, r represents the coefficient of determination.

[0084] Specifically, in this embodiment, the decision threshold is set to 0.8. This embodiment does not impose specific limitations on the value of the decision threshold. Those skilled in the art can set it freely, as long as it satisfies the analysis of the state correlation coefficient. The value of the decision threshold should be within [0.7, 0.9].

[0085] Specifically, in step S63 of this embodiment, by modeling the synergistic effect of humidity and carbon dioxide concentration in long-distance transportation, the expression of the state correlation coefficient is dynamically adjusted to solve the model failure problem under sudden environmental changes.

[0086] Please see Figure 4 As shown, this is a food data processing device based on big data in this embodiment, including:

[0087] The data acquisition module is used to collect environmental data, food data, and spatiotemporal data during the cold chain transportation of food, as well as the state of food after the cold chain transportation.

[0088] The fusion module is used to perform heterogeneous fusion of the collected data to construct a spatiotemporal matrix and a food correlation matrix. The fusion module is connected to the collection module.

[0089] The analysis module is used to perform dynamic feature extraction and environmental coupling weight allocation on the spatiotemporal matrix to obtain temperature features, microbial features and environmental coupling features. The analysis module is connected to the fusion module.

[0090] The early warning module is used to concatenate temperature features, microbial features, and environmental coupling weights into a feature vector, and also to determine the data status and issue an early warning based on the feature vector. The early warning module is connected to the analysis module.

[0091] The update module is used for the analysis process of updating the data status based on food status, environmental data, and spatiotemporal data. The update module is connected to the early warning module.

[0092] This application also provides a computer-readable storage medium storing instructions that, when run on a computer, cause the computer to perform the big data-based food data processing method as described in the above method embodiments.

[0093] It will be understood by those skilled in the art that all or some of the steps and systems in the methods disclosed above can be implemented as software, firmware, hardware, and suitable combinations thereof. Some or all of the physical components can be implemented as software executed by a processor, such as a central processing unit, digital signal processor, or microprocessor, or as hardware, or as an integrated circuit, such as an application-specific integrated circuit. Such software can be distributed on a computer-readable medium, which can include computer storage media (or non-transitory media) and communication media (or transient media). As is known to those skilled in the art, the term computer storage media includes volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information (such as computer-readable programs, data structures, program modules, or other data). Computer storage media includes, but is not limited to, RAM, ROM, EEPROM, flash memory or other memory technologies, CD-ROM, digital versatile disc (DVD) or other optical disc storage, magnetic cartridges, magnetic tape, disk storage or other magnetic storage devices, or any other medium that can be used to store desired information and is accessible to a computer. Furthermore, as is known to those skilled in the art, communication media typically contain computer-readable programs, data structures, program modules, or other data in modulated data signals such as carrier waves or other transmission mechanisms, and may include any information delivery medium.

[0094] Obviously, the above embodiments of the present invention are merely examples for clearly illustrating the present invention, and are not intended to limit the implementation of the present invention. For those skilled in the art, other variations or modifications can be made based on the above description. It is impossible to exhaustively list all the implementation methods here. All obvious variations or modifications derived from the technical solutions of the present invention are still within the protection scope of the present invention.

Claims

1. A big data-based food data processing method, characterized by, The application relates to a data state analysis method for food cold chain transportation. The method comprises the following steps: Collecting environmental data, food data and space-time data in the food cold chain transportation process; Fusing the collected data to construct a space-time matrix and a food correlation matrix; Aligning the collection time of the environmental data and the space-time data, and sequentially numbering the collection time of the environmental data and the space-time data in time order to obtain collection number t, and fusing the environmental data and the space-time data at the same collection time to construct a space-time matrix E(t), wherein E(t)=[T(t), H(t), C(t), L(t)], T(t) represents temperature, H(t) represents humidity, C(t) represents carbon dioxide concentration, and L(t) represents altitude; Associating food types with corresponding microbial growth rates and food correlation matrix deterioration critical temperature to construct a food correlation matrix; The temperature in the space-time matrix within 1 hour is dynamically characterized based on a length of a preset analysis period to obtain temperature characteristics, and an expression of the temperature characteristics is: In the expression, F(t) represents the temperature characteristics, v represents a first extraction parameter, v∈N and v≤V, V represents a second extraction parameter, V=3600 / t1, t1 represents the length of the preset analysis period, and λ represents a temperature decay factor. The temperature and food correlation matrix in the time-space matrix is dynamically characterized based on the initial microorganism content to obtain a microorganism feature, and the expression of the microorganism feature is G(t)=G0×e G1(i)×[T(t)-G2(i)] / G2(i) , wherein G(t) represents the microorganism feature, G0 represents the initial microorganism content, i represents the food type, G1(i) represents the microorganism growth rate, and G2(i) represents the spoilage critical temperature. Performing dynamic feature extraction and environmental coupling weight distribution on the space-time matrix to obtain temperature features, microbial features and environmental coupling features; Performing environmental coupling weight distribution on the carbon dioxide concentration and the altitude in the space-time matrix to obtain environmental coupling features, and the expression of the environmental coupling features is W(t)=beta1*(1-L(t) / Lmax)+beta2*C(t) / Cmax, wherein beta1 represents an altitude weight, beta2 represents a carbon dioxide weight, beta1+beta2=1, Lmax represents a preset maximum tolerance altitude, and Cmax represents a preset carbon dioxide critical concentration; Splicing the temperature features, the microbial features and the environmental coupling weight into a feature vector, judging the data state according to the feature vector and giving a warning; Taking the dot product of the feature vectors of adjacent collection numbers as a risk feature value, and formulating a feature value tolerance interval according to the risk feature value to judge the data state; Extracting the risk feature value in the food cold chain transportation process, calculating the average value and the standard deviation of the extracted risk feature value, and formulating a feature value tolerance interval according to the average value and the standard deviation of the extracted risk feature value, wherein the feature value tolerance interval is set as [mu(R(t))-3*sigma(R(t)), mu(R(t))+3*sigma(R(t))]; Comparing the state correlation coefficient with a correlation threshold value, updating the feature tolerance interval to [mu(R(t))-(3-S)*sigma(R(t)), mu(R(t))+(3+S)*sigma(R(t))] when the state correlation coefficient is greater than the correlation threshold value to update the data state analysis process, and vice versa, R(t) represents the risk feature value, mu() represents the average value of the data in the brackets, sigma() represents the standard deviation of the data in the brackets, and S represents the state correlation coefficient; Collecting the food state after the food cold chain transportation; Updating the data state analysis process according to the food state, the environmental data and the space-time data.

2. The big data-based food data processing method of claim 1, wherein, The temperature feature, the microbial feature and the environment coupling weight are spliced to obtain a feature vector X(t), X(t) is set as [F(t), G(t), W(t)], and a dot product of feature vectors of adjacent collection numbers is taken as a risk feature value, and the expression of the risk feature value is R(t)=X(t)·X(t-1); If the risk feature value of the current analysis period does not belong to the feature value tolerance interval, it is determined that the data state is abnormal; Otherwise, it is determined that the data state is normal; According to the data state and the risk feature value, early warning is performed, when the data state is abnormal, if the risk feature value is less than the first feature threshold, severe early warning is performed, if the risk feature value is greater than or equal to the first feature threshold and less than the second feature threshold, moderate early warning is performed, and if the risk feature value is greater than or equal to the second feature threshold, mild early warning is performed; When the data state is normal, if the risk feature value is less than the first feature threshold, moderate early warning is performed, and if the risk feature value is greater than or equal to the first feature threshold, no early warning is performed. 3.The big data-based food data processing method of claim 2, wherein, The risk characteristic values are divided into normal characteristic values and deteriorated characteristic values according to food states, the number of normal characteristic values not belonging to the characteristic value tolerance interval is counted as the first-class abnormal number, the number of deteriorated characteristic values not belonging to the characteristic value tolerance interval is counted as the second-class abnormal number, and the state correlation coefficient S is analyzed, the expression of the state correlation coefficient S is S = σ(R1(t)) Nr1 / NR1 / σ(R2(t)) Nr2 / NR2 , wherein R1(t) represents the normal characteristic value, Nr1 represents the first-class abnormal number, NR1 represents the number of normal characteristic values, R2(t) represents the deteriorated characteristic value, Nr2 represents the second-class abnormal number, and NR2 represents the number of deteriorated characteristic values. The regression analysis is performed with the transportation time length as the independent variable and the product of the humidity and the carbon dioxide concentration as the dependent variable to analyze the regression equation of the product of the humidity and the carbon dioxide concentration with respect to the transportation time length, and the determination coefficient of the regression equation is calculated. If the determination coefficient is greater than a determination threshold, the analysis process of the state correlation coefficient is not updated, otherwise, the expression of the state correlation coefficient is updated as S = r x σ(R1(t)) Nr1 / NR1 / σ(R2(t)) Nr2 / NR2 , wherein r represents the determination coefficient.

4. A big data-based food data processing apparatus applied to the big data-based food data processing method according to any one of claims 1 to 3, characterized by, It comprises: A collection module is used to collect environment data, food data and space-time data in the food cold chain transportation process and food state after food cold chain transportation; A fusion module is used to perform heterogeneous fusion on the collected data to construct a space-time matrix and a food association matrix; An analysis module is used to perform dynamic feature extraction and environment coupling weight distribution on the space-time matrix to obtain temperature features, microbial features and environment coupling features; An early warning module is used to splice the temperature features, the microbial features and the environment coupling weight into a feature vector, and is also used to determine the data state and perform early warning according to the feature vector; An update module is used to update the analysis process of the data state according to the food state, the environment data and the space-time data.

5. A storage medium, characterized by The computer is caused to execute the food data processing method based on big data according to any one of claims 1-3 when the instructions stored therein are run on the computer.

Images